An Economic History of Twentiet - Enrique Cardenas

An Economic History of Twentieth-Century Latin America
Volume 3: Industrialization and the State in Latin America: The Postwar Years
E. Cardenas; J. Ocampo; R. Thorp
ISBN: 9780230595682
DOI: 10.1057/9780230595682
Palgrave Macmillan
Please respect intellectual property rights
This material is copyright and its use is restricted by our standard site license
terms and conditions (see palgraveconnect.com/pc/info/terms_conditions.html).
If you plan to copy, distribute or share in any format, including, for the avoidance
of doubt, posting on websites, you need the express prior permission of Palgrave
Macmillan. To request permission please contact rights@palgrave.com.
Edited by Enrique Crdenas,
Jos Antonio Ocampo and Rosemary Thorp
An Economic History of
Twentieth-Century Latin
America
St Antonys Series
General Editor: Richard Clogg (1999 ), Fellow of St Antonys College, Oxford
Recent titles include:
Enrique Crdenas, Jos Antonio Ocampo and Rosemary Thorp (editors)
AN ECONOMIC HISTORY OF TWENTIETH-CENTURY LATIN AMERICA
Volume 1: The Export Age
Volume 2: Latin America in the 1930s
Volume 3: Industrialization and the State in Latin America
Jennifer G. Mathers
THE RUSSIAN NUCLEAR SHIELD FROM STALIN TO YELTSIN
Marta Dyczok
THE GRAND ALLIANCE AND UKRAINIAN REFUGEES
Mark Brzezinski
THE STRUGGLE FOR CONSTITUTIONALISM IN POLAND
Suke Wolton
LORD HAILEY, THE COLONIAL OFFICE AND THE POLITICS OF RACE AND
EMPIRE IN THE SECOND WORLD WAR
The Loss of White Prestige
Junko Tomaru
THE POSTWAR RAPPROCHEMENT OF MALAYA AND JAPAN, 194561
The Roles of Britain and Japan in South-East Asia
Eiichi Motono
CONFLICT AND COOPERATION IN SINO-BRITISH BUSINESS, 18601911
The Impact of the Pro-British Commercial Network in Shanghai
Nikolas K. Gvosdev
IMPERIAL POLICIES AND PERSPECTIVES TOWARDS GEORGIA, 17601819
Bernardo Kosacoff
CORPORATE STRATEGIES UNDER STRUCTURAL ADJUSTMENT IN ARGENTINA
Responses by Industrial Firms to a New Set of Uncertainties
Ray Takeyh
THE ORIGINS OF THE EISENHOWER DOCTRINE
The US, Britain and Nassers Egypt, 195357
Derek Hopwood (editor)
ARAB NATION, ARAB NATIONALISM
Judith Clifton
THE POLITICS OF TELECOMMUNICATIONS IN MEXICO
Privatization and StateLabour Relations, 192895
10.1057/9780230595682 - An Economic History of Twentieth-Century Latin America, Edited by Enrique Cardenas, Jose A. Ocampo and Rosemary Thorp
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Ccile Laborde
PLURALIST THOUGHT AND THE STATE IN BRITAIN AND FRANCE, 190025
Craig Brandist and Galin Tihanov (editors)
MATERIALIZING BAKHTIN
C. S. Nicholls
THE HISTORY OF ST ANTONYS COLLEGE, OXFORD, 19502000
Anthony Kirk-Greene
BRITAINS IMPERIAL ADMINISTRATORS, 18581966
Laila Parsons
THE DRUZE BETWEEN PALESTINE AND ISRAEL, 194749
M. K. Flynn
IDEOLOGY, MOBILIZATION AND THE NATION
The Rise of Irish, Basque and Carlist Nationalist Movements in the
Nineteenth and Early Twentieth Centuries
Karina Sonnenberg-Stern
EMANCIPATION AND POVERTY
The Ashkenazi Jews of Amsterdam, 17961850
Shane ORourke
WARRIORS AND PEASANTS
The Don Cossacks in Late Imperial Russia
Azar Gat
BRITISH ARMOUR THEORY AND THE RISE OF THE PANZER ARM
Revising the Revisionists
St Antonys Series
Series Standing Order ISBN 0333711092
(outside North America only)
You can receive future titles in this series as they are published by placing a standing order.
Please contact your bookseller or, in case of difficulty, write to us at the address below with
your name and address, the title of the series and the ISBN quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke,
Hampshire RG21 6XS, England
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
An Economic History of
Twentieth-Century
Latin America
Volume 3
Industrialization and the State in Latin America:
The Postwar Years
Edited by
Enrique Crdenas
Professor of Economics and Rector
Universidad de las Amricas-Puebla
Jos Antonio Ocampo
Executive Secretary
United Nations Economic Commission for
Latin America and the Caribbean (CEPAL/ECLAC)
and
Rosemary Thorp
Reader in the Economics of Latin America
St Antonys College
Oxford
in association with
St Antonys College, Oxford
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Editorial matter, selection and Chapter 1 Enrique Crdenas, Jos Antonio
Ocampo and Rosemary Thorp 2000
Chapter 7 Enrique Crdenas 2000
Chapter 9 Jos Antonio Ocampo and Camilo Tovar 2000
Chapters 26, 8, 10, 11 Palgrave Publishers Ltd 2000
All rights reserved. No reproduction, copy or transmission of
this publication may be made without written permission.
No paragraph of this publication may be reproduced, copied or
transmitted save with written permission or in accordance with
the provisions of the Copyright, Designs and Patents Act 1988,
or under the terms of any licence permitting limited copying
issued by the Copyright Licensing Agency, 90 Tottenham Court
Road, London W1P 0LP.
Any person who does any unauthorised act in relation to this
publication may be liable to criminal prosecution and civil
claims for damages.
The authors have asserted their rights to be identified
as the authors of this work in accordance with the
Copyright, Designs and Patents Act 1988.
First published 2000 by
PALGRAVE
Houndmills, Basingstoke, Hampshire RG21 6XS and
175 Fifth Avenue, New York, N. Y. 10010
Companies and representatives throughout the world
PALGRAVE is the new global academic imprint of
St. Martins Press LLC Scholarly and Reference Division and
Palgrave Publishers Ltd (formerly Macmillan Press Ltd).
ISBN 0333633423
This book is printed on paper suitable for recycling and
made from fully managed and sustained forest sources.
A catalogue record for this book is available
from the British Library.
Library of Congress Cataloging-in-Publication Data
An economic history of twentieth-century Latin America / edited by Enrique
Crdenas, Jos Antonio Ocampo, and Rosemary Thorp.
p. cm.
Includes bibliographical references and index.
Contents: v. 1. The export age / edited by Enrique Crdenas, Jos Antonio
Ocampo, and Rosemary Thorp v. 2. Latin America in the 1930s : the
role of the periphery in world crisis / edited by Rosemary Thorp v. 3.
Industrialization and the State in Latin America : the postwar years /
edited by Enrique Crdenas, Jos Antonio Ocampo, and Rosemary Thorp.
ISBN 0333913043 (v. 1 : cloth) ISBN 0333633415 (v. 2 : cloth)
ISBN 0333633423 (v. 3 : cloth)
1. Latin AmericaEconomic conditions. 2. Latin AmericaEconomic
conditions1918 I. Crdenas, Enrique. II. Ocampo, Jos Antonio.
III. Thorp, Rosemary.
HC125 .E37365 2000
330.98'0033dc21
00040449
10 9 8 7 6 5 4 3 2 1
09 08 07 06 05 04 03 02 01 00
Printed and bound in Great Britain by
Antony Rowe Ltd, Chippenham, Wiltshire
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ForCarlosD azAlejandro
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
This page intentionally left blank
10.1057/9780230595682 - An Economic History of Twentieth-Century Latin America, Edited by Enrique Cardenas, Jose A. Ocampo and
Rosemary Thorp
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Contents
ListofTablesandFigures ix
ForewordtoAnEconomicHistoryofTwentieth-CenturyLatinAmerica xii
NotesontheContributors xiii
1 Introduction 1
EnriqueCardenas,Jose AntonioOcampoandRosemaryThorp
2 TechnologicalLearning,InstitutionBuildingandthe
MicroeconomicsofImportSubstitution 36
JorgeKatzandBernardoKosacoff
3 ECLAandtheTheoryofImportSubstitutingIndustrialization
inLatinAmerica 58
E.V.K.FitzGerald
4 TheInfluenceofInternationalFinancialInstitutionsonISI 98
RichardWebb
5 TheIndustrializationofChileduringProtectionism,194082 114
RicardoFfrench-Davis,OscarMunoz,Jose MiguelBenavente
andGustavoCrespi
6 ImportSubstitutionandGrowthinBrazil,1890s1970s 154
MarcelodeP.Abreu,AfonsoS.BevilaquaandDemosthenes
M.Pinho
7 TheProcessofAcceleratedIndustrializationinMexico,192982 176
EnriqueCardenas
8 IndustrializationinVenezuela,193683:TheProblemof
Abundance 205
PabloAstorga
9 ColombiaintheClassicalEraofÌnward-LookingDevelopment',
193074 239
Jose AntonioOcampoandCamiloTovar
10 Import-SubstitutingIndustrializationinArgentina,194080:
ItsAchievementsandShortcomings 282
vii
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
viii Contents
11 ImportSubstitution,EconomicIntegrationandtheDevelopment
ofCentralAmerica195080 314
IsaacCohen
Index 335
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ListofTablesandFigures
Tables
1.1 Worldeconomicgrowth 7
1.2 Exportgrowthrate 24
3.1 TheexternalconstraintaccordingtoECLA 85
3.2 Aggregatecapital,labourandoutputratiosinLatin
America,195094 86
3.3 Ratesofaccumulation(GFCF/GDP) 86
3.4 ThefiscalstructureinLatinAmerica 88
3.5 FiscalincidenceinLatinAmerica 89
5.1 Chile:rateofgrowthofGDP 115
5.2 Chile:rateofgrossfixedinvestment 116
5.3 Chile:productivityofcapital 116
5.4 Chile:sectoralgrowthofGDPandshareofindustry 119
5.5 Chile:averagegrowthrateofvalueaddedperindustrialworker 124
5.6 Chile:valueaddedperworkerinChileandtheUS,197094 124
5.7 Chile:percentageandaverageexpenditureonresearchand
developmentbyinstitutionalsectors,196592 129
5.8 Chile:costperdollarofimports,inrealterms 137
6.1 Brazil:coffeepriceregressions,18801960 159
6.2 Brazil:GDPandindustrialoutputgrowthrates,193080 162
6.3 Brazil:GDPshares(%),191080 162
6.4 Brazil:distributionofindustrialvalueadded,191959 163
6.5 Brazil:percentageshareofcoffeeandmanufactured
exportsintotalexports,193080 169
6.6 Brazil:lifeexpectancyatbirth,inyears,194080 171
7.1 Mexico:sourcesofindustrialgrowth,192939 181
7.2 Mexico:sourcesofeconomicgrowthofindustry 185
7.3 Mexico:sourcesofindustrialdemandgrowth 188
7.4 Mexico:sourcesoffinancingoftheeconomy,195070 196
7.5 Mexico:positionofthe10mostimportantindustrial
branches,196081 200
8.1 Venezuela:sharesofselectedindustriesinmanufacturing 214
8.2 Venezuela:GDPbyeconomicsectors:annualaveragegrowth
andshares 221
9.1 Colombia:compositionofeconomicactivity 241
9.2 Colombia:externaltraderatiosandcomposition 243
ix
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
x ListofTablesandFigures
9.3 Colombia:evolutionofinfrastructure 249
9.4 Colombia:compositionofindustrialGDP 251
9.5 Colombia:sourcesofgrowthinthemanufacturingsector 253
9.6 Colombia:sourcesofcorporatefinancing 255
9.7 Colombia:tradepolicyindicators 259
9.8 Colombia:effectiveprotectionaccordingtoISICclassification 261
9.9 Colombia:growthofagriculturalsector 264
9.10 Colombia:maineducationalandhealthindicators 274
9.11 Colombia:qualityofhousing 275
9.12 Colombia:averagewagesforselectedsectors 277
10.1 Argentina:theexpansionofmanufacturingindustry,193049 283
10.2 Argentina:theparticipationoflabourindomesticGDP 287
10.3 Argentina:theexpansionofmetalworkingandchemical
productioninthe1950sand1960s 289
10.4 Argentina:averageannualgrowthratesofindustrialvalue
added 292
10.5 Argentina:industrialoutputduringthe1970splateau
(1970=100) 295
10.6 Argentina:capitalflight(inUS$millions) 295
10.7 Argentina:structureofindustrialvalueadded,1974and1990 296
10.8 Argentina:rawmaterialprocessingplantserectedduringthe
finalstagesoftheISIprocess 297
10.9 Argentina:flowofforeignprivatecapitaltoArgentine
manufacturingaccordingtothelegalregime 298
10.10 Argentina:foreignprivatecapitalbyindustry,195461 299
10.11 Argentina:exportof`pure'technologybyArgentinefirms
undertheformofturnkeyproductionfacilitiessoldin
LatinAmerica 309
11.1 CentralAmerica:totalGDPandpercapitalrealgrowthrates 316
11.2 CentralAmerica:GDPcomposition 317
11.3 CentralAmerica:incomedistributionbyhouseholds 324
11.4 CentralAmerica:theshareofintraregionaltradeintotaltrade 326
Figures
1.1 GDPgrowthandcomposition 17
1.2 Convergencepatterns 18
1.3 Relativelivingstandardindex(relativetoUSA) 21
1.4 Exportsas%ofGDP 23
1.5 Nettransferofresourcesas%of1980realGDP 27
2.1 Technologicallearningandimprovementsinproduction
duringtheimportsubstitutionprocess 47
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ListofTablesandFigures xi
5.1 Chile:Chileanbalanceoftradepattern 122
5.2 Chile:nettradebalancebysubsections 123
7.1 Mexico:exchangerateovervaluationandtheimportcoefficient 188
7.2 Mexico:evolutionofkeyrealpublicpricesforindustry 191
7.3 Mexico:evolutionofindustrialstructure,196081 199
8.1 Venezuela:valueaddedinmanufacturing,193689 212
8.2 Venezuela:realexchangerate,192086 223
8.3 Venezuela:labourproductivitybysectors,192090 224
8.4 Venezuela:insupportofindustrialization 225
9.1 Colombia:fixedinvestmentinmachineryandequipment 244
9.2 Colombia:termsoftradevsrealimportexchangerate 254
9.3 Colombia:Colombianindustrialprices/internationalprices 262
10.1 Argentina:totalproductionofgraininArgentina,191084 288
10.2 Argentina:performanceoftheindustrialsector,196474 308
10.3 Argentina:totalexports,industrialexportsand
manufacturingexportsofindustrialorigin,1964and1974 308
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ForewordtoAnEconomicHistoryof
Twentieth-CenturyLatinAmerica
Thethreevolumesofthismini-seriesreflectacollaborativeeffortthatbeganin
1982,whenagroupofeconomistsandeconomichistoriansmettoconsiderthe
effect of the 1929 Depression on Latin America, and the explanation of the
continent's rapid recovery. Carlos D az Alejandro was the leading light of the
group until his premature death in 1985. This initial group effort led to the
study LatinAmericainthe1930s,publishedin1984.Subsequent collaboration
was made possible by the inspiration of Enrique V. Iglesias, President of the
Inter-American Development Bank,whoseideaitwastofundamajorproject
to write an economic history of Latin America for the twentieth century,
accessibletoawideaudiencebutbasedonthebestLatinAmericanscholarship.
Hisinvitationtometoleadtheprojectandwritetheoverviewvolumegaveme
themoststimulating16 monthsofmyacademiccareer.Newfriendsjoinedthe
group, and we came together in the course of 1997 in a series of workshops
organized to support the project. These workshops resulted in the books
published here as Volumes 1 and 3. Volume 2 is a reprint of the 1930s
study. The overview volume was published by the IDB in 1998 as Progress,
Poverty and Exclusion: An Economic History of Latin America in the Twentieth
Century(available inEnglish, Spanish, Frenchand Portuguese from IDBbook-
store, www.iadb.org/pub, and in English through Johns Hopkins University
Press,www.press.jhu.edu.).
My thanks for the opportunity to lead this stimulating project go first and
foremosttoEnriqueIglesiashimself,totheInter-AmericanDevelopmentBank
andtotheEuropeanUnionforfunding.TimFarmiloewasourenthusiasticand
patient editor. I owe a particular vote of thanks to my co-editors, who made
themselves available against exceptional pressures and were remarkable part-
ners.MargaretHancox,CatalinaSanintandElviraRyanborethehugeburden
ofthefinaleditingandprocessing.
ROSEMARY THORP
LatinAmericanCentre
OxfordUniversityandStAntony'sCollege
xii
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
NotesontheContributors
Marcelo de Paiva Abreu is Professor of Economics at the Department of
Economics, Catholic University of Rio deJaneiro, and formerDirector for the
Department,199097.
PabloAstorgahasaDPhilinEconomicsfromtheUniversityofOxford.Heisa
ResearchFellowatStAntony'sCollege,Oxford,andattachedtoIESA,Caracas,
Venezuela.
Jose Miguel Benavente has a BA in Industrial Engineering and an MSc in
Economics fromthe Universityof Chile, andan MScin Economics forDevel-
opment from the University of Oxford. He is Lecturer in Economics, Depart-
ment of Economics, University of Chile, and senior consultant for ECLAC,
MinistryofEconomicsandCORFOinChile.
Alfonso S. Bevilaqua is amember of the Department of Economics, Catholic
UniversityofRiodeJaneiro(PUC-Rio),Brazil.HehasaPhDinEconomicsfrom
theUniversityofCalifornia,Berkeley.
Enrique Cardenas is Professor of Economics and Rectorof Universidad de las
Americas-Puebla,since1985.
Isaac Cohen was director of the Washington office of the United Nations
Economic Commission for Latin America and the Caribbean (ECLAC) and is
nowaprivateconsultantbasedinWashington,DC.
GustavoCrespiisLecturerinEconomicsattheUniversityofChileandsenior
consultantforECLACMinistryofEconomicsinChile.
Ricardo Ffrench-Davis is Principal Regional Adviser of ECLAC, and Professor
of Economics at the University of Chile. He is a former Director of Policies of
theCentralBankofChileandDirectoroftheCentreforEconomicResearchon
LatinAmerica.
E.V.K.FitzGeraldisReaderinInternationalEconomicsandFinanceatOxford
UniversityandaProfessorialFellowofStAntony'sCollege,Oxford.
Jorge Katz, previously Professor of International Economics at the University
of Buenos Aires, is currently Regional Adviser on Industrial and Technical
DevelopmentatECLACinSantiago,Chile.
Bernardo Kosacoff obtained a BA in Economics at the University of Buenos
Airesin1972;hewasaVisitingSeniorFellow,StAntony'sCollege,Oxford,UK.
xiii
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
xiv NotesontheContributors
(1995); Senior Economist, United Nations Economic Commission for Latin
America and the Caribbean, Buenos Aires (since 1983); Professor of Industrial
Organization and Political Economy at the University of Buenos Aires (since
1984)andtheNationalUniversityofQuilmes(since1993).
Oscar Munoz is an economist, formerly President of CIEPLAN, and presently
Executive Secretary of the Forum for Economic Development, Ministry of
EconomicAffairsinChile,andaProfessorattheUniversityofChile.
Jose Antonio Ocampo is currently the Executive Secretary of the United
Nations Economic commission for Latin America and the Caribbean (CEPAL/
ECLAC). He is a former Finance Minister of Colombia, Director (Minister) of
theNationalPlanningDepartmentofColombiaandMinisterofAgricultureof
Colombia. He has also been the Director of research centres in Colombia
(FEDESARROLLOandCEDE).
DemosthenesM.PinhoisattachedtotheCentralBankofBrazil.HehasaPhD
inEconomicsfromtheUniversityofCalifornia,Berkeley.
RosemaryThorpisReaderintheEconomicsofLatinAmericaattheUniversity
of Oxford, Fellow of St Antony's College and Director of the Latin American
Centre.SheisalsoamemberofQueenElizabethHouse,Oxford.
Camilo Tovar is currently Adviser to the Executive Secretary of the United
NationsEconomicCommissionforLatinAmericaandtheCaribbean(CEPAL/
ECLAC). Previously, he served as Adviser to the Finance Minister and to the
DirectoroftheNationalPlanningDepartmentofColombia.
RichardWebbrecentlyco-authoredahistoryoftheWorldBankpublishedby
theBrookingsInstitutionandiscurrentlyPresidentofBancoLatinoinPeru.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
1
Introduction
EnriqueCardenas,Jose AntonioOcampoandRosemaryThorp
In the 1990s, `protection', ìmport substitution' and ìntervention' have
becomedirtywords,partofthe`leyendanegra'ofLatinAmericandevelopment
in the postwar period. Like most `black legends', it is at once both true and
false. This book attempts a deeper and as far as possible dispassionate look at
thecontroversialyearsbetweentheendoftheSecondWorldWarandthepoint
whenalmosteverywhereinLatinAmerica,althoughatvaryingdatesindiffer-
ent countries, a discontinuity occurs in which the postwar `style of develop-
ment'ceasedtoplayacentralroleintheeconomicevolutionoftheregion.As
wewellknow,thisprocesseventuallyledtoanewstageinwhich`themarket'
assumedthepredominantroleinthedevelopmentstrategy.
The analysis is based on seven case studies covering 11 countries. These
studieswerediscussedintwoworkshops,inOxfordinSeptember1996 andin
Paipa, Colombia, in May 1997.
1
The guiding concepts that emerged in the
debate are developed in two conceptual chapters which open the book. On
the micro side, as Katz and Kosakoff in Chapter 2 make clear, the crucial
concept is the èvolutionary' or `maturing process' which characterizes the
development of both technological capabilities and institutions. On the
macro side, Fitzgerald explores in Chapter 3 the issue of accumulation by
theprivateandpublicsectors,anditsrelationbothtopostwarindustrialization
andthe`fiscalcrisis'ofthestate.Thisprovidesa`positioning'ofthetheorizing
of ECLAC, (the UN Economic Commission for Latin America and the Carib-
bean),theworkofwhichisatitsstrongestatheoryofcapitalaccumulation.A
closing chapter by Webb presents a historical perspective on the thinking of
multilateralfinancialinstitutionsatthetime.
The book tries to answer several questions: what are the distinguishing
features of the `style of development' which characterized this period? Is it
adequatelycapturedbythecommonconceptofìmport-substitutingindustria-
lization'(ISI)?Dothedifferentnationalexperiencesgiverisetoasinglestereo-
type? What external and internal forces shaped development patterns? What
1
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2 Introduction
was the role of ideas, and particularly what role did ECLAC's thinking play?
Given existing constraints, what were the real alternatives to inward-looking
industrialization? What were the benefits and costs of the path adopted? To
what extent were the debt crisis and adjustment problems of the 1980s a
reflection of such costs? Were such problems a result of the internal contra-
dictions of the development model followed, or rather of political instability,
macroeconomic financial mismanagement or exogenous reasons? As we will
see, theanswerstothesequestionsvarybycountry andsubperiod,andarein
factlessneatthangenerallyassumed.
The following section of this introduction explores the core topic of the
volume. We then elaborate the context, both national and international, to
whichpostwarindustrializationwasresponding,andfinallyreviewtheoverall
resultsforLatinAmerica.
Thetopicofthevolume
Our subject is the period of accelerated industrialization that followed the
Second WorldWar in large andmedium-sized Latin American countries, with
ècho'effectsinthesmallercountries.Inmanycases,thepostwardevelopment
pattern hadclearcontinuitieswith the1930s andevenbefore, afact whichis
brought out in several of the case studies. This period has typically been
referred to as ìmport substituting industrialization', but this is not a very
helpful label, since it is as much about a new and expanded role of the state
as it is about import substitution. In addition, modern manufacturing first
appeared in several countries of the region in a significant scale in the late
nineteenth and early twentieth centuries, as a response to the expansion and
integration of domestic markets induced by increasing integration into the
internationaleconomy.Forobviousreasons,thatfirstspurtofindustrialization
waslargelystimulatedbythesubstitutionofgoodshithertoimported,mostly
consumer goods, and so qualifies as ìmport substituting industrialization'.
Moreover, in several cases (Brazil, Chile, Colombia and Mexico being notable
examples), protection actually promoted such domestic linkages during the
exportage,inthelatenineteenthandearlytwentiethcenturies,indeedaspart
ofastylizedinternationalpracticeatthetime(seetheintroductionandcoun-
trycasesinthecompanionvolumeontheexportage).
Ìmportsubstitution'isanimperfectlabelformanyadditionalreasons.Inthe
postwaryearstheexportsectorcontinuedtoplayafundamentalrole,although
increasinglyasasourceoftheforeignexchangerequiredtofinancetheimports
of capital and intermediate goods which were not produced domestically. In
most mineral export economies most notably Venezuela throughout the
periodofanalysismineralrentswereamajorsourceofcapitalaccumulation.
Moreover, in many, particularly small, countries industrialization did not
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas,Jose AntonioOcampoandRosemaryThorp 3
totally replace but was actually superimposed on the primary export model,
whichcontinuedtodominatedevelopmenttrends,asthechapterbyCohenon
CentralAmericainthisvolumemakesclear.Insomemediumandlargecoun-
tries, export promotion was introduced as an essential component of the
development strategy at some stage of the industrialization process, which in
these cases thus became a `mixed model' that combined import substitution
with export promotion. The model was also `mixed' in several countries at
differenttimesinthesensethatitactivelypromoted,withsimilarinstruments
to those used to encourage industrialization, several agricultural activities,
which servedto generateexport income orproduced foodstuffsfor agrowing
urban population or inputs for the manufacturing sector. On top of that,
despite protectionism, there was often no net import substitution during the
process and import substitution was not always or consistently over time the
leading source of growth, even in new industries, whereas domestic demand
generallyplayedamoreconsistentrole.
Therefore,theessenceoftheperiodcoveredinthisvolumeisbestexpressed
as `state-led industrialization' or àccelerated industrialization',
2
a process
accompanied by a thorough transformation of the particular economies and
societies. Indeed, accelerated industrialization was only possible through the
integrationofdomesticmarkets,andagreaterandmoreubiquitousroleofthe
state.Itdevelopedtogetherwithurbanizationandanincreasingroleoflabour
unions,entrepreneurialassociationsandstateenterprises.Itincludedawhole
rangeofgovernmentalsupportthroughfiscal,tradeandexchangeratepolicies.
Althoughmanyofthesepoliciesfirstappearedduringthe1930sasaconsequence
oftheGreatDepressionandsomeevenpredatedit(seethetwopreviousvolumes
inthisseries),itwasonlyinthepostwarperiodthatmoreconsciouspolicieswere
developedandimplementedtospeedupindustrialization.
These policies were pushed up to the point variable among countries
where the diminishing returns from their use became evident. As we have
pointed out, this led in some cases to a rationalization of the development
strategy towards a `mixed model', maintaining, nonetheless, high levels of
stateintervention.Theoldformsofinterventionwerekeptbutweredetached
increasinglyinmanycountriesfromadevelopmentstrategyofanysort,asthe
short-term management of booms and crises came to dominate totally the
economic scene. This was increasingly true in the years following the first oil
shock. In this sense, the strategy of àccelerated industrialization' was effect-
ivelyabandonedinmanycountriesintheearlyormid-1970s,eventhoughits
policy manifestations maintained an inertia of their own. Throughout this
process,therewasagradualshiftinpolicy-makingideologyawayfromprotec-
tionandtowardsopennessandbeliefinthevalueofexposuretointernational
competitive pressures. However, the Chilean case aside, the new ideology
became dominant only in the second half of the 1980s and, particularly, in
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
4 Introduction
the1990s.Thus,alongperiodelapsedinmanycountriesbetweentheeffective
end of the era of state-led industrialization and its replacement by the new,
market-orientedstrategy.
The process of accelerated industrialization must thus be viewed together
with the transformation of the state and society which accompanied it. The
politicaleconomydimension,bywhichnewactorsappeared,suchasentrepre-
neurial organizations, labour unions, peasant movements and technocrats, as
well as the transformation of political parties and movements to which these
processesled,grewindepthandcomplexity.Majorsocialconflictsbuiltupin
several countries: the Cuban revolution accelerated this process, but it was
probably only in the 1970s that social conflict became the dominant feature,
tothepointofseriouslyundermininginvestorconfidenceinseveralcountries.
Externalandinternalforcesshapingdevelopment
Theexternalcontext
Theoutstandingcharacteristicofthepostwarscenewasthenewdominanceof
the USA. While European countries had suffered terrible destruction, the pro-
ductive capacity of the USA had increased 50 per cent during the war and in
1945 produced more than half the worldwide total of manufacturing goods,
andwascapableofquicklyconvertingproductivecapacityfromwartopeace-
timeproduction.Stillmoresignificant,theUSAownedhalfoftheworldsupply
ofshipping,comparedwithonly14percentin1939,andsuppliedone-thirdof
worldexportswhiletakingonlyone-tenthofworldimports(Ashworth,1987,
p.266).
TheUSAalsohadarelativelyclearagenda.Whereasintheinterwaryearsthe
signals pointing to the need for change in the international system were
therebutwereweakandconflicting,astheSecondWorldWardrewtoaclose
the international system was clearly perceived to have broken down and to
requiremajorinstitutionalchange.LedbytheUSA,talkshadgoneonatmany
levelsduringthewaritselftopreparetobuildthepeace.Theeconomicagenda
comprised:(i)institutionaldevelopmentstoensurestableandexpandingtrade
and capital flows; (ii) the use of those institutions to reduce discriminatory
controls; and (iii) a sufficiently rapid recovery of the main trading nations to
complementtheinstitutionalrenewaloftheinternationalsystem.Theagenda
shaped the Bretton Woods agreement of 1944, which saw the creation of the
International Monetary Fund (IMF) and the World Bank (IBRD), specifically
aimingatareturntoasystemofstableexchangeratesandtoanassuredsupply
of long-term capital, with a progressive liberalization of the European econo-
mies. The `dollar shortage' made that process a very gradual one, leading to
practices which, for many years, continued to be far more protectionist than
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
thetheory.
3
Asitbecameclearthattheseactionsalonewouldnotbeenoughto
kick-starttheEuropeaneconomies,amoreradicalUSrolewasconceivedwith
theMarshallPlanin1947.LengthynegotiationseventuallyaddedtheGeneral
Agreement on Tariffs and Trade (GATT), signed in 1947 in Geneva by 23
nations (including Argentina, Chile and Cuba), but failed to add the Interna-
tionalTradeOrganisation(ITO).
4
WhatwasonlytobecomeclearwithtimewashowfardowntheUSagenda
wasthepositionofLatinAmericaintheearlypostwaryears.Theembodiment
of this was the series of conferences which began with Chapultepec in 1945
before the war ended. The eighth Inter-American Conference in Chapultepec
was a disillusion, in that economic issues were mostly postponed to a confer-
ence to deal with them specifically. Nonetheless, at least Latin America
appeared to be on the agenda, and it was possible to sign pious statements
about free trade while reserving the right to make just about any specific
exceptions they should see fit to make. On the issue of most concern to the
USA conditions for direct foreign investment the Latin Americans were
rather interested in responding. But then the economic conference was post-
poned so often that Rabe (1978) christened it the èlusive' conference. By the
ninthInter-AmericanConferenceinBogota in1948,aftertheinventionofthe
Marshall Plan and audacious suggestions that the time was ripe for a similar
effortinLatinAmerica,GeneralMarshall'schiefpreoccupationwastopersuade
the Latin American nations that the USA had too much on its shoulders as a
reconstructor of world peace and security to be considered a major source of
finance for the rest of the hemisphere.
5
The corollary was that, therefore, the
essentialthingwastoputinplacegoodsolidguaranteessothatprivatecapital
couldcomeanddothejob.Theclashesoverthe`Calvodoctrine'onthelack
ofresidentaliens'rightstoappealtotheirhomegovernmentweresuchthat
thereafter the US private sector strongly counselled the abandonment of a
conference and the concentration on bilateral negotiations on direct foreign
investmentlegislation(Rabe,1978,p.289).Thisiswhathappened.
A further characteristic of the postwar period was the emerging consensus
thatindustrializationofdevelopingcountriesprovidedbothinterestinginvest-
mentopportunitiesandgoodsales ofmachineryandequipment.Meanwhile,
ontheLatinAmericanside,theampleforeignexchangereservesinplaceasthe
warfinishedvanishedsoonerthananyonethoughtpossible,andprovokedan
abrupt resort to import quotas. When foreign firms found that they were
completely excluded from the market through quotas, thisproduced a strong
incentivetomovetolocalproduction,asitbecamesuddenlytheonlyoption.
So, the early building of the postwar boom and the shaping of Latin
America's mode of insertion went hand in hand. As Webb indicates, the
WorldBankwasquitetolerantofprotectionismindevelopingcountriesnot
sowithrespecttoEuropepreferringamixofexportpromotionandèfficient'
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
6 Introduction
import substitution; it also quietly favoured protectionism in the specific
industries which the Bank or the International Financial Corporation were
financing. It was also less influential in policy formulation than it became
lateron.InitsearlieryearsandevendecadestheIMFalsodidnotaltogether
dislikeprotection,andindeedvieweditasausefulcomplementaryinstrument
tocorrectpaymentsimbalances.
In due time, of course, Latin America became more central to US policy. As
had been true in the early postwar years, this was intimately associated with
howimportanttheregionwasinthestruggleagainstcommunism.EarlyLatin
American initiatives tocreate aregional development bankinitiallyfound no
supportfromtheUSA.In1958,atthemeetingoftheInter-AmericanEconomic
and Social Council, the USA announced that it would support the initiative,
and the 1959 Cuban revolution gave a new urgency to the initiative. An
announcement toestablishtheInter-American DevelopmentBank(IDB), and
withinittheSocialProgressTrustFund,specificallydesignedtolendtosectors
neglected by existing financial organizations education, health, agriculture,
and so on was signed in 1960. This was soon followed by a more ambitious
initiative, the Alliance for Progress which, under the leadership of President
Kennedy, came formally into existence with the Charter of Punta del Este in
August 1961. This initiative also became closely interlinked with the postwar
patternsofstateintervention.Thekeythemeswereintegration,planning,tax
reform, agrarian reform and modernization, and investment in the social
sectors, many of them central to the agenda which the United Nations Eco-
nomicCommissionforLatinAmerica(ECLA
6
)hadbeenpushingforwardsince
the1950s.Manyinitiativesintheseareaswereputinplace,theflowofoutside
funds increased and the IDB was able to direct financing to the neglected
sectors. But the flow of funds was less than promised, and the tied character
ofUSaidandtheconditionthatithadtobespentonUSgoods,àdditional'to
whatwouldhavebeenboughtanyway,becameasourceofsignificantfriction.
Bytheendofthe1960s,itwasclearthatexpectationsunleashedatthestartof
thedecadebytheAllianceforProgresshadnotbeenfulfilled.
Meanwhile, the world economy had entered a period of unprecedented
prosperity. The depressed conditions of the 1930s followed by the wartime
stimulustoinnovationhadleftabacklogoftechnologywaitingtobeapplied.
Theseopportunities,therapidsuccessofEuropeanreconstruction,thegradual
but steady return to freer trade and the successful application of macroeco-
nomicstabilizationtools,amongotherfactors,helpedtounleashthecentury's
fastest economic growth in the world industrial centres. The growth of the
welfare state and, in many countries, of state enterprises, led to a large-scale
expansionofthestateindevelopedmarketeconomies,whichwasaccompan-
ied in parallel by the expansion of centrally-planned economies. In 195073,
theeconomicgrowthoftheindustrializedcountriesaccelerated,notonlywith
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table1.1 Worldeconomicgrowth
187013 191350 195073
GDP GDP GDP GDP GDP GDP
growth growth growth growth growth growth
(%) percapita (%) percapita (%) percapita
(%) (%) (%)
WesternEurope 2.1 1.3 1.4 0.9 4.7 3.9
Westernoffshoots 3.9 1.8 2.8 1.6 4.0 2.4
SouthernEurope 1.5 1.1 1.3 0.4 6.3 4.9
EasternEurope 2.4 1.0 1.6 1.2 4.7 3.5
LatinAmerica 3.3 1.5 3.4 1.5 5.3 2.5
Asia 1.1 0.6 1.0 0.1 6.0 3.8
Africa 1.1 0.4 3.0 1.0 4.4 2.0
World 2.1 1.3 1.9 0.9 4.9 2.9
Source: Maddison(1995).
respect to the interwar period but also with respect to the previous period of
rapidexpansion,18701913.Moreover,thiswaspartofandlargelypropelled
an unprecedented, generalized world economic expansion. As Table 1.1
makes clear, annual world economic growth reached 4.9 per cent in 195073
vs 2.1 per cent in 18701913, according to Maddison's estimates (1995). This
was also accompanied by the most rapid expansion of world trade in history:
real world exports grew at an annual rate of 7 per cent in 195073 vs 3.4 per
centin18701913(seeMaddison,1995,tableI-4).Indeed,thegrowthoftrade
continued to be very rapid even after growth in the industrialized economies
and the world economy at large decelerated, after the first oil shock. The
growthofmultinationalcorporationswasalsopartofthisprocess.Thegrowth
of trade and direct investment was finally accompanied, with a lag, by the
reconstruction of an international financial system, which had been severely
hitbythecrisisofthe1930s.
By the early 1970s, when the era of `state-led industrialization' in Latin
America was coming to an end, the international economy had experienced
significantchangeswithrespecttowhathadbeentypicalintheearlypostwar
years. A dynamic economy had replaced the crippled world economy of the
interwar period and the international trade and financial systems had been
rebuilt. Multinational corporations had become a major agent in the interna-
tionaleconomy.Thestatehadbecomeacentralplayerinmarketeconomies
and obviously the central player in non-market ones but one of the oldest
formsofstateintervention,protectionism,wasundergoingagradualprocessof
negotiatederosion.TheBrettonWoodsinstitutions,GATT,theUnitedNations
organizationsandseveralregionalbodiesmostimportantlythosesupporting
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
8 Introduction
European integration had become well-established, bringing a presence of
stateinterventiontoamoregloballevel.
Rapidgrowthhad`trickleddown'tothedevelopingcountriesthroughboth
trade and investment channels. The expansion of world trade was largely
concentrated in transactions among industrialized nations, and tariff reduc-
tions maintained a structure that discriminated against the imports of light
manufacturedandsomeagriculturalproducts.However,theexpansionopened
windows of opportunity for exports from the periphery, which were rapidly
takenadvantageofbysomecountriesinthedevelopingworldparticularlyin
EastandSoutheastAsia.Also,thegrowthoftheinternationalfinancialsystem
concentratedlargelyonfinancialtransactionsamongdevelopedcountries,but
since the late 1960s, at least, it began to offer large-scale opportunities for
externalfinancingasanalternativetomultilateralbanksandbilateralagencies.
The developing countries underwent, as a result, their own phase of rapid
expansion,asTable1.1indicates,andLatinAmericawascertainlypartofthat
story.However,withsomenotableexceptions,GDPpercapitagrewlessinthe
`periphery'thaninthe`core'oftheworldeconomy.Thus,incomegapstended
to increase in the world economy during its `golden age'. New mechanisms
weredesignedtoenhancethediffusionofprogress,bothataninternational
thatis,theGeneralizedSystemofPreferencesandcommodityagreementsor
regional (that is, the Alliance of Progress) level, but they were generally per-
ceived to be insufficient. The call for a `new international economic order', a
centraldebateofthemid-1970s,wasaclearreflectionofthatfact.
Thecall also demonstrated how far from the international agenda were the
ideas of economic liberalization as the engine of progress of the developing
worldattheendofourperiodofanalysis.Theseideashadmadesomeheadway,
bothineconomicthinkingandinnationalpracticeinsomecountries,butby
thestandardsofthelate1990stheclassictreatiseofthetimethatarguedforthe
reduced use ofprotectionism asadevelopment tool(Little etal.,1970),could
be read today as a defence of moderate state intervention! Indeed, it could
be viewed as further from current ideas than from Rau l Prebisch's report
to UNCTAD a few years earlier (Prebisch, 1964). As Webb argues in his
chapter, the World Bank continued to defend, up to the late 1970s, the ideas
that industrialization was essential to economic development, and that
protectionism was an essential stage in the process of industrialization,
ideas which were clearly exposed by its most influential thinker at the time,
HollisB.Chenery.
Theinternalcontext
There have been several attempts to classify the different stages of industrial-
ization.InChenery'swell-knownclassification,emphasisisplacedonwhether
theprocessreliesonproductionforthedomesticmarketvsexportsinwardvs
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
outwardorientationandonwhenspecificindustriesappearintheprocessof
industrialization èarly' vs `late' manufactures (see, for example, Chenery et
al., 1986). The latter classification is associated to the idea, which was wide-
spread in the period we are analysing, that there was a first èasy' phase of
import substitution during which mostly non-durable consumer goods were
produced,andlaterphasesinwhichimportsubstitution`deepened'producing
non-durable consumer as well as intermediate and some capital goods. Based
on our case studies, four stages in the evolution of industrialization in Latin
Americacanbedifferentiated.
Thefirststagecameasanaturalbyproductofexportexpansion.Theexpan-
sionandintegrationofdomesticmarkets,asaresultofrisingexportrevenues
andthedevelopmentofmoderntransportationnetworks,wascombinedwith
thatofmodernmonetarysystems,capitalandwagelabourmarkets.Aswehave
pointed out, in several countries state protection actually encouraged the
`domestic linkages' of the dynamic export sectors, as well as their linkages to
domesticagriculture.
The second was an èmpirical'stage of state-led industrialization, in which
policy-makers followed their intuition and external conditions created the
appropriatepricestructuretostimulatedomesticratherthanexternaldemand.
The1930sandtheSecondWorldWarprovideagoodexampleofthisstage,but
the First World War had been an important precedent. The collapse of com-
moditypricesandcapitalflowsinthe1930sledtoawaveofdevaluationsand
protectionismwhichshiftedrelativepricesinfavourofdomesticindustriesand
agricultural production. Indeed, exchange controls, protectionism and the
proliferationofbilateraltradeagreements,whichenhancedthecomplexityof
tariffstructures,becamesuchwidespreadpracticesintheindustrializednations
inthe1930sthatitseemednaturalandevennecessarytopracticethemunder
the circumstances. In particular, bilateral trade agreements introduced direct
controls of foreign trade operations to guarantee bilateral trade equilibria,
practiceswhichwereunknownbefore.
Thisledtothedevelopmentofdirectimportcontrolsatamoregenerallevel.
On the other hand, in all cases but Mexico, the Second World War isolated
nationaleconomies,forcingthemtoexploitexistingcapacity,andtoproduce
or create the plans to produce a wider variety of products which were not
available given war shortages. The concept that some èssential' or `strategic'
industries should be encouraged was, thus, a natural byproduct of the war.
After the experience of the 1930s and the Second World War, protectionism,
tariffdiscrimination,directtradecontrols,importsubstitutioninmanufactur-
ingandagriculture,andplanstoencourageèssentialindustries'becamefairly
widespread practices, but they were more a byproduct of the deficient func-
tioningoftheinternationaleconomicsystemthana`developmentstrategy'of
anysort.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
10 Introduction
Theendofthewarreopenedthepossibilityofimportingallkindsofgoods,
expanding output even further, and saw the gradual evolution of the third
stage. Rising commodity prices supported this process; but the repressed
demand was soon manifested in an avalanche of imports. Dollar foreign
exchangereservesevaporatedandmostcountriesreactedbyimposingorhard-
ening trade restrictions and exchange controls. Inconvertibility in European
countrieswasanadditionalconstraintonthewayadjustmentsoperatedinthe
earlypostwarperiodforthosecountriesforwhichEuropewasthemainexport
market. Following the experience of the èmpirical phase', early postwar bal-
ance of payments crises now established a pattern by which each crisis
increased the previous levels of protection. Now, however, a more conscious
industrializationstrategycameintoexistence,theso-calledISImodel,basedon
tariff and non-tariff protection of industry and some agricultural activities,
changesintheinternaltermsoftradeagainsttraditionalprimaryexports,the
design of development banks to support these activities, strong public sector
investment in infrastructure and in `strategic' sectors, including energy and
other raw materials, and the rationing of foreign exchange, with multiple
exchange rates to channel it preferentially to the imports of non-competitive
intermediate and capital goods. An essential characteristic of the model was
thatnewlayersofprotection,topromotenewsectors,wereimposedonolder
layersassociatedwithpreviousimportsubstitutionphases.Itshouldbenoted
that,evenatthetime,themodelwasnotexemptfromcriticism,notevenfrom
ECLACwhichwasinstrumentalinjustifyingandrationalizingthestrategy.At
alltimes,therewerevoicesofwarningonthelimitsofthestrategy,particularly
on the rising costs of new industries being promoted, which affected older
industries adversely, and on the diminished capacity to reach the explicit
goal of reducing import dependence. This `classic' stage of inward-looking
development `desarrollo hacia adentro' dated from the late 1940s to the
early1960sinmostmediumandlargereconomies.
Finallycamethe`mature'stage.Duringthisstage,countriesdifferedsubstan-
tiallyintherelativeweightoftwodifferentstrategies.Thefirstwastogivemore
emphasis to export promotion, generating what we have called a `mixed'
model. It was characterized by the imposition of a new layer of in some
cases, very high export incentives on old layers of protection, generally
accompanied by a moderate rationalization of the latter. The second was to
furtherdeepenimportsubstitutioninintermediateandcapitalgoods.Among
the countries covered in this volume, Argentina, Chile and, particularly,
Colombia were closer to the first, Mexico and Venezuela closer to the second
supported in the 1970s by rising oil revenues in both cases whereas Brazil
veryactivelymixedbothuptotheeveofthedebtcrisis.Mexicohadfolloweda
more `mixed' strategy in its own classic period, 194862, as the chapter by
Cardenas in this volume make clear, and effectively moved backwards to a
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
moretraditionalemphasisonimportsubstitutioninitsmaturephase.Among
the countries not covered in this volume, Peru is another example of such a
backward shift. Central America effectively never moved into the advanced
stages of import substitution and, rather, followed still another `mixed stra-
tegy':importsubstitutionatthesubregionallevel,superimposedontheprim-
aryexportmodel.
As we have pointed out so far, at some point short-term macroeconomics
totallyreplacedlong-termstrategiesinmanycountries,and/orthelatterwere
swamped by rising social conflicts which then became the dominant force
determining social as well as economic events. However, the Chilean case
aside,theolderlayersofhighprotectionandthemorerecentofexportpromo-
tioninthosecountrieswheretheywerefullydevelopedsurviveduptothe
liberalization period which took off in the mid-1980s and speeded up in the
regioninthe1990s.Althoughdeeplydefendedbytheirbeneficiaries,theywere
increasinglydetachedfromarealdevelopmentstrategyofanysort.
Theprocessofindustrializationwenthandinhandwithothermajorchanges
occurring in Latin American societies and economies. The restructuring that
took placein the1930s (see theprevious volume) continued during andafter
the war. The state had acquired a number of policy instruments, such as
monetary discretion, and a more diverse tax base, while the abandonment of
thegoldstandardhadleftfinancialauthoritiesfreetomanoeuvretheexchange
rate.Insomecases,thisledtotheillusionthatmoneycreationwasoneandthe
sameaswealthcreation.Thestatehadalsotakenonitsshoulderstherespon-
sibilityforpromotingeconomicdevelopment.Itdidsobyenlargingeconomic
infrastructure in someinstancesor by establishingstate enterprises, either for
defencepurposesasinArgentinaunderPero nortoassuredomesticsuppliesof
certain`strategic'inputs,asinBrazil,Mexicoandmanyothercountries.Italso
createdpublicsectorcommercialanddevelopmentbanks,orencouragednew
privatefinancialinstitutionsandforcedallofthemtochannelfundstopriority
sectors. The role of the state was also enlarged to incorporate new entrepren-
eurs, often actually `creating' them under the umbrella of large government
contractsrelatedtotheconstructionofroads,publicutilities,damsandsoon,
whichthereforeimpliedacertainworkingrelationshipthatoftensurpassedthe
businesscomponentandenteredthepoliticalsphere.
Inseveralcountries,thepostwarperiodwitnessedanaccelerationofpopula-
tiongrowthcausedbythedeclineofinfantmortalityrates.Withinafewyears,
that implied a corresponding increase in the labour force, creating enormous
pressures for the economy to grow. The Latin American population increased
from 126 million in 1940 to 210 million twenty years later, and 139 million
more wereaddedinthe1960sand1970s. Duetofalling mortalityratesanda
lagged transition in fertility, population growth peaked at an average level of
2.72.8 percent a year from the mid-1950s to the mid-1960s, with over 3 per
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
12 Introduction
centinseveralcountries.Indeed,LatinAmericaandAfricafacedthestrongest
populationpressures,afactwhichprobablycontributedtodepressedgrowthin
percapitaGDP(seeTable1.1).Citiesalsomultipliedinsize,creatingincreasing
demandsonpublicutilities,schools,hospitalsandthelike.Therateofurban-
izationwas41.6 percentin1950,buthadalreadyreached5673percentinthe
four `success stories' of the export age (Argentina, Chile, Cuba and Uruguay).
By1970,57.7percentofthepopulationlivedinurbanareas(75to82percent
inArgentina,Chile,UruguayandVenezuela)andmostLatinAmericancapitals
hadbecomelarge,sophisticatedmetropolitanareas.
7
Thiswasaccompanied,as
we will see, by significant improvements in human development and the
incorporationofaproportionofthepopulationintothebenefitsofamodern
economy, but also by the persistence of high levels of inequality in the dis-
tributionofincomeandassets.
TheroleofECLAC
AconsensusamongparticipantsintheworkshopsisthattheroleofECLAChas
been exaggerated in the mythology of ISI unfortunately in part because for
some in the US State Department and elsewhere Prebisch became paradox-
ically,givenhisratherconservativebackgroundidentifiedasthespokesman
for ideas coming out of Moscow and Prague, and thus a victim of Cold War
paranoia.
8
In fact, as we haveseen, the development of many of the building
blocksofstrongstateinterventiontookplaceasaspontaneousprocessduring
the èmpirical' phase of import substitution and even during the export age,
and corresponded to international practice at the time. Thus, all clearly pre-
dated the creation of ECLAC in 1948. Large governments were also not an
innovation of ECLAC or Latin America, either, but rather a widespread inter-
nationalpracticeinthepostwarperiod.
However, the wider importance of ECLAC is undoubted. It helped to bring
aboutalmostthefirstglimmeringsofcontinentalself-awareness,inparticular
by providing theoretical and empirical support to the evolution of a regional
identity.
9
It played a significant role in spreading modern economic analysis
and statistical techniques, and in developing institutions such as schools of
publicadministrationandplanningministries.Induecourse,itwasimportant
indevelopingthenotionofprotectionattheregionallevel,anditthusplayed
a central role in the design of the Latin American Free Trade Association
ALALC,accordingtoitsSpanishabbreviation,laterLatinAmericanIntegration
Association the Andean Group and, most notably, the Central American
CommonMarket,asthechapterbyCohenmakesclear.Italsohelpedtopress
forreformsinthesocialarea,manyofwhichsetthestagefortheAlliancefor
Progressinthe1960s.
Attheconceptuallevel,ECLACgavealogictoandhelpedtorationalizethe
strategy of development which had been springing up in the region. It built
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
uponatheoryofcapitalaccumulation,notatheoryofstaticefficiency.Indus-
trial development was primarily seen as the mechanism to transfer technical
progress to peripheral countries in the face of the slow growth in primary
commodity markets. Capital accumulation, structural changes associated
with industrialization and the transfer of technological innovations from the
centrewere,thus,seenasthreefacetsofauniqueprocess.Itwasalsoseenasa
mechanismtoabsorblabourdisplacedfromruralareasandtomanagemacro-
economic constraints associated with foreign exchange scarcity. As Fitzgerald
makes clear, these ideas on industrialization were different from the infant
industry argument, as well as from those associated with the role of dynamic
scaleeconomiesinmorerecenttheoriesoftradeandgrowth.Tariffprotection
was seen as a way to guarantee a stable demand for new sectors and public
sector investment as the way to guarantee the provision of skills and infra-
structure.Theproblemsofthisstrategywerealsorecognized,andsomepolicies
weredesignedtoovercomethemregionalintegration,the`mixed'strategy
butotherswereseenasintrinsictoperipheralindustrialization.
The institution also played a negative role at the level of ideas and percep-
tions. Most important, it never developed, during our period of analysis, the
ideasofprotectionasatime-boundinstrumentandtheneedtotieincentives
toperformance;thelatterplayedan importantroleinrapidEastAsianindus-
trialization, according to some interpretations. It was thus lenient on the
`geological' layer over layer pattern which protection adopted and the
inefficiencies and political economy which it generated. It also tended to
view new developments in international trade with a pessimistic lens. None-
theless, it shifted gradually in the 1960s and, particularly, in the 1970s, to a
favourable view of export promotion (Bielschowsky, 1998). Several times it
manifested distrust for relative price adjustments and, particularly, for active
foreign exchange rate policies, which were certainly instrumental in encour-
agingstructuraldiversification(seebelow).
However,itshouldbeemphasizedagainthatECLAC'sviewsonindustrializa-
tion, protection and state intervention agreed with contemporary wisdom.
Moreover,intheregionaldebateitalwaysremainedinthecentre,moremod-
erate than the more radical dependency school visions that sprang from its
work,aswellasMarxistthinking.Itscharacterasanintergovernmentalinstitu-
tioncertainlydeterminedthat.
Whatwerethealternatives?
A distinct conclusion of the project, which comes clearly from the previous
sections,isthatitishardtoimaginecountrieschoosingotherthantheydidin
theearlypostwar years.Ontheonehand,atthetime,anddespitefavourable
short-term trends, continued reliance on primary exports did not look like a
goodalternativeinthelightofpasttrendsandinstabilityinthetermsoftrade.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
14 Introduction
Indeed,afterthemid-1950sthereneweddownwardtrendofcommodityprices
generatedanewwaveofbalanceofpaymentscrises,whichfurtherencouraged
regional protectionist trends. The coexistence of GATT with renewed agricul-
tural protectionism was an additional argument, particularly for those coun-
tries which exported temperate zone agricultural products. The export of
manufacturingproductsfromdevelopingcountrieswasaverydistantalternat-
ive, although the Mexican experience during the war had made a first start.
Highlevelsofprotectionwerestilltheruleintheindustrializedcountriesandit
wasclearlynecessarytoundergoalongperiodofcontinuousgrowthininter-
nationaltradetoconvincecountriesandauthoritieswhichhadlivedthrough
itscollapsetoregarditasareliablealternative.ThefailureoftheInternational
TradeOrganizationwasalsoawrongsignal,aswellasthatoftheotherinstitu-
tions Keynes hadenvisaged as accompanying thefamiliar Bretton Woods cre-
ations, including a commodity buffer fund and a soft aid facility through the
UnitedNations.Afurtherrealityintheearlypostwaryearswasinconvertibility,
as a reflection of the `dollar shortage'. Sterling export earnings, in particular,
couldnotbeconvertedintodollarstobuyUSmachinery.
Ontheotherhand,theoptionforindustrializationhadalreadybeenmadede
factointhelargercountries,inanunmistakablesensesincethe1930s,andthis
was clearly reflected in the positions of emerging industrial groups in Latin
America and in their growing political weight. Politically, also, urban growth
and continuing migration were making industrialization crucial. During the
SecondWorldWar,theUSAbackedthepushforexpandedstateinvolvementin
economicaffairsandevenadirectroleinindustry.Afterthewar,theofficialUS
line, and the framework for the new Bretton Woods institutions, was to pull
back from state involvement but, as Webb's chapter reveals, the international
financialinstitutionswerenothardonthatline,atleastwithrespecttodevel-
oping countries. As we have seen, the multinationals' views on protection
changed rapidly as they began to perceive the extent of the profits to be
made in `tariff-hopping'. Under Truman's Four Point Plan, launched in 1949,
the industrialization of the developing world was an important policy object-
ive, and the technical missions that followed were intended to design devel-
opment plans including the devising of foreign exchange controls and tariffs
(MaxfieldandNolt,1990).Finally,aswehaveseen,economicideasatthetime
and indeed for a long period continued to favour state intervention, and
indeed,inthisregardaswellasinpractice,probablytherealoptionwascentral
planningthatis,moreratherthanlessstateinvolvement.
Adifficultpartoftheassessmentofalternativesconcernstheroleofexternal
pressures.WastheenthusiasmforprotectionamongLatinAmericanadminis-
tratorsmainlyaresultofexclusionfromthefinancialaidavailableforpostwar
reconstruction, rather than a response to endogenous forces within Latin
America?Fitzgeraldarguesinhiscontributiontothisvolumethattherewasa
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
markedwillingnesstoconformtofreetraderulesaslongasitseemedthateven
a Marshall Plan for Latin America was not beyond the pale. But by Havana it
was clearly beyond the pale, and countries rapidly changed their tune and,
moretothepoint,theirpractice.
The situation had, of course, changed considerably by the 1960s when the
internationaleconomydidstarttoofferanalternativeinarealsense.Indeed,
the shift to what we have called a `mixed model' indicates that many Latin
American countries did not ignore it. The country studies provide, indeed,
many instances of this change of direction. Chile was the first to introduce
thecrawling pegin1965, andcombinedthiswith anextensivesetofpolicies
developedoverthenextfewyears,includingtaxrebates,tariffreductionsand
rationalization of the tariff structure. CORFO was active in these years with
programmes that laid the basis for subsequent exports in fishery, forestry and
freshfruits.Colombiahadadoptedinthelate1950sandearly1960saseriesof
export incentives that were rationalized in 1967, when the crawling peg was
introduced and wide-ranging institutional reforms in the foreign trade and
exchange regimes were adopted. Mexico actually introduced its maquiladora
programmeinthesameyearasTaiwan(1965).Theintroductionofthecrawl-
ing peg in Brazil in 1968 had been preceded by export incentives since 1965.
Venezuela had certainly not ignored her opportunities in oil throughout. At
thesametime, inthesmallercountries,theopportunitiesinmore traditional
commoditymarketshadnotbeenignored,andexportsofcoffee,bananasand
cotton were generating export dynamism in Central America from the mid-
1950s.ThisisalsotrueofsomeSouthAmericancountriesnotincludedinthe
casestudiesEcuador,inparticular.Indeed,aswewillsee,therewasasenseofa
`goldenage'inmanyLatinAmericancountriestowardstheendofourperiodof
analysis,fromsomeyeararoundthemid-1960stothefirstoilshock.
Examplesabound,however,inwhichsenioradvisersrecommendeddevalu-
ations, tax reforms or other policy measures in circumstances where, with
hindsight,theinstitutionalconditionsandmacro/microrelationswouldhave
allowed successful policy alternatives, but the options were rejected for polit-
ical reasons. Many individual examples can also be quoted concerning the
pushing of import substitution programmes when it was already clear that
market size could not sustain them. Obviously, the very ambitious industrial
investmentplansofBrazil,MexicoandVenezuelafollowingthefirstoilshock
are the most remarkable examples, although some actually partly paid off in
thelongterm,evenintermsofexportexpansion.Outsideourcasestudies,Peru
isprobablythemostoutstandingexampleofacountrywhichoptedforamore
inward-oriented policy in the late 1960s, against regional and international
trends and without adequate export backing, and failed. Decisions to plan
import substitution at a subregional level by a political allocation of the pro-
duction of certain goods to specific countries, in the Andean Group and the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
16 Introduction
Central American Common Market, were definite failures. But the most frus-
trating experience was, obviously, the resistance to liberalizing competitive
imports in regional and subregional integration agreements the Central
American Common Market aside. It is a historical paradox that the great
invention of ECLAC to rationalize import substitution, regional integration,
succeeded only when fully-fledged trade liberalization was undertaken in the
1990s, that is, when import substitution as a development strategy had
beenabandoned!Itis,indeed,adoubleparadox,asorthodoxthinkingpushing
for trade liberalization considered integration agreements to be close to an
aberration.
More generally, the major problem was the reduced flexibility that built up
duringtheyearsofstate-ledindustrialization.Thisiscloselyassociatedwiththe
politicaleconomywhichcharacterizedtheprocess.Aswehavepointedout,a
remarkablefeaturewastheabsenceoftheideasofprotectionasatime-bound
instrumentandtheneedtotieincentivestoperformance.Thiswasobviously
supported by the political economy that characterized the process, in which
protection to a specific sector was a permanent conquest. The pessimistic
exportlenswasalsoafeature,althoughdecreasinglysoinmostcountries.An
interestingcounterfactualexercisewouldbetoaskwhether,intheabsenceof
the oil shock and the `macro mess' which followed, the model would have
evolvedinadifferentdirectionasaresult,forexample,ofthesuccesswhich
the`mixedmodel'wasstartingtoshowinsomecountries.Nonetheless,bythat
timemacroeconomicsandpoliticsweretakingoverthescene.Wewillreturnto
thisanalysislateron.
Theoverallresults
10
Thebenefits(withsomequalifications)
LatinAmericaneconomicperformanceduringthethreedecadesthatfollowed
theSecondWorldWarwasremarkable,inducingawidespreadtransformation
ofsociety.Ontheonehand,continentalGDPgrewat5.6 percentperyearin
194574,or2.7percentpercapita.
11
ThiswasunparalleledforLatinAmerica,
just as the developed country record was the fastest in recorded time. In the
intermediate period between this phase of long-term growth and the debt
crisis, growth was only slightly slower 5 and 2.5 per cent, respectively
12
though it was more than two percentage points lower than the peak rate of
growth of 196874 (see below). The manufacturing sector was the engine
of growth, growing 6.8 per cent a year in 194574, reaching a peak share of
GDP of 26 per cent in 1973, 7 percentage points more than in 1945. Sectoral
behaviour is vividly summarized in Figure 1.1b, where industry gains at the
expense of primary production, as also occurred in most developing areas in
theworld.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
%

o
f

G
D
P

a
t

1
9
8
0
s

c
o
n
s
t
a
n
t

p
r
i
c
e
s
1
9
3
9
1
9
3
9
1
9
4
1
3%
26%
25%
24%
23%
22%
21%
20%
19%
18%
17%
16%
Figure1.1 GDPgrowthandcomposition
Source: ECLAC.
The temporal pattern of GDP growth was a very rapid recovery in the early
postwarperiod,whichwasinterruptedbyaseriesofbalanceofpaymentscrises
in the late 1940s and early 1950s. Growth then stabilized for the region as a
whole around a rate somewhat above 5 per cent. In the late 1960s and early
1
9
4
1
1
9
4
3
1
9
4
5
1
9
4
7
1
9
4
9
1
9
5
1
1
9
5
3
1
9
5
5
1
9
5
7
1
9
5
9
1
9
6
1
1
9
6
3
1
9
6
5
1
9
6
7
1
9
6
9
1
9
7
1
1
9
7
3
1
9
7
5

1
9
4
3
1
9
4
5
1
9
4
7
1
9
4
9
1
9
5
1
1
9
5
3
1
9
5
5
1
9
5
7
1
9
5
9
1
9
6
1
1
9
6
3
1
9
6
5
1
9
6
7
1
9
6
9
1
9
7
1
1
9
7
3
1
9
7
5
1
9
7
7
1
9
7
9
1
9
8
1
1
9
8
3
1
9
8
5

1%
1%
3%
5%
7%
9%
Current growth rate
Three years centred moving average
1939-45 Latin Amrica six largest economies
Three years centred moving average L.A. (6)
17 EnriqueCardenas,Jose AntonioOcampoandRosemaryThorp
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
18 Introduction
1970s there were clear signs of acceleration (see Figure 1.1a), and the period
P
e
r

c
a
p
i
t
a

G
D
P

g
r
o
w
t
h
,

1
9
4
5
7
4

(
%
)

ended in a `golden age' with peak rates of growth in 196874 7.2 per cent
ayear,or4.2percentpercapita.Thisreflectedafairlybroadpatternofgrowth
acceleration. There was also a discernible reduction in the volatility of
growthinthelastphasesofthepostwarboom.
Growth was not uniform among countries, either. A certain `convergence'
was achieved due to the fact that the three `success stories' of the export age
(Argentina, Chile and Uruguay) grew at a slower rate. This is also true if the
fourth`successstory'oftheexportage,Cuba,wasaddedtothepicture,butno
comparable long-run GDP series for this country exists.
13
These cases aside,
however,therewasnoclearconvergencepattern.Veryfastgrowthwasexperi-
encedbybothcountrieswithlowinitialpercapitaGDP,notablyBrazil,butalso
by countries which started with a medium-level per capita GDP, particularly
Mexico and Venezuela. On the other hand, Bolivia, Honduras, Paraguay and,
particularly, Haiti constitute a sample of countries with low initial per capita
GDP which grew at slow rates and, thus, performed contrary to convergence
patterns(Figure1.2).
Along with growth came labour productivity gains, which translated into
higherrealwagesofthoseemployed,andastrongerlabourunionmovement.
Labour productivity increased at an annual rate of 3.4 per cent in 195073,
which exceeded that of the USA but was below those typical in the most
dynamic developed market economies and the Asian NICs.
14
This reflected
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.5
1.0
Argentina
Bolivia
Brazil
Colombia
Costa Rica
Chile
Ecuador
El Salvador
Guatemala
Haiti
Honduras
Mexico
Nicaragua
Panama
Paraguay
Peru
Dom. Rep.
Uruguay
Venezuela
0 100 200 300 400 500 600 700 800
Per capita GDP, 1945
Figure1.2 Convergencepatterns
Source: ECLAC.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
bothcapitalaccumulationandtechnicalchange.Accordingtoarecentestim-
ate,realfixedcapitalstockgrewatanannualratecloseto6 percentin195073
and, indeed, up to the debt crisis. Total factor productivity increased at an
estimated growth of2.5 percent per year in 195073, slightly above the rates
typicalintheUSAbut,again,belowtheaverageoftheindustrializedeconom-
iesand,particularly,the`periphery'oftheEuropeancommunityandtheAsian
NICs.
Productivitygrowthwas,infact,partofalargerprocessoftransformationof
productivestructures,whichinvolvedthedevelopmentofsubstantialtechno-
logical capabilities including technical knowledge in the narrow sense aswell
as quality control practices, entrepreneurial, managerial and labour skills,
working habits and other behavioural patterns. This is emphasized in the
contributions by Katz and Kosacoff to this volume, as well as in Katz (1987,
1998)andTeitel(1993),amongothers.Astheyemphasize,thesetechnological
capabilities built up through an èvolutionary', path-dependent process in
which accumulated experience played the crucial role. It involved `learning
by doing', adaptation and induced generation of technologies as well as the
developmentofmodernadministrationtechniquesandworkrelations.Indeed,
the development of the modern entrepreneurial and labour classes of Latin
America,whichdemonstratedaremarkableabilitytosurvivethemacroshocks
of the 1980s and to adapt to globalization in the 1990s, is indistinguishable
fromstate-ledindustrialization.Inanarrowersense,thedevelopmentofman-
ufacturing exports since the 1960s and their acceleration since the 1980s can
hardlybeunderstoodwithouttakingintoaccountthecapabilitiesaccumulated
inthepreviousphaseofindustrialization,thatis,asasuccessfultransitionfrom
importsubstitutiontoexports(TeitelandThoumi,1986).Thiswasalsotrueof
someagriculturalexports,asOcampoandTovarillustrateintheircontribution
onColombiaforthisvolume,butasimilarstorycanbetoldofBrazilandother
countries.
Institutional development underwent a similar, dynamic process.
15
This
conceptisusedhereinabroadsense,toincludenotonlythecreationofpublic
and private agencies, but also the development of norms, rules and customs
that determine the behaviour of social actors. Modern state institutions in
Latin America, both in the economic and social areas are, indeed, largely the
productofthisstageofdevelopment.Developmentbanks,forexample,came
into being at variable dates from the late 1930s on. Some of them were to
experience decay, as in Argentina, but others remained strong and acquired
expertise, as with the BNDES (Banco Nacional do Desenvolvimento Econom-
ico) in Brazil. The ability to innovate in the exchange rate area and in the
designofexportpromotioninstrumentsinthetransitiontothe`mixed'model
inmanycountriesisalsoademonstrationofsuccessfulinstitutionalflexibility.
In the private sector, entrepreneurial associations, at a formal level, and the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
20 Introduction
extension of modern business and labour relations, at a more informal one,
experienced significant steps forward during this period. The Southern Cone
aside,modernsocialsecurity,health,educationalandlabourtraining systems
were also largely the product of state-led industrialization; in the Southern
Cone countries, these systems also became much more comprehensive than
before.
Agriculturewasnotabsentfromthisstoryofproductivitygrowthandinsti-
tutionaldevelopment.Despiteitslowergrowthrecord,agriculturalproduction
grew at an annual rate of 3.5 per cent in 195075, by no means a slow rate
relative to international patterns.
16
Although macroeconomic and pricing
policiesgeneratedatmanyjuncturesastrongbiasagainstit,thedevelopment
of new state institutions to support agriculture was remarkable in most coun-
tries. These included credit, marketing and technological services, sometimes
evenmoredevelopedthanthosedesignedtosupportindustrialdevelopment.
Thisisparticularlythecaseoftechnologicalserviceswhichwerequiteeffective
in introducing new crops and improving varieties and cultural practices, not
least those introduced by the `green revolution'. As a result of that, non-
traditional agricultural products were added to the export basket in several
countries(soybeans,orangejuice,cutflowers,cottonandsoon).
17
Taxpolicy,
including lower tariffs on agricultural inputs and machinery, was generally
supportive to agriculture. An open agrarian frontier and large state-financed
irrigation infrastructure were also crucial for growth. The expansion of the
agrarian frontier, as well as the nature of some new agricultural technologies
andofmanufacturing developmentingeneralgeneratedinmanycases,how-
ever,animportantenvironmentaldegradationwhichwasnotfacedbygovern-
mentpoliciesduringourperiodofanalysisand,indeed,onlybecameasource
ofconcernmuchlater.
Also, the benefits of agricultural expansion were quite uneven across coun-
tries. Among those studied in this volume, Brazil, Colombia, Venezuela and
most Central American countries stand out as stories of rapid agricultural
growth,andArgentinaandChileascountrieswithslowgrowth.Mexicounder-
wentaninitialsuccessintheearlypostwarperiodfollowedbyasignificantlag
inagriculturaldevelopmentinthemoreadvancedstagesofstate-ledindustri-
alization.Benefitsalsoconcentratedinlargeandmedium-sizedfarms.Thus,the
initialstructuralheterogeneitythatcharacterizedruralareaswidenedthrough-
out our period of analysis. Agrarian reforms were limited in their scope, not
leastbecauseofstrongresistancebypowerfuleconomicandpoliticalinterests.
Contrary to the fears that were constantly expressed, employment genera-
tionwasreasonablydynamic.Thechallengeswereconsiderable,asbothpopu-
lation growth and urbanization were very rapid. The non-agricultural labour
forcegrew atan annual rate of 4 percent in 195080, faster than that experi-
encedbytheUSAin18701910.
18
Althoughtheveryrapidgrowthoftheurban
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
labourforcewasreflectedingrowinginformalityinthecities,thereductionof
employment in traditional agriculture took place at a faster rate, generating a
reduction in overall underemployment, from 46 to 38 per cent in 195080,
according to the estimates by Garc a and Tokman (1984). In a sample of 14
countries analysed by these authors, nine followed this pattern, some with
more (Brazil, Colombia, Costa Rica, Mexico, Panama and Venezuela) and
others with lesser intensity (Chile, Guatemala and Peru). Also, despite rising
informality, Argentina and Uruguay kept levels of underemployment signific-
antlybelowtheregionalaverage.
Finally, the standard of living experienced asignificant increase during this
stage of development. Indeed, an estimate of living standards relative to the
USA,whichfollowingearlyversionsoftheUNDP's(UnitedNationsDevelop-
ment Programme) Human Development Index combines GDP per capita,
adultliteracyandlifeexpectancy,indicatessignificantadvancementsincethe
1920sandafairlyrapidcatch-upduringtheperiodofstate-ledindustrialization
(Figure 1.3).
19
The record in terms of poverty reduction and, particularly,
income distribution was certainly more mixed, but unfortunately no overall
estimatesorevencomparablestatisticsexistfortheperiodasawhole.Poverty
probablydeclined,althoughwithsignificantlagsinseveralcountries,whereas
40
45
50
55
60
65
70
75
LA(6)
LA(13)
LA(20)
1
9
0
0
1
9
0
3
1
9
0
6
1
9
0
9
1
9
1
2
1
9
1
5
1
9
1
8
1
9
2
1
1
9
2
4
1
9
2
7
1
9
3
0
1
9
3
3
1
9
3
6
1
9
3
9
1
9
4
2
1
9
4
5
1
9
4
8
1
9
5
1
1
9
5
4
1
9
5
7
1
9
6
0
1
9
6
3
1
9
6
6
1
9
6
9
1
9
7
2
1
9
7
5
1
9
7
8
1
9
8
1
1
9
8
4
1
9
8
7
1
9
9
0
1
9
9
3
1
9
9
6

LA(6): Argentina, Brazil, Mexico, Chile, Colombia, Venezuela
LA(13): Latin America LA(6) Cuba
LA(20): Latin American countries
Figure1.3 Relativelivingstandardindex(relativetoUSA)
Source: AstorgaandFitzGerald(1998).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
22 Introduction
income distribution remained highly skewed in most countries and experi-
enced opposite patterns towards the end of the period in different countries.
ECLAC's first overall estimate of poverty, available for 1970, indicates that 40
percentofLatinAmerica'shouseholdswerepoor;thisproportiondeclinedto
35percentin1980,alevelwhichhasnotbeenreachedagaininthefollowing
decades. In any case, the benefits from development remained significantly
concentrated at the end of our period of analysis. A correct rough statement
wouldbethatsocialprogressbasicallytrickleddowntoa`middleclass',which
included the urban population employed by the state and large and middle-
sized private firms and some small entrepreneurs, with the extent of that
`middleclass'varyingconsiderablyfromcountrytocountry.
Thecosts(andthecorrespondingqualifications)
Themajordrawbackofstate-ledindustrializationwas,undoubtedly,itsinabil-
itytofullyexploitthebenefitsfromdynamicworldtradeinthepostwarperiod.
The postwar boom in commodity prices led to an increase in Latin America's
shareinworldtrade,butthesharewasbacktoprewarlevelsbythelate-1950s.
Since then, it fell dramatically, from around 7 per cent of world exports to
slightly over 4 per cent in the 1970s.
20
This implied not only a relatively
marginal participation in the dynamic world trade in manufactures, but also
afallinLatinAmerica'sshareincommoditymarkets,hertraditionalstrength.
Also,asFigure1-4makesclear,theratioofexportstoGDPfelldramatically
throughthe1950sand1960s,beforerisinginthe1970s.However,thisglobal
picture is strongly affected by the largest countries and, particularly, by Brazil
andVenezuela.Amorecompletestoryofexportperformancebysubperiodand
countrysizeisprovidedinTable1.2.Afewcountriesaside(Ecuador,Peruand
Venezuela),the194555exportrecordwasdismal;risingcommoditypricesup
to the mid-1950s generated, however, an environment of rising export pur-
chasing power, despite very slow real export growth. However, the picture
improved significantly from the mid-1950s. Indeed, a simple average of
quantum export growth rates exceeded that of GDP growth in 195565 and
wasroughlysimilartothelatterin196573.Forsmalleconomies,exportexpan-
sionexceededGDPgrowthbyasubstantialmarginfromthemid-1950sto1973.
Thisisconsistentwiththeideathat,forthesecountries,importsubstitutionwas
effectivelysuperimposedontheprimaryexportmodel.Interestingly,compara-
tive export growth patterns became quite diverse among large and medium-
sizedcountriesin196573andamongsmallcountriesin197380.
The policy shifts of the 1960s in several countries towards a `mixed model'
had in fact positive effects in terms of export dynamism. Rising intraregional
tradeassociatedwithintegrationagreementshadasimilareffect(seebelow).In
particular, manufacturing exports rose from 3.6 per cent of total regional
exports in 1960 to 16.3 per cent in 1975 and 17.4 per cent in 1980, growing
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
(a) Latin America's exports as % of GDP
(b) Exports as % of GDP by country sizes (simple averages)
Figure1.4 Exportsas%ofGDP
Source: ECLAC.
at an annual rate (in value terms) of 22 per cent throughout these two dec-
ades.
21
Inthelargestcountries,thisincludedexportsofmachineryandequip-
ment, primarily to other Latin American countries, and even of pure
technologyintheformoflicensesandengineeringservices(Katz,1998).New
agricultural products soybeans, orange juice, fresh fruits and vegetables, cut
flowers,andsoonwerealsoaddedtotheexportbasketinseveralcountries.
1
9
5
1
1
9
5
1
1
9
5
3

1
9
4
5
1
9
4
7
1
9
4
9
1
9
8
7
1
9
8
9

5
10
15
20
25
30
35
1
9
4
5
1
9
4
7
1
9
4
9
1
9
8
7
1
9
8
9

%

o
f

G
D
P

a
t

c
o
n
s
t
a
n
t

1
9
8
0

p
r
i
c
e
s

Large countries
Medium countries excluding Venezuela
Small countries
1
9
5
3
1
9
5
5
1
9
5
5

1
9
5
7
1
9
5
7

1
9
5
9
1
9
5
9

1
9
6
1
1
9
6
1
1
9
6
3

1
9
6
3
1
9
6
5
1
9
6
5
1
9
6
7

1
9
6
7
1
9
6
9
1
9
6
9

1
9
7
1
1
9
7
1
1
9
7
3

1
9
7
3
1
9
7
5
1
9
7
5
1
9
7
7

1
9
7
7
1
9
7
9
1
9
7
9

1
9
8
1
1
9
8
1

1
9
8
3
1
9
8
3
1
9
8
5

1
9
8
5

10
12
14
16
18
20
22
24
26
28
30
%

o
f

G
D
P

a
t

c
o
n
s
t
a
n
t

1
9
8
0

p
r
i
c
e
s

Total
Total excluding Venezuela
Simple average excluding Venezuela
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
24 Introduction
Table1.2 Exportgrowthrate(%)
194555 195665 196673 197480
Largecountries 2.0 3.1 7.2 6.9
Argentina 0.5 2.5 5.0 3.1
Brazil 0.2 3.6 13.3 6.2
Mexico 4.8 3.1 3.4 10.0
Medium-sizedcountriesexcluding 1.6 5.3 3.3 5.8
Venezuela
Colombia 2.3 3.4 5.9 4.6
Chile 0.6 3.5 1.8 12.5
Ecuador 6.5 7.1 13.7 2.3
Peru 4.2 7.8 2.2 6.4
Uruguay 3.2 5.6 3.0 9.9
Smallcountries 0.6 6.1 8.0 2.1
Bolivia 4.1 1.2 11.5 3.9
Paraguay 4.1 8.0 3.3 11.7
CostaRica 5.3 5.7 12.3 2.5
ElSalvador 3.5 9.8 4.6 3.9
Guatemala 0.9 10.0 7.9 4.9
Honduras 0.6 9.4 6.8 3.7
Nicaragua 11.3 7.6 5.6 4.9
Venezuela 4.2 7.8 2.2 6.4
Simpleaverages
largeandmedium-sizedcountries 1.8 4.6 4.7 6.3
excludingVenezuela
smallcountries 1.5 7.0 7.4 2.6
totalexcludingVenezuela 1.7 5.8 6.1 4.4
total 1.8 5.8 5.5 4.7
Memo:
GDPgrowthrates
Simpleaverages
largeandmedium-sizedcountries 5.2 4.3 5.5 4.7
excludingVenezuela
smallcountries 4.6 4.5 5.3 4.1
totalexcludingVenezuela 4.9 4.4 5.4 4.4
total 5.2 4.5 5.3 4.3
Source: ECLAC.
The inability to rationalize the complex `geological' pattern of import protec-
tionhadimportant costs.Forestablishedindustries,importprotectionceased
toplayapositiveroleasanincentiveforcapitalaccumulation,andincreasingly
became a pure source of rent and/or a useful protection from cyclical or
permanent exchange rate overvaluation. It distorted, moreover, the structure
of incentives which the model itself required, particularly the relative price
shifts necessary to induce import substitution in new sectors and exports in
mature manufacturing and primary producing activities. It implied that, to
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
encourage additional import substitution, the only effective instruments
becamequantitativerestrictionsandevenoutrightprohibitionofcompetitive
imports. This cumbersome system of protection was, thus, self-defeating in
terms of its explicit objectives: to generate capital accumulation and an ade-
quate diversification of the structure of production. Moreover, rather than
reduce dependence on imported inputs and technology, the strategy may
have increased it, as ECLAC made systematically clear in its analyses at the
time.Thiswasthebasisfortheattemptstorationalizethestructureofprotec-
tionthatsomecountriesundertooksincethe1960s.
Protection was also an obstacle to regional integration. As we have pointed
out, this was certainly paradoxical, as integration was viewed as the major
means to reduce the costs from import substitution. It certainly facilitated
the dynamic growth of intraregional trade and manufacturing exports, espe-
cially in the 1960s and 1970s.
22
Nonetheless, the benefits that ECLAC envi-
sionedfromthatprocesswereonlycapturedpartially,throughthecreationof
larger markets for complementary goods. The other part, the benefit from
restricted competition in subregional or regional markets, was not
captured,astheliberalizationofcompetitivegoodswasgenerallyverylimited
a major exception was the Central American Common Market. Moreover,
theattemptstoplantheproductionofnewcomplementaryinvestmentsaimed
at the regional or subregional markets were, almost invariably, outstanding
failures.
The history of exchange rate management during the period analysed had
complexitiesofitsown.Theriseofmultipleexchangeratesystemsintheearly
postwarperiodmadethemafairlyclosesubstitutefortradepolicy.Theability
to implicitly tax competitive imports and traditional exports and subsidize
complementaryimportsusingexchangeratesasaninstrumentwasattractive,
bothinspeedandadministrativeconveniencerelativetothecomplicationsof
legislation. Export taxation was a particularly difficult issue and, indeed, in
mostcountriesdiscriminatoryexchangeratesweretheonlymeansavailableto
achieveitorrathertohideit.Multipleratesservedtomaketheprotectionist
regimemorecomplex,furtherconfusingthepricesignalswhichitwastryingto
transmit.Inthisarea,however,thereweresignificantimprovementsfromthe
mid-1950sunderstrongIMFpressureand,particularly,inthe`mature'stage
whenmost multipleexchange rateregimes were eliminated orsimplifiedand
thecrawlingpegwasintroducedinseveralinflationarycountries.
Theshifttowardsamoreflexibleexchangerateininflationarycountrieswas
instrumental to avoid overvaluation and to encourage export diversification.
Indeed, at this stage, as in the 1980s, it was clearly demonstrated in several
countries in the region as in Asia that dynamic export growth could be
compatiblewithimportprotection,providedthatacompetitiveexchangerate
wasavailable.Themoreorthodoxview,accordingtowhichthedismantlingof
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
26 Introduction
protection would automatically lead, through the induced real exchange rate
devaluation, toexportpromotion,wasradically discredited bytheexperience
oftheSouthernConecountriesinthelate1970sandbyotherexperienceslater
on, in which trade liberalization was accompanied by capital inflows which
led,duringseveralyearscrucialtothetradereforms,toexchangerateappreci-
ation rather than devaluation. The long transition to a more competitive
exchange rate was thus mediated, in these cases, by a balance of payments
crisisgeneratedbylargeovervaluationandcurrentaccountdeficitsintheearly
liberalizationperiod.Chileinthe1970sand1980sisanotableexampleofthis
lengthyandcostly`transitionpath'.
Giventhestrongemphasisonovervaluationasafeatureofimportsubstitu-
tion industrialization, it is paradoxical that few analyses of the exchange rate
dynamics during the period of state-led industrialization are available. A rare
exampleisJrgensenandPaldam(1987),whichanalysestheevolutionofthe
real exchange rate in the eight largest Latin American countries in 194685.
Theirconclusionsindicatethattherewasnotalong-termappreciationtrendof
therealofficialexchangerateinanyofthesecountries.Onthecontrary,intwo
Brazil and Venezuela there was a long-term real devaluation and, most
importantly, in several of them, there was a discrete devaluation of the real
exchange rate in the early postwar period which had permanent effects.
23
Even if it is argued that this early devaluation was only partially compens-
ating for a strong initial overvaluation, these conclusions indicate that
the deepening of protection was accompanied, in most countries, by real
devaluation, basically achieved by means of discrete jumps with permanent
relative price effects. A more worrisome feature was, thus, the very sharp
fluctuationsaroundthelong-termtrendsoftherealexchangerate,particularly
in the more inflationary economies. This certainly had adverse effects in two
areas: it generated an additional demand by import-competing sectors for
protection, as a defence against cyclical real exchange rate appreciation, and
it made exporting profits highly unstable, generating adverse incentives to
24
export.
Exchange rate instability was associated, as we have pointed out, with
domestic inflation. In inflationary economies, the temptation to delay deval-
uationtoavoiditsinflationaryeffectswasveryhighevenfortheIMFinsome
conjuncturesafactwhichgenerallyledtosharperdevaluationslateron.Prior
tothe1970s,however,inflationtendedtobemoderatealthoughwithsharp,
short bursts of high inflation and was concentrated in a few countries,
notably Brazil and the Southern Cone, Bolivia (after her revolution in the
1950s) and Paraguay in the early 1950s. On the contrary, remarkable price
stability was the rule in Mexico, Venezuela, Ecuador, the Central American
countries and the Antilles, where a stable exchange rate for long periods of
timewasviableasaresult.OthersColombiaandPeru,forexamplewerein
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
an intermediate position. However, inflation trends and cycles and, thus,
fluctuations in the real exchange rates also reflected other characteristics of
the macroeconomic functioning of Latin American economies at the time,
most importantly those associated with external shocks and financing, and
savingsinvestmentimbalances,particularlyonthefiscalside.
A notable feature of the quarter of the century after the Second World War
was the absence of significant external financing. As Figure 1.5 indicates, net
resourcetransferswereslightlynegativethroughoutthe1950sand1960s.The
lack of adequate means to finance deficits, together with the temptation to
delaydevaluationsforfearoftheirinflationaryeffects,generatedagreattemp-
tationtouseprotectionandmultipleexchangeratesasmechanismsofadjust-
ment.Inthosecountrieswhichhadbetteraccesstoexternalfinancing,Mexico
inparticular,thisfactorwasnotpresent,butforeigndebtissuesbecamecrucial
early on. The timing of balance of payments crises was surprisingly similar,
reflectingcommonexternalshocks.Therewasafirstwaveearlyinthepostwar
period associated, as we have seen, with the repressed import demand during
the war and the consequent rapid run on international reserves. There was a
second wave in the mid-1950s, a result of the end of the postwar commodity
boom.Athird,lesscommonandlessexternallydeterminedsetofcrisestook
place in the mid-1960s, particularly in countries with multiple exchange rate
regimes.
5
4
3
2
1
0
1
2
3
4
5
%

o
f

G
D
P

1
9
5
0
1
9
5
2
1
9
5
4
1
9
5
6
1
9
5
8
1
9
6
0
1
9
6
2
1
9
6
4
1
9
6
6
1
9
6
8
1
9
7
0
1
9
7
2
1
9
7
4
1
9
7
6
1
9
7
8
1
9
8
0
1
9
8
2
1
9
8
4

Figure1.5 Nettransferofresourcesas%of1980realGDP
Source: ECLAC.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
28 Introduction
Thisperiodofexternally-determinedfinancialausteritywasfollowedbythe
extremely sharp debt cycle of the 1970s and 1980s. The `macro mess' which
characterizedboththeboomandbustisawell-knownstory.Wassuchmacro-
economicmanagementabyproductoftherigiditieswhichstate-ledindustria-
lizationhadaccumulatedforexample,thelackofanadequateexportbaseto
adjust, together with the exhaustion of import substitution opportunities or
thedifficultiesfacedinreducingthesubsidieswhichthatstyleofdevelopment
required? Although answering this question positively could make some
25
sense, we prefer a different line of reasoning which gives macroeconomic
management inthefaceofsharpexternal financialfluctuations adynamicof
its own, rather independent of the underlying `style of development'.
26
We
haveseveralreasonsfordoingso.TheexporteconomiesofLatinAmericahad
similardifficultiesinmanagingasharpcycleinthe1920sand1930sandeven
before,inthe1890s.Also,inthe1970sandthe1980s,Chile,whichhadalready
openedupitseconomy,wasequallyunabletomanageasimilarcyclewhereas
more closed economies, notably Colombia, were in part able to avoid it.
Finally, in the 1990s open economies have been equally unable to manage
sharpexternalfinancialfluctuations.Thepropensitytoboomandbustinthe
faceofunstableexternalfinancingis,thus,afairlygeneralphenomenon,and
certainlynotaparticularfeatureofstate-ledindustrialization.
Several episodes from the 1920s and 1990s, as well as the Southern Cone
particularly the Chilean experience in the 1970s, indicate that private sav-
ingsinvestment imbalances during debt booms can be even more harmful
thanfiscaldeficits,duetotheirrepercussionsinthedomesticfinancialsystems
whichcanbeextremelycostlyandmaytakealongtimetoovercome.Indeed,
theabsenceofdramaticdomesticfinancialcrises,suchasthosethattheregion
experiencedinthe1930sandagaininthe1980sand1990s,indicatesthatthe
domestic `financial repression' characteristic of state-led industrialization had
at least that advantage! It certainly also had others, particularly the ability to
supply domestic long-term financing. It indicates that in Latin America,
domestic financial crises have been basically associated with the aftermath of
externaldebtbooms.
Thetendencytooverburdenthestatewithfiscalresponsibilitieswithoutan
adequate counterpart in resource mobilization was another central feature of
the Latin American macroeconomic scenario during our period of analysis,
which erupted in a fairly generalized fiscal crisis in the 1970s. Fitzgerald
arguesinhiscontributiontothisvolume(see,also,Fitzgerald,1978)thatthis
wasnotintrinsictoimportsubstitutionorstate-ledindustrialization,butshould
rather be located in a particular form of state intervention which had three
distinguishing features: (i) a rise in government expenditures as a proportion
ofGDP,butwithalowerproportionofwelfareexpenditurethaninindustrial-
ized countries; (ii)ashiftin tax composition awayfromproperty andincome
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
taxesintoindirectandwagetaxation;and,asaconsequence,(iii)risingborrow-
ingrequirementstofinancetransferstotheprivatesectorratherthantoredis-
tributeincome,whichcanhardlybecalledaformof`populism'.Accordingto
hisanalysis,thiswasintrinsictothecollapseofthemodel.Tothisissueweturn
next.
Routestothedemiseofthemodel
A lengthy part of the discussion in the workshops concerned the extent to
which, case by case, the costs that we have discussed in the previous section
actuallyledtothebreakdownofstate-ledindustrialization.Thecomplexityof
thediscussion arisesfromthe superimpositionof thedebtcrisis, with itsown
externalandinternaldynamics.Obviously,asidefromthedebtboomandbust,
somecrucialexternalissueswerealsoimportantinthetransitiontofreemarket
policiestheradicalshiftineconomicideologies,theinfluenceoftheseshifts
oninternationalfinancialinstitutions,whichinteractedwiththeirgreatinflu-
ence in the turbulent 1980s, the crisis of the welfare state in industrialized
countries, the fall of communism, and so on but we will concentrate on
internal dynamics. Inthis regard,the debateand the paperssuggested afour-
foldschemeforthedemiseofthemodel.
First, it could be that the limitations internal contradictions of inward-
looking industrialization were so severe that even without the debt crisis the
model would have been unsustainable. The high costs of deepening import
substitution in the 1970s in Brazil, Mexico and Venezuela, as well as the very
high costs of a late adoption of the model in Peru, are perhaps the best
examplesinthisregard.Nonetheless,itisunlikelythatwithoutthedebtcrisis
anyoftheseeconomieswouldhavecollapsedbythesheerweightofinefficient
investment decisions; actually, it is unclear that some of these investment
decisions would have been made at all in the absence of the excess interna-
tionalliquiditywhichcharacterizedthe1970s.Inothercountries,itshouldbe
recalled,themodelhadbeeneffectivelyabandonedintheearlyormid-1970s,
although its instruments survived for a longer period of time as passive ele-
ments of short-term macroeconomic management without a clear long-term
strategy. In others, particularly the small countries of Central America, the
model wasneverfully implemented anditwassuperimposed onatraditional
primaryexportmodel.
A second alternative cannot be altogether disregarded: that through an ad-
equate, gradual adjustment of policies, the model would have evolved into a
morebalancedstrategy,hadnottheoilshocksandthedebtboomandcrisisof
the 1970s come in between, with the accompanying `macro mess'. The facts
thatthemodelwasevolvingtowardssome`mixedstrategy'inseveralcountries,
andthatitunderwentitsown`goldenage'inthelate1960sandearly1970s,are
evidencethatsucharoutecouldhavebeenfeasible.Indeed,somethingcloser
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
30 Introduction
to the equally protectionist and state-led, but outward oriented model of the
Asiantigerscouldhaveevolvedinthefaceoftherevealedsuccessinbuildingan
exportbasewhichseveralcountrieswerestartingtoshow.Anotherevidenceis
the story of Colombia, which did not fall into major macro difficulties in the
1970sandfollowedthisintermediaterouteandinfactrebuiltherownfairly
successful`mixedstrategy'inthesecondhalfofthe1980s.Butthisalternativeis
purelycounterfactualandthusincapableofproofordisproof.
Thethirdpossibility,whichplayedacentralrole,isthatthesharpfinancial
boomandbust,andthemacroeconomicinstabilitywhichaccompaniedthem,
strainedalreadyvulnerablestructuressomuchthatthedebtcrisisbecamethe
deathknellofthemodel.Thesharpturnaroundofexternalfinancinggenerated
internal distribution issues that the system was unable to manage. Weakened
state structures were overburdened and finally unable to respond. The fiscal
imbalanceswhichhadbuiltupbecameunmanageable,anadditionalsourceof
uncertainty which discouraged investment and encouraged capital flight, as
Fitzgerald argues. Investment decisions which would have had a poor return
had the economies grown became a major source of loss with economic col-
lapsealthough,aswehavesaid,somepartlypaidoffinthemediumterm.
The persistence of the debt crisis was obviously crucial for this result. D az-
Alejandro(1988p.310)expressedthisviewbrilliantly:
whatcouldhavebeenaseriousbutmanageablerecessionhasturnedintoa
major development crisis unprecedented since the early 1930s mainly
becauseofthebreakdownofinternationalfinancialmarketsandanabrupt
change in conditions and rules for international lending. The nonlinear
interactions between this unusual and persistent external shock and risky
orfaultydomesticpoliciesledtoacrisisofseveredepthandlength,onethat
neithershocksnorbadpolicyalonecouldhavegenerated.
Indeed,itiscertainlypeculiarthataningrainedfeatureofinternationalfinan-
cing,itssevereinstability,servedasthedeathknellofboththeexportageand
state-ledindustrialization!
The final alternative has to do with politics and political economy. They
certainly played a role in the story of state-led industrialization in several
countries, as well asin the story of itsdemise. Clear examples come from our
case stories. Throughout the period of state-led industrialization, private
investment in Chile was weak, and this can be traced to the inability of
successivepoliticalleaderstoovercomebusiness-class distrustoftheprogress-
ive left wing. Despite healthy reforms, the country was caught between
strongconservativeforcesandradicalizingtendencies,whichendedupinthe
UnidadPopularfailure.Argentinabringsforthparallelreflections;socialconflict
built up since the end of the export era, as Cortes Conde argues in his
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
contribution to the first volume in this series. Against this conflictive back-
ground, the divisions introduced by the first Peronist government seemed
unresolvable and affected political and economic events for decades. Indeed,
they played a central role in the turbulent 1970s, which led to the equally
turbulent1980s.
Military conceptions of national security lay behind many of the import
substitutioninitiativesinArgentina,fromPero nonwards,asKatzandKosacoff
argue in their contribution on Argentine industrial development in this
volume. It also played a central role in Brazilian industrialization, including
the very intensive drive that preceded the debt crisis. It was also behind the
belated Peruvian experiment with classical state-led industrialization in the
1970s.InthecaseofMexico,protectionismwasattheheartoftherelationship
of the PRI (Partido Revolucionario Institucional) with the private sector: a
carrot but also an instrument of political control. In Venezuela, the problems
of`sowingtheoil'generatedastrongsenseofrentabundanceaswellasstrong
forms of `rent seeking' which lay behind that country's own investment mis-
allocation. On the contrary, the weight of agricultural interests was certainly
behindthemorebalancedmodelthatColombiafollowed.Thiswasalsotrueof
CentralAmericawhere,nonetheless,sizealsomattered.
Thisparticularpoliticaleconomywasobviouslynobetterinotherregards.In
Central America, the conflicts that exploded into civil wars in the 1970s and
1980swererootedintheprimaryexportmodeland,ofcourse,playedacentral
roleinthedemiseofitspeculiarmarriagewithweakstate-ledindustrialization.
In Colombia, traditional rural conflicts have been one of the factors behind
persistentviolencefordecades.InBrazil,thefearofagrarianreformwasbehind
the support of the elites for the 1964 military coup which strengthened in its
ownwaythestate-ledindustrializationagenda,aswehaveseen.
The`Washingtonconsensus'viewofimportsubstitutionindustrializationas
amassivedetour,a`policymistake'whichcouldsomehowhavebeenavoided,
is grossly ahistorical and thus inappropriate, as we have shown. Its implicit
viewofthecollapseofstate-ledindustrializationasamanifestationofitsown
inadequaciesisasinadequateasthetraditionalMarxistprescriptionthatcapit-
alism would collapse because of its own internal contradictions. Beyond the
critique of this simplistic view, we have sketched a stronger argument: that
pure economic explanations are an inadequate route to understand both the
excessesofstate-ledindustrializationandalsoitsdemise.
Notes
1 We want to acknowledge warmly the effort and generosity of all participants, who
notonlytookpartwithvigourbuthaveallowedtheampleuseoftheirideasinthe
writing up of this chapter. Some of them also provided lengthy remarks and doc-
umentation,aswellascommentsonthetextofthisintroduction.Weareparticularly
gratefultoRicardoFfrench-DavisandJorgeKatzinthisregard.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
32 Introduction
2 Aparticipantinoneoftheworkshops,DavidPollock,Prebisch'spersonalassistantfor
severalyears,considersthatPrebischpreferredthelabelàcceleratedinternalindus-
trialization'anduseditfrequentlyindiscussions.
3 It is important to emphasize that the return to free markets was expected to take
some time until the mid-1950s certainly. Also, it was not anticipated that capital
accountswouldbeliberalized.
4 The`HavanaCharter'whichsetuptheInternationalTradeOrganisation,(ITO)was
never ratified by many of its signatories, including the USA. It was an unrealistic
proposal attempting immediate free trade with no serious institutional backing to
transactsuchachange(seeDam,1970).TheGATTwasasaresultrequiredtoplaythe
role envisaged for the ITO, but it did not incorporate the organizational and sub-
stantivecontentthattheITOwasintendedtocontain.
5 DepartmentofStateBulletin,11April1948.Thiswasassociatedwiththedifficulties
thattheUSgovernmenthadingettingtheMarshallPlanapprovedbyCongress.See
onthisandontheLatinAmerican,particularlytheBrazilian,discussionsatthetime,
Furtado(1989),ch.II.
6 ECLA later incorporated the Caribbean to become ECLAC. CEPAL is its Spanish
acronym.
7 ForgeneralpopulationtrendsseeECLACCELADEIDB(1996).Populationestimates
aretakenfromrecentissuesofCELADE,DemographicBulletin.
8 OneviewinthedebatestressedthattherathernegativeviewofECLACtakenbyUS
policy-makers in the first decadeor so ofECLAC's existencewas a mixtureof those
who considered its economics misguided if not wrong, and those who did indeed
worryaboutitspoliticalorientation.AnotherviewstressedtheUSdislikeofasemi-
autonomoussecretariatundertheUN.
9 In his semi-autobiographical book, Furtado (1989) provides a fascinating early his-
toryofECLAC.Anevaluationofitscontributionsinthelightofcontemporaryideas
is provided by Hirschman (1971), ch. 13, Cardoso (1977), Fishlow (1985) and Love
(1994).Rodr guez(1980)providesthebestoverallaccountofitsthinkingduringits
`classical'periodandBielschowsky(1998)themostcompletesurveyoftheevolution
ofECLAC'sideasovertime.
10 For additional, recent evaluations of postwar performance, see Bulmer-Thomas
(1994), Ffrench-Davis, Munoz and Palma (1994) and Katz (1998). The recent work
by Hofman (1998) provides growth-accounting exercises for the major Latin Amer-
icaneconomiesintheregioninthetwentiethcentury.
11 ECLAC (1978b). Unless otherwise indicated, the figures in the rest of this and the
next paragraph come from this source, the only one that has estimates of GDP (at
factorcost)fortheregionasawhole(excludingCuba)intheperiodanalysed,usinga
similarbaseyear(1970)andmethodology.
12 ECLAC(1991),using1980asthebaseyear.
13 According to the imperfect series available, Cuba's per capita GDP fell by 31 per
cent in the 1920s and 1930s and only partially recovered in the 1940s, prior
to a phase of stagnation in the 1950s and 1960s. Thus, by 1970 it was still 19
percentbelowthelevelreachedin1920.SeethestatisticalappendixinThorp(1998).
14 Labour productivity as well as total factor productivity and capital accumulation
estimates below refer to the six major Latin American economies, according to
calculations by Hofman (1998). Comparisons are taken from Maddison (1995),
Ffrench-Davis,MunozandPalma(1994)andHofman(1998).
15 Foramoreextensivediscussionofinstitutionaldevelopmentduringstate-ledindus-
trialization,seeThorp(1998),ch.5.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
16 Foranoverallviewoftrendsinagriculturethroughoutthisperiod,seeECLAC(1978).
17 These results came sometimes with a lag. As we have pointed out, the cases of
forestry,fishingandfreshfruitsinChile,thesuccessexportstoriesinrecentdecades,
wererootedineffortswhichtookoffinthe1960s.
18 Thisappreciation,aswellastheremainderofthisparagraph,isbasedonGarc aand
Tokman(1984).
19 Thorp(1998),basedonestimatesbyShaneHunt;andAstorgaandFitzgerald(1998).
20 IMFseriesindicatethattheshareofLatinAmericainworldtradeincreasedfrom7.9
percentin1938to12.2percentin1948,11.7percentin1950andbackto7.8per
centin1960.UNCTADseriesindicatethatitwas10.9percentin1950,6.7percentin
1960andanaverageof4.3percentinthe1970s.
21 SeeECLAC,StatisticalYearbookofLatinAmericaandtheCaribbean,severalissues.These
data refer to SITC groups 58, which excludes some processed agricultural goods,
beverages, tobacco manufactures and oil derivatives. According to the even more
restrictive UNCTAD definition, which differentiates semi-manufactures and manu-
factures, the latter's share increased from 4.9 per cent in 1962 to 14.3 per cent in
1980,growingatarealrateof12percentoverthesetwodecades.Seeafullanalysisin
ECLAC(1992).
22 SeeFishlow(1984)andFfrench-Davis,MunozandPalma(1994).Theshareofintrar-
egionalin totalexportsincreasedfrom8.8 percentin 1960to 16percent in1975;
this share was broadly similar just before the debt crisis, when intraregional trade
experienced a strong depression. A well-known pattern was the higher share of
manufactures in intraregional exports than in overall trade, particularly for the
smallandmedium-sizedcountries.
23 Mexico in 1948, Peru in 194950, Brazil in 1953, Chile in 1956, Colombia in 1957
andVenezuelain1961.
24 An interesting conclusion of D az-Alejandro's (1976) classic study of Colombia's
foreign trade regime was that the instability of the real exchange rate and export
incentives prior to 1967 had large adverse effects on non-traditional exports. Thus
thecrawlingpeghadasignificanteffectonexportdiversificationbyprovidingstable
incentives,despitethefactthatdirectexportsubsidiesactuallydecreased.
25 This was a typical line of reasoning in the 1980s, which emphasized that open
economies,suchasthoseinAsia,wereabletoadjustfasterinthe1990s.Intheline
ofrecent(1997)Asianevents,thatreasoninglooksinconclusiveatbest.
26 See Bacha and D az-Alejandro (1982) for a now classic long-term view of external
financingissuesinLatinAmerica.
References
Ashworth, W. (1987) A Short History of the International Economy since 1850, 4th edn,
London:Longman.
Astorga,P.andFitzGerald,E.V.K.(1998)`TheStandardofLivinginLatinAmericaDuring
the Twentieth Century', Centro Studi Luca d'Agliano and Queen Elizabeth House
DevelopmentSudiesworkingpaperno.117,May.
Bacha, E. and D az-Alejandro, C. F. (1982) Ìnternational Financial Intermediation: A
Long and Tropical View', reprinted in A. Velasco (ed.), Trade, Development and the
World Economy, Selected Essays of Carlos F. D az-Alejandro, 1988 (ch. 8), Oxford: Basil
Blackwell.
Bethell,L.(ed.)(1994)TheCambridgeHistoryofLatinAmerica,VolVI,PartI,Cambridge:
CambridgeUniversityPress.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
34 Introduction
Bielschowsky,R.(1998)`Cincuentaan os osdepensamientodelaCEPAL',inCincuentaan
depensamientoenlaCEPAL,Santiago:FondodeCulturaEcono micaCEPAL.
Bulmer-Thomas,V.(1994)TheEconomicHistoryofLatinAmericasinceIndependence,Cam-
bridge:CambridgeUniversityPress.
Cardoso,F.H.(1977)`Laoriginalidaddelacopia:laCEPALylaideadeldesarrollo',Revista
delaCEPAL,secondsemester.
Chenery, H., Robinson, S. and Syrquin, M. (1986) Industrialization and Growth: A Com-
parativeStudy,NewYork:OxfordUniversityPress.
Dam, K. W. (1970) The GATT: Law and International Organization, Chicago: Chicago
UniversityPress.
D az-Alejandro, C. F. (1988) `Latin American Debt: I Don't Think We Are in Kansas
Anymore', reprinted in A. Velasco (ed.), Trade, Development and the World Economy,
SelectedEssaysofCarlosF.D az-Alejandro,1988(ch.15),Oxford:BasilBlackwell.
(1976)ForeignTradeRegimesandEconomicDevelopment,NationalBureauofEconomic
Research,NewYork.
ECLAC (1992)El comerciode manufacturas de America Latina:Evolucion yestructura1962
89,EstudioseInformesno.88,Santiago.
(1991) America Latina y el Caribe: Series regionales de cuentas nacionales a precios
constantesde1980,CuadernoEstad sticono.15,Santiago.
(1978a) 25 anos en la agricultura de America Latina: Rasgos principales 195075, Cua-
dernono.21,Santiago.
(1978b)SerieshistoricasdelcrecimientodeAmericaLatina,CuadernoEstad sticono.3,
Santiago.
ECLAC, CELADE and IDB (1996) Impacto de las tendencias demograficas sobre los sectores
socialesenAmericaLatina,Santiago.
Fishlow,A.(1985)Èlestadodelacienciaecono micaenAmericaLatina',inIDB,Progreso
EconomicoySocialenAmericaLatina,ch.5,Washington,DC.
(1984)`ReciprocalTradeGrowth:TheLatinAmericanExperience',inM.Syrquin,L.
Taylor and L. Westphal (eds), Economics, Structure and Performance: Essays in Honor of
HollisChenery,Orlando:AcademicPress.
Ffrench-Davis, R., Munoz, O. and Palma, G. (1994) `The Latin American Economies,
195090', in L. Bethell (ed.), The Cambridge History of Latin America, Vol. VI, Part I,
Cambridge:CambridgeUniversityPress.
Fitzgerald, E. V. K. (1978) `The Fiscal Crisis of the Latin American State', in J. F. J. Toye
(ed.),TaxationandEconomicDevelopment,London:Cass.
Furtado,C.(1989)Lafantas aorganizada,Bogota:TercerMundoEditores.
Garc a, N. and Tokman, V. (1984) `Transformacio n ocupacional y crisis', Revista de la
CEPAL,no.24,December.
Hirschman,A.O.(1971)ABiasforHope,NewHaven,Conn.:YaleUniversityPress.
Hofman,A.(1998)`LatinAmericanEconomicDevelopment:ACausalAnalysisinHistor-
icalPerspective',GroningenGrowthandDevelopmentCentre,monographseries,no.3.
Jorgensen,S.L.andPaldam,M.(1987)`TheRealExchangeRatesofEightLatinAmerican
Countries 19461985: An Interpretation', in Geld und Warung (Monetary Affairs), vol
3(4),December.
Katz,J.(1998)`TheDynamicsofTechnologicalLearningduringtheISIPeriodandRecent
Structural Changes in the Industrial Sector of Argentina, Brazil and Mexico', mimeo,
ECLAC.
(1987) `DomesticTechnologyGenerationin LDCs: A Reviewof Research Findings',
inJ.Katz(ed.),TechnologyGenerationinLatinAmericanManufacturingIndustries,London:
Macmillan.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Little,I.,Scitovsky,T.andScott,M.(1970)IndustryandTradeinsomeDevelopingCountries:
AComparativeStudy,NewYork:OxfordUniversityPress.
Love, J. L. (1994) Èconomic Ideas and Ideologies in Latin America since 1930', in A.
Maddison(1995),MonitoringtheWorldEconomy18201992,Paris:OECDDevelopment
Centre.
Maxfield,S.andNolt,J.(1990)`ProtectionismandtheInternationalizationofCapital:US
Sponsorship of ISI in the Philippines, Turkey and Argentina', International Studies
Quarterly,vol.34,pp.4981.
Prebisch, R. (1964) Nueva politica comercial para el desarrollo, Mexico: Fondo de Cultura
Econo mica.
Rabe, S. (1978) `The Elusive Conference: U.S. Economic Relations with Latin America,
194552',DiplomaticHistory,vol.2(3),pp.27994.
Rodr guez,O.(1980)Lateor adelsubdesarrollodelaCEPAL,Mexico:SigloXXI.
Teitel,S.(1993)IndustrialandTechnologicalDevelopment,Washington,DC:Inter-American
DevelopmentBank.
Teitel,S.andThoumi,F.(1986)`FromImportSubstitutiontoExports:TheManufacturing
Experience of Argentina and Brazil', Economic Development and Structural Change, vol.
34.
Thorp,R.(1998)Progress,PovertyandExclusion:AnEconomicHistoryofLatinAmericainthe
XXth Century,WashingtonDC:Inter-AmericanDevelopmentBankandJohnsHopkins
Press.(AlsoinSpanishandPortuguese.)
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2
TechnologicalLearning,Institution
BuildingandtheMicroeconomicsof
ImportSubstitution
*
Structuralism,theneo-classicalschoolandthestudyofeconomic
growth
The history of economic thought is marked by a deep divide between
the structuralist current originating in the German historicist school and the
positivistapproacharisingfromtheworkofthinkerssuchasHumeandSmith.
Whileinitiallybothrespondedtothedistinctproblemsfacedbybothcountries
in the world order of nations around the beginning of the Industrial Revolu-
tion,thedifferenceshavepersisted,andstillformthebasisoffierceconfronta-
tiononideologicalandintellectualmattersattheheartoftheprofessiontoday.
The first approach advocates an interventionist policy, which derives from
Germany's need to close the technological gap between it and
British best practice. To achieve this, the approach establishes the state as the
central agent of social organization, coordinating and directing individual
economic relations. The second approach, on the other hand, favours unhin-
deredfreetradeandseeslaissez-faireasthemostsuitablewayofachievingthe
optimumallocationofavailableresources.
Over the years there have been many contributions on both sides of this
argument. In the process of transformation of these ideas, those proposed by
ECLAC in the 1950s should be seen as a significant moment in structuralist
thinking,asare,forexample,thoseproducedbytheFrenchregulationistthin-
* The ideas expressed here are the sole responsibility of the authors. They are both grateful for
comments from those who participated in the seminar on Latin America's Economic History in the
NineteethCentury,heldinPaipa,ColombiainMay1997,especiallythosecommentsfromRosemary
Thorp, Enrique Cardenas, Jose Antonio Ocampo and Arturo O'Connell which made it possible to
improvetheoriginalversionofthetext. ThechapterhasbeenpublishedinDesarrolloEconomico, vol.
37,1998.Permissiontoreprintisacknowledged.
36
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
JorgeKatzandBernardoKosacoff 37
kers,
1
or by the evolutionists supported in the writing of Joseph Schumpeter,
especiallythosefoundinCapitalism,SocialismandDemocracy,writtenin1942.
2
Various authors have attributed the little cross-fertilization visible between
the two streams of economic thought to the lack of mathematical rigour
in structuralist ideas.
3
We consider, however, that the problem goes much
deeper, and is related to epistemological issues which are harder to
resolve and which ultimately make up the sense and substance of the social
sciences.
Toastructuralistthinkerthedevelopmentofone'sowntechnological com-
petenceandcapabilitiesandthecreationofnewinstitutions
4
arekeyplayersin
anygivensociety'sprocessoflong-termeconomicgrowth.Thistypeofprofes-
sionalconsiderscompaniesandindividualsasimperfectlyinformedregarding
theoptionstheyface,andashavinganincompleteperceptionofthecostsand
benefits associated with each of their potential decisions. They also consider
them to be acting in bounded rationality, that is, that they are looking for
benefitsbutwithoutnecessarilymaximizinggain.Economicplayersoperateby
trial and error, looking for and testing new operating processes which might
allowthemtoimprovetheirperformanceovertime.Theyaccumulateexperi-
ence as they undertake new activities or expand existing ones. This all makes
up a maturing process in which time and history play a fundamental role,
conditioning the complexity of the activities the economic agents can carry
out efficiently. There are increasing returns to scale, externalities, inappropri-
abilities and various other anomalies that prevent the price system from
achievingthesociallyoptimalallocationofresources.
Inthiscontexttheexpansionofagivensociety'sindustrialproductionbase
constitutesoneofthecentralfactorsdeterminingthedegreeoforganizational
andtechnicalcomplexityachievedbythissociety,giventhatindustrialdevel-
opment requires more sophisticated forms of division of labour. Expansion
encourages the creation and diffusion of norms and habits of behaviour
which together give form and content to a huge productive institutional
culture thatpermeatesthewhole breadth ofthecommunity.Such aculture
which brings in technological knowledge as well as the capacity for manage-
mentandpatternsofbehaviouroflabourmakesupaveryimportant`social
capital',whichconditions,andinturnisconditionedby,theevolutionarypath
whichsocietyfollows.Itisinthestructureofthissocialcapitalthattheabove-
mentionedanomaliesplayafundamentalrole.
Thus, technological learning and the development of new institutions in
themanysensesmentionedabovemakeupcentralaxesforanystructuralist
explanation of a given community's long-term process of economic develop-
ment.Thestateappearsinthiswayofthinkingasaleadingplayerguidingthe
allocation of resources with a vision of a social optimum which transcends
anythingthefreemarkethastooffer.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
38 TheMicroeconomicsofImportSubstitution
In contrast, neo-classical economics offers a very different description of
economicdevelopment,andofwhyasocietygrowsanddevelopsoveraperiod
of time. This description arises from the `representative firm', an extremely
stylizedcharacterizationofwhatconstitutesacompanyandthefeatureswhich
determine its behaviour. A neo-classical company enjoys perfect information
regardingitsfuturepotential,hasfullknowledgeoftheprobabilitiesofsuccess
or failure of each and every action it might pursue, works in perfect factor
marketsandmaximizesprofitsfromacollectionofexogenousdatawhichfully
describebothproductionfunctionstowhichithasfreeaccessandpricesof
factorsofproduction.Thesedataaretakenasgiveninitsmaximizationcalcu-
lations.
There is no such thing as bounded rationality in this world, or increasing
returnstoscale,externalitiesorpublicgoods.Thelogicofcompetitiveequilib-
riumisnotcompatiblewithsuchanomaliesinthesmoothworkingoftheprice
system.Theproductionfunctionisgeneric,notcompany-specific.Thereisno
tacit unformalized knowledge. When learning is allowed to take place, it
happens in a deterministic manner, apparently without the uncertainty and
trial and error which usually surround attempts at attaining new scientific-
technical knowledge. The neo-classical world allows for no social institution
butthemarket,theonlyplaceforthesocialinteractionofindividualeconomic
agents; it takes place without the interference of direct interdependencies
between them. The price structure offers them not only incentives but also
informationnecessaryandsufficientinformationonthebasisofwhichthey
canmakeday-to-dayeconomicdecisions.
5
The above-mentioned micro-foundations are necessary if we are to have
aggregate savings, investment, consumption or production functions, based
on which the global running of the economy can be discussed. At the
1988 conference in Stockholm to mark his being awarded the Nobel Prize for
Economics, Robert Solow described the neo-classical analytical setting as fol-
lows:
Theideaistoimaginethattheeconomyispopulatedbyasingleimmortal
consumer, or a number of identical immortal consumers...she, or the
dynasty,issupposedtosolveaninfinite-timeutility-maximisationproblem.
...For this consumer every firm is just a transparent instrumentality, an
intermediary, a device for carrying out intertemporal optimization subject
onlytotechnologicalconstraintsandinitialendowments.Thusanykindof
marketfailureisruledoutfromthebeginning,byassumption.Thereareno
strategic complementarities, no coordination failures, no Prisoner's Dilem-
mas. The end result is a construction in which the whole economy is
assumed to be solving a Ramesy optical-growth problem through time,
disturbed only a stationary stochastic shocks to tastes and technology. To
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
thesetheeconomyadaptsoptimally.Inseparablefromthishabitofthought
istheautomaticpresumptionthatobservedpathsareequilibriumpaths.
(Solow,1988,p.310)
The above paragraphs clearly demonstrate the different microeconomic insti-
tutionalandhistoricalbasesonwhichneo-classicalandstructuralisteconom-
ics are founded, as they bear on a society's process of long-term economic
development. The factors determining economic growth and the very nature
ofthephenomenondescribingeachofthesesystemsofthoughtdiffersignific-
antly.It,therefore,isjustifiabletoquestionwhethertheneo-classicalmethodis
really useful for understanding what has taken place in peripheral countries
duringtheimportsubstitutingstage,thatis,inthefourdecadesbetween1940
and1980.
The question seems relevant tous in as muchas various writers inthe neo-
classicaltradition,usingtheneo-classicalmodelofgrowthasacounterfactual
setting, have over the last decade presented an evaluation which has been
extremely critical of what has taken place in the periphery especially in
LatinAmericaduringthepostwarperiod.
6
Tocaricaturetheargumentalittle,
wecouldevensaythatforthesewritersallthatimport-substitutingindustrial-
ization(ISI)couldoffertoperipheralcountrieshasbeencorruptpublicofficials
andrent-seekingbusinessmen,onlyabletosetupinefficientproductionplants,
incapable of facing the threat of international competition. Although they
accept that the process hasbeen more successful in Southeast Asia, their con-
clusionsonstateinterventionandindustrialpolicyarenegative,andtheyprefer
insteadtoemphasizethevirtuesofthemarketasanallocatorofresources.
7
Itis,however,appropriatetoasktowhatextenttheneo-classicaldescription
of the events of the import substitution period actually derives from the very
particular lenses through which these writers look at the world. These lenses
simply prevent them from seeing the enormous complexity to be found
beneath the processes of generation, adaptation, diffusion and the use of
technological knowledge, and from understanding for example that there
is significant interdependence between the use and creation of technology,
since no comprehensive or documented blueprint or engineering manual
exists.
8
Rather the use of a certain package of information generally requires
adaptationtothecontext,andthisinturndemandsadhocgenerationofnew
localized technical knowledge. By using an overly simplistic model which
suggests the existence of generic production functions freely available in a
stock or `shelf' of technologies to which the whole world hasaccess, the neo-
classicalmodel simplyisolatesitself fromthepossibilityofunderstandingthe
historicalandculturalcomplexityoftechnologicallearning,andtheprofound
influence of institutions on the path of a community's learning. Something
similar happens if one refuses to admit that the behaviour of the individual
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
economic agent is conditioned not only by relative prices but also by an
extensive list of regulatory norms, behavioural patterns, perceptions and so
on, which vary from community to community and often even between
individuals.Themodelfailstorecognizetheenormousimportanceofinstitu-
tions other than the market in giving shape to the evolution of a society and
therolethehistoricaldimensionplaysthroughout.
Thus, we should not be surprised that between structuralists and neo-class-
icists there are significant differences of interpretation of what took place
duringtheperiodofISIand,moregenerally,concerningatheoryofeconomic
growthandthatthelatterhaveattachedlittlesignificancetodynamicfactors
relatedtothedevelopmentofasociety'sinternaltechnologicalcapabilities,nor
totheroleofinstitutions.Toastructuralisteconomist,suchthingsarecrucial
toanunderstandingoftheevolutionaryprocesswhichagivensocietyisgoing
through. Instead, mainstream professionals have chosen to attach greater sig-
nificancetoaspectsrelatedwiththestaticallocationofresources,whichleads
to analysis which prioritizes a competitive equilibrium model, and margin-
alizesthosedynamicaspectswhichhavetodowiththematuringofproductive
forces.Thisistosaythey situatethemselvesclosertoParetothantoSchump-
eter,andalthoughtheymaybeabletoilluminateoneimportantaspectofthe
reality in a way which is certainly valid they entirely fail to capture the
significanceofanother,atleastassignificantasaninterpretationofwhattook
placeinthedevelopingworldoverthelongrun.Atheorywhichtrulyhelpsus
to understand the evolution of the peripheral world and its competitive inte-
grationintotheworldshouldcertainlyincorporatebothperceptionsofreality
and,furthermore,considerthewaysinwhichtheyareinterdependent.
Inthisworkweintendtoexaminetwoimportantissueswhichneo-classical
orthodoxy has more or less ignored. First we will look at the origins of the
endogenous technological development of peripheral countries, and then
examine the role played by institutions in conditioning and being condi-
tionedbyasociety'sevolutionoverthefourdecadescoveredbytheimport-
substituting policy, that is until the beginning of the attempts at external
openness and the deregulation of the region's economies which began at
variouspoints,tovariousextentsandwithvaryingdegreesofsuccessinthe
1970sand1980s.
Looking at these matters it becomes obvious that ISI did not have the
thoroughly negative consequences that conventional neo-classical interpreta-
tion would suggest, and that it is necessary to adopt a more balanced stance
thanthatofferedtousbytheprofessionalmainstream,andonelessderogatory
ofthesubstitutionprocess.Paripassuwiththeunendingindustrialexpansion
ofcompaniesintheregion,completebranchesofactivitydevelopedtheirown
technological base and stock of technical and business knowledge, as well as
training of the labour force, patterns of labour behaviour, ways of organizing
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
production, mechanisms for social interaction, systems of mutual trust and
interdependency between productive agents, which allowed them to signific-
antly improve the relative productivity of factors, closing the divide which
traditionallydifferentiatedthemfromtherestoftheworld.
Based on thismaturing process, whole firms and branches of industry have
gainedgroundintheinternationalcompetitivescene,reachingothermarkets
and business scenes in the developed world, exploiting technological and
business potential, economies of scale and various types of synergy originally
developedfortheinternalmarket.Largenumbersofnewinstitutionsinthe
multiple sense in which we use the term here began to be diffused through
the production system and social structure, giving rise to the explosion of a
productioncultureofgreatimportancetotheframeworkofalong-termevolu-
tionaryvision.
Itcanbearguedthatthismaturationprocesshasnot,intheLatinAmerican
case, been as profound and evenly distributed over the productive and social
spectraasitseemstohavebeeninSoutheastAsiancountriessuchasKoreaor
Taiwan. The question of why this happens is oneof indisputable importance,
andoneforwhichwestillhavenosatisfactoryanswer.Thedifferentialroleof
educational development, the higher rates of savings and investment, the
impact of the increased pressure from competition which seems to have pre-
vailedintheAsiancountries'domesticmarkets,aswellasanenormousrange
ofanthropological,geopoliticalrelatingtotheconflictbetweenEastandWest
and even religious factors, should all be borne in mind if one is to attempt
some explanation of the differences observed in long-term performance
between groups of countries. However this does not mean that we should
deny the importance of what took place in the case of Latin America, or
relegatetoasecondaryleveltheroleplayedbythedynamiceffectsofstructural
change derived from import substituting industrialization. From the point of
viewofourhistoricalanalyticalreadingofwhattookplace,aswhenassessing
thecost/benefitsofthesubstitutionstrategy,itwouldseemreasonabletoadopt
amorebalancedviewofrealitythanthatofferedbyneo-classicalorthodoxy.
Inthenextsectionwe willexaminetheissue oftechnological learningand
thecumulativedevelopmentoflocaltechnologicalcapabilities.Wheredothe
signalsandincentiveswhichinitiatethematuringprocessatamicroeconomic
level come from? What evolutionary consequences do these have? Following
thiswewilldothesamewiththeareaofinstitutions,attemptingtorecoverthe
historical-cultural,andthereforecountry-specificdimensionofeachnational,
andevenregional,case.Thefinalaimofthisworkistolocatethereaderinthe
technologicalandinstitutionalcontextinwhichISItookplace,soastoprevent
anahistoricalandnon-institutionalreadingofthefacts.Onlyinthiswaycan
wedevelopamoresuitabletheoreticalwayofdealingwiththem.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Thedevelopmentoflocaltechnologicalcapabilities
ThecountriesofLatinAmericaemergedfromtheSecondWorldWarextremely
isolatedfromtheirglobalcontext.Atthetime,manyofthemwerecontrolled
by strongly nationalistic military governments for whom the climate of the
ColdWarandtheappealofplanningstrictdisciplineinmilitarytrainingfrom
the time of Klausewitz translated into the granting of high priority status
to the defencesector andto those industriesclassified as heavy. This explains
why the industrial policy agenda was so biased towards steelworks, coal or
petrochemicalsincountriessuchasArgentina,Brazil,ChileandMexico.
9
The bias also worked in favour of the state monopoly of many of these
productionactivitiesoratleastofthehighdegreeofpublicsectorparticipa-
tioninthemanufacturingplantsbeinginstalledallovertheregion.Thispublic
sectorbiaswasatthetimealsoquitesignificantincountriessuchasFranceand
Englandwherestateindustrywasbeingproposedasasolutiontoproblemsof
economies of scale, ùpstream' or `downstream' interdependencies of the pro-
ductionstructure.
InLatin Americathese were years of shortages in the markets forconsumer
durables, energy inputs and capital goods. Queues, long waiting periods,
rationing, lack of energy and so on, contributed to the establishment of a
sellers'marketinwhichsellerswereabletoturnshortagestotheiradvantage.
At the same time these were the years in which the various economies of the
regionbegantogrowrelativelyquicklyfollowingthedifficultiesofwartime.In
somecasesthistookplacewithoutsidehelpprovidedbydevelopedcountries
inthesettingoftheColdWar.InthissensethecasesofKoreaandTaiwanare
typical,withthethreatofMaoZedong'sChinatriggeringtheseotherexternal
influences.Inothercasesitwasdirectforeigninvestmentbeginningearlierin
MexicothanintheotherLatinAmericancountriesthatmobilizedindustrial
production. In a third group, the increased dynamism of manufacturing
resulted from the existence of large international reserves. Argentina is the
prototypeofsuchacase.
Thisexpansiveatmosphere,thenaturalprotectionresultingfromtheperiod
of war, the tariff protection (or, in many cases, the simple prohibition of
imports) which began to be implemented over these years as an incentive
to setting updomestic productivecapabilities andtheavailability ofsubsid-
ized credit all induced many firms in Argentina, Brazil, Mexico, Chile and
Colombia as well as in Spain, Portugal, Korea and Taiwan to initiate local
production of consumer durables and low-complexity capital equipment,
based on engineering designs one or even two decades behind the interna-
tionalstate-of-the-art.
Substitution policies were often considered to be transitory. They were a
short-term response to the changes taking place on an international scale
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
following the break-up of multilateralism and convertibility which had been
typical of the years of the gold-standard regime. The experience of the First
World War, in which natural protection also acted as an incentive to import
substitution, but which produced a clear step back in terms of domestic pro-
ductionofmanufacturesoncetheworldreturnedtonormalifthatwasinfact
whathappenedseemedtosuggestanephemeralstateofaffairs.
However,onthisoccasionmattersweredifferent.Internationallysignificant
factorsweretheextensionofwartime,thegradualdeepeningintheclimateof
East-West confrontation during the Cold War, the change taking place in the
centralcountriesconcerningtheroleofthestateasamotorofgrowthandas
the ultimate guarantee of communal well-being stemming from the ideas of
Keynes,Beveridgeandotherimportantthinkersoftheperiod.InLatinAmerica
we note the presence of highly nationalistic military governments in many
countries. We note also the gradual consolidation of a convenience pact
between the emerging protected industrial enterprise sector and the new
urbanproletariat which was beginning to grow aroundthe large metropoli of
theregion.Thispactallowedbothsetsofactorstotakeasignificantportionof
the surplus in the primary sector. All these factors meant that things were
differentthistime,andthatthesubstitutionprocesswouldextendanddeepen
during the 1950s. Along with the large public companies dealing with heavy
industry,transportandtelecommunications,andthedefencesectors,asignific
ant production and enterprise base also began to grow, made up of small and
medium-sizedfamilybusinesses,manyofwhichwereestablishedbyEuropean
immigrantsSpanish,Italian,Germanandsoonwhohadsome(incomplete)
knowledge of prewar technology in the areas of electro-mechanics, chemicals
andsoon. Basedon this,onusedandrepairedcapital equipment, onlocally-
made machinery and product designs often copied from similar European or
USmachinesofthepreviousoneortwodecades,manyofthesefirmsbeganto
produce relatively simple capital equipment locally electric motors, water-
pumps, farm machinery and so on, as well as chemical, pharmaceutical and
rubber products, which had previously been imported. It is obvious that they
startedwiththeleastsophisticatedofthepreviouslyimportedgoods, andthe
industrywasbasedonadomesticdemandwhichdidnotinsistonquality,nor
wasitdemandingasregardsdeliverytimesorevenprices.Therewerequeues,
sellers' markets, and little of the internal competition required to make the
generalrulesofthemarketworkasinconventionaltextbookmodels.
Two major actors in the industrial setting of the ISI world emerged and
strengthenedduringthisperiod:ontheonehandthelargepubliccompanies,
and on the other the small and medium-sized enterprises owned by national
capital,whichachievedhighratesofgrowthattheendofthe1940sandinthe
1950sunderthenewincentivesregimeprevalentinsociety.Atthesametimea
sectorconsistingoflargefirmswithlocalcapitalalsogrew,primarilyrelatedto
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the exploitation of natural resources, which much later was to become a very
important player following the debt crisis around the end of the 1970s and
thebeginningofthe1980s.Thisgroupoffirms,whichwouldgivewaytothe
constitution of large national-capital conglomerates which are now very
important in every country in the region, was not a very significant force at
this point, but should be noted owing to their later role in consolidating the
frameworkoftheregionalproductionsystem.
A fourth important player local subsidiaries of transnational companies
burstontothescenearoundtheendofthe1950sinArgentinaandBrazil,and
somewhat later in the 1960s and 1970s in the smaller countries of the
region.
10
It was not that the above-mentioned player was not present before,
but that given the incentive of external protection, of import permits and
various other types of subsidy, those years now saw the setting-up of a huge
numberofnewmanufacturingplantsdedicatedtoproducingforthedomestic
market. Many of those from previous periods were focused on exploiting
natural resources often in an enclave situation, that is, not well-integrated
into the local productive and institutional setting and when they produced
for the domestic market they tended not to benefit from protection from
external competition.
11
The appearance of this new layer of transnational
firms introduced significant changes to the model of industrial organization
currentatthetime,inasmuchasthesefirmsbroughtwiththemnewproduc-
tion designs, new technological processes, new kinds of work organization.
Nonetheless, their entrance into the market often noticeably changed their
structureandbehaviour,suchasintheconsumerdurablemarketsofArgentina,
Brazil or Mexico, where many firms based on local capital were quickly dis-
placedbythearrivalofthetransnationals.
12
TheincorporationofLatinAmer-
icaintotheprocessofinternationalizationofproductionwasinstepwiththe
prevalence of Fordist models of production which began to spread as multi-
plantfirmsbecameestablishedinthoseeconomieswithhightariffprotections
(Dunning,1988).
The institutional framework described above, the scarcity and fragmentary
nature of the technological capabilities then available and the difficulty until
thelate1950sofobtainingcapitalequipmentandmoreup-to-dateproduction
designs,meantthattheindustrialstructuredevelopinginLatinAmericainthe
postwar years was less advanced and integrated than that in the developed
world. The latter, in turn, was going through a process of transformation
brought about by the transfer from a wartime to a peacetime economy. A
new generationof capitalgoods, consumer durablesandso on brokeinto the
worldmarkets.Thisoriginatedintheconversionofcentralcountries'military
expenditure to expenditure on investment and development, and the indus-
trial application of many technologies developed over the two previous
decades, because of armaments and space programmes. This indisputably
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
helped to widen still further the gap separating the peripheral world's new
industrial production from what was being produced at the time in Europe
andtheUSA.
Inspiteofthelargeamountswrittenforandagainstthesubstitutionmodel,
themicroeconomicunderstandingofthismodelhasadvancedrelativelylittle.
So it is worth our pausing here briefly to look at a few aspects of micro
behaviour and structure which were influencing the path of learning and the
developmentoftheperipheralcountries'internaltechnologicalcapabilities.In
thefollowingsectionweseektoexaminethesequestions.
Thefirstfeaturetobetakenintoaccountisthesizeoftheindustrialplants.In
generalthesewerenolargerthan,say,10percentoftheprototypicalsizefora
developed country, for similar goods. The lay-out of the plant and the organ-
izationofmanufacturingworkweremuchlesssophisticated,notonlybecause
ofadaptationtothedifferentrelativepricesoffactorsofproduction,butalso,
and mainly, because of the lack of adequate information, equipment and
organizationalknowledge.Hadtherebeengreatertechnologicalcapitalatthe
heart of society this doubtless would have been used. The capital goods used
wereoftensecondhandorhomemade,thedegreeofverticalintegrationofthe
industrialestablishmentswasfarhigherbecauseoftheimmaturityofthelocal
production structure andthe lackof independentsuppliers ofspare partsand
so on. On this level, and on the level of direct interdependency between
producers which a higher degree of complexity in industrial structure would
have allowed, one can observe significant differences between the developed
countriesandtheworldofISI.Thedegreeof`roundaboutness'wassignificantly
lower in the periphery, as was the extent of the social division of labour and
internalizedeconomiesofscale.Themixofmanufacturedproductstendedto
be far greater than would be found in a specialized plant in a developed
country. This implies small-scale production of many different products and,
ultimately,alackofeconomiesofscale.
Boththestaticefficiencyandthedynamicpathoftechnologicallearningin
suchasettingareboundtoaccumulateidiosyncraticfeaturesandproblemsas
timepasses.Unquestionablythedifferencesbetweenthisproductiveworldand
thatofthedevelopedcountriesarenumerous.Itisnotmerelyaquestionofthe
relative prices of factors of production, but a systemic phenomenon which
concernsthewholesocialorganizationofproduction,theevolutionaryphase
which a given society is passing through. Competition between the two sys-
tems is certainly difficult, though it is obviously impossible a priori to discard
the possibility that in a significant number of individual cases, perhaps even
atthescaleofwholebranchesofactivity,firmsmightbeablegraduallytoclose
the relative gap separating them from the industry of more developed coun-
tries.Thiswaseffectivelywhathappenedinalargenumberofcases,aswewill
seebelow.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Averysmall-scalemanufacturingplant,ahighdegreeofverticalintegration,
an ad hoc and almost backward factory lay-out, imperfect knowledge and
understanding of the principles of organization of labour, and an excessively
broad production mix with small quantities of many different products, cer-
tainlyexplainthehighinitialcostperunit,thelowqualityoftheproducts,and
anexcessiveburdenofoperationaldowntime.Althoughdomesticwageswere
low,thetypeofproductsabletobeproducedandthehighlocalcostsdidnot
favourexportationtomorecompetitive,sophisticatedmarkets.
Itisalsoimportant tobearinmindthatthisset ofinitialconditions gener-
atedmultiplephysicalsignalsinengineeringterms,thatisaswellasrelative
price signals, which induced many local firms to attempt technological
improvementsaimedatproductdesign,manufacturingprocessesandwaysof
organizing labour. Obviously the greater productivity potential arising from
these attempts was significant, given the initial high degree of operational
inefficiency and the need to produce more up-to-date goods with respect to
the international state of the art. Many companies reacted to these signals by
creating ad hoc groups and departments to work on technical assistance, on
process engineering, production planning and organization and so on. Their
fundamentalaimwastogenerateadditionalsourcesoftechnologicalinforma-
tion, based on which product design, production processes, organization of
labour, engineering lay-outs, quality control techniques, and production
norms in factories could all be improved. In numerous cases these groups
designedmachinerywhichwassoonproducedinthemaintenanceworkshops
withinthecompaniesthemselves.
Such in-house productivity improvement appeared through a process of
development of internal technological capabilities which was highly specific
to each industrial setting, butwhich wassimultaneously diffused in the com-
munitythroughthefreecirculationofworkersandtechniquesviamarketsand
the professional circuit. The unblocking of bottlenecks and quality improve-
ments made it possible to begin to consider foreign markets, an idea which
wouldhavebeenunthinkableonlyafewyearsearlier.Itisalsothecasethatthe
impulse towards external markets often came as much from public policies
whichBrazil,ArgentinaandMexicobegantoimplementinthe1960sasfrom
cyclicalfallsininternaldemandwhichtookplacewiththerecurrenceofstop
gocyclesinthoseeconomiesmostaffectedbytheinstabilityofforeignfinance.
Itcanhardlybesaidthatinthefirstfewyearstheseindustrialcompanieshada
realdesiretoexport,ashappenedinthecountriesofSoutheastAsiaunderthe
ironhandofmilitarygovernmentsatleastasauthoritarianasanytobefound
in Latin America, but extremely strongly focused on the idea of national
excellenceinthefieldofhighlytechnologicallysophisticatedmanufacturing.
Wewillnowattempttostylizethemicroeconomicbehaviourdescribedupto
this point. Let us consider the case of a firm which decides to create a small
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
departmentoftechnicalproductionassistanceinwhichitemploysbetween5
and10techniciansand/orprofessionalsinordertocreatemoretechnological
knowledge so as to improve product designs, production engineering and
organization of their labour force. The activity of these professionals and
technicians contributes to the gradual modification of the establishment's
operational routines. It does this by means of a flow of minor technological
changes of the type described in the literature on the microeconomics of
technological change by writers like Hollander (1965), David (1975), Stiglitz
andAtkinson(1969)andothers.Whatwehaveexplainedsofarcanbeshown
graphicallyasinFigure2.1.
Figure2.1showsthathavingstartedfromasuboptimalpositioninwhichit
was far from achieving satisfactory physical performance from the available
production capabilities, the firm succeeded in gradually improving these by a
systematicflowofincrementalimprovementsinitsoperationalroutine.These
are minor technological changes which cumulatively result in a significant
improvementinmanufacturingproductivity.
Neo-classicaldevelopmenttheorydoesnotattachgreatimportancetothese
dynamic issues of the maturing of the internal technological capabilities of
relatively less-developed countries. It rather chooses to examine from the
point of view of free access to the stock of international technology issues
concerning theoptimumselection ofproductiontechniques andstaticineffi-
ciencyintheallocationofresourcesassociatedwiththeemploymentofmore
Real returns
Expected returns
Learning curve
Initial productivity
'Optimal process'
by the book
A
Initial operating
conditions
t T
X
*
* *
*
Q
Figure 2.1 Technological learning and improvements in production during the import
substitutionprocess
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
capital-intensivetechniquesthanthosejustifiablebytherealopportunitycost
or shadow prices of resources. It ignores the phenomenon of maturing pro-
cesses resulting fromtechnological learning. It doesnotthink in terms of the
existenceoffirm-specificknowledge,butratherofagenerictechnologywhich
doesnotneedtobeadaptedtothespecificuseeachfirmmakesofit.Theneo-
classical metaphor describes a stylized situation in which the firm has perfect
knowledge of and access to the production function which it operates. No
allowance is made for the possibility that competitive firms close to one
another may operate with different production functions, with dissimilar
modelsoforganizationofthelabourforce.
In contrast, various factory-level studies completed in a number of Latin
Americancountriesoverthelastthreedecadeshavemadeitpossibletorecon-
struct a setting of microeconomic behaviour which differs significantly from
that offered to us by the conventional neo-classical model. The above-
mentioned studies (Katz, 1986 and 1987; Kosacoff, 1993; Lall, 1993; Teitel,
1984; Dom nguez and Brown, 1995a and 1995b; Katz and Vera, 1997) show
that the efforts at technological adaptation carried out within each manufac-
turingplantwiththeintentionofimprovingproductdesign,productionengi-
neering and organization of labour explain a significant fraction of the
observed productivity increases. In many cases some two-thirds of these
increases took place as a result of engineering experiments carried out in
ordertoimproveroutinesonthefactoryfloor.Atthesametimeitisimportant
toobservethatthisrateofproductivityincreaseoftenendsupbeingfarhigher
thanthatobservedattheinternationallevelintheareaofactivityinquestion,
allowing the firm to achieve a gradual approximation towards international
levels ofqualityandefficiency.Inthesecases thescenarioisoneinwhich we
canseeaclearnarrowingoftherelativegapofproductivityandquality.
The firm's search for new technological information seems to follow a de-
finitepatternovertime,movingfromengineeringdesignsfornewproductsto
processtechnologyand,sometimelater,toknowledgeabouttheorganization
oflabour (Katz,1986).Studiesoftimingandmovements,ofthecoordination
of production lines, of the optimization of the lay-out of the manufacturing
plant and so on, have been the business of the second decade of many firms'
productive histories, when their level of technological learning was already
advanced, and with improvements in production design and quality already
achieved(Katz,1986).Whilethereweremanyfirmswhichundertookactivities
inthe1960stoimproveproductiondesignandproductionprocesses,effortsto
solve problems regarding the organization of manufacturing labour seem to
havebeeninitiatedonlysomeyearslater,inthe1970s.
Themodelofmicroeconomicbehaviourexaminedhithertomakesitpossible
to understand the accumulation of technological capabilities within the firm
anditsgradualimpactovertimeonmanufacturing productivityaswellason
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
thedegreeoforganizationalandtechnologicalsophistication.Italsoallowsus
todescribehow,asaresultofthesechanges,thecomparativeadvantageofthe
firmkeepschangingdynamically,asdoesitspotentialforengagingwithinter-
national competition, especially in those sectors in which changes in the
international technological frontier are not too abrupt. In many of the cases
mentioned above this has in fact taken place, and as a result of the
maturing process the firms increased their export commitments, moving
from insignificant amounts to figures of the order of 15 to 20 per cent of
their total sales going to other markets. These markets were initially those in
neighbouring countries, and only later those of the more industrially devel-
oped countries. What is more, the gradual increase in internal technological
capabilities and the growing degree of technological sophistication of many
of these firms meant that some of them not only moved into exports of
final products but also exported pure technology in the form of complete
manufacturing plants, delivered as turnkey projects to companies in other
countriesintheregion,aswellaslicensesforproductsand/orprocesseslocally
designed and/or adapted to domestic production conditions, and services of
technical production assistance to companies with lower levels of technolo-
gicaldevelopment.
Itisobviousthatthetechnologicallearningtowhichwehavebeenreferring
has taken place in given macroeconomic and sectoral settings, in a specific
institutionalandregulatoryatmosphere,andwithinaparticularframeworkof
dynamic interdependencies with other local firms, suppliers of capital equip-
ment and services associated with production. These macro/microeconomic
links,likethesystemicinterdependenciesreferredtointhelastparagraph,are
very poorly understood by the conventional neo-classical model, which only
conceives of arm's-length relations and universal automatic behaviour pat-
terns. In contrast, our studies have shown that there are very distinct institu-
tional frameworks, regulatory frameworks and patterns of interdependency
betweenindividualeconomicagentswhichconditioneachoftheirstrategies,
the type of technological efforts undertaken in their plants, and the impact
these efforts have on productivity, dynamic comparative advantage and
employment.
In contrast, we may note the distinct microeconomic base underlying the
differingconductofAsianproducerswho,fromamoreorlesssimilarstarting
point, and having to survive the changes of a complex maturing process
marked by each industry's technological restrictions, nevertheless progressed
along a more agile and successful evolutionary path than that achieved by
Latin American firms. A higher rate of savings and investment, a different
degree of competitive pressure, more significant and coercive government
policy forcing firms to move into external markets, a systemic scenario
of rapid improvement in skilled human resources and in society's general
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
technological infrastructure as a result of public spending, provide different
explanationsforwhattookplaceinthetworegions.
A further level of complexity in the discussion is that we cannot simply
assumethattheseforceswerepresentintheonecaseandabsentintheother,
giving way to a general picture of failure in Latin America. Rather, an evolu-
tionaryprocesslikethatdescribedherecanbedetectedinnumerousArgentine,
Brazilian, Mexican and Colombian firms, as well as in companies in Spain,
India or othercountries which explicitly pursued policies of supporting man-
ufacturingdevelopment.Productssuchascars,farmequipment,capitalequip-
ment with low or medium levels of technological complexity, and refined
chemicals, appear among those in which an evolutionary process of the type
describedhereseemstohavebeenatthebaseoftheindustrialdevelopmentof
the peripheral world during the 1970s. This evolutionary process can be seen
onlyasaclearconsequenceoftheexpansionofthemanufacturingsector,and
itissurprisingthattheneo-classicalwritershavehaddifficultyincapturingthe
dynamicimpactofthistransformation oftheproductivesetting,insteadpay-
ing more attention to matters of static inefficiency in the allocation of
resources. These inefficiencies were certainly also present as a result of the
strategies which supported the development of manufacturing production,
butareonlyapartofwhattookplace,andprobablytheleastinteresting.
Our characterization of the substitution process is definitely very different
fromthatofferedtousbytheneo-classicalmainstreamintheprofession,and
covers a large number of dynamic historical factors which the neo-classical
tradition simply misses. Rather than issues of static inefficiency in the alloca-
tion of resources, our analysis places emphasis on long-term dynamic aspects
relatedtothematuringofthesociety'sinternaltechnologicalcapabilities.
Justasinthecaseoftechnologicallearning,whereamicroeconomicapprox-
imation of the real world in developing countries allows us to identify and
describe a great variety of facts and variables which the neo-classical research
agendasimplyleavesaside,thesamehappenswhenweapproachthesubjectof
institutionaldevelopment.Itisanother`blackbox'aboutwhichconventional
theoryhaslittleornothingtosay.Wewillexplorethissubjectinthefollowing
pages.
InstitutionbuildingduringtheISIyears
As with the matter of technological change and the development of local
technical capabilities, the study of technological change in a given society
requiresagreateffortofhistoricalcontextualization.Institutionsunderstood
asnormswhichregulatethebehaviourofindividualeconomicagents,astheir
behaviouralhabits,andfinallyasentitiesoragencies,publicaswellasprivate,
which intervene in a community's day-to-day life, such as banks, unions,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
universitiesandsoondonotoperateinavacuumbutinahistoricalperiod,in
acertainorganizationalandproductiveculture.
Itis,however,alsothecasethatinstitutionschangegraduallyovertimeand,
despitekeepingtheirname,oftenchangetheirrolewithincommunitylife,pari
passu with the changes taking place in its customs, in its degree of openness
andsoon.InthewordsofPaulDavid,Ìnstitutionsarethecarriersofhistory'
(David,1994).
Thisbeingthecase,andinordertofurtherourunderstandingoftheinstitu-
tional setting which characterizes the years of industrial development in the
postwar period, it will be useful to carry out a counterfactual experiment and
imagineourselvesattheclosingstagesofthegold-standardregime.Letusask
whatwasthetypicalinstitutionalframeworkinthemultiplesenseusedhere
oftheperiod,andhowinstitutionschangedorwerereplacedfollowingthe
collapseofthenear-automaticgold-standardregime.
Itisimportanttounderstandthattheregulatoryandinstitutionalframework
of the gold-standard model is more automatic, more conditioned and simpli-
fied than that which any of the Latin American countries had to construct
duringandfollowingthewarinordertoadapttothenewrulesofthegame.It
worked based on the free convertibility of currencies and on multilateralism,
allowingthecompensationofpositiveandnegativecurrentaccountbalances.
Monetarypolicyplayedonlyapassiverole,andthecentralbankhadnoactive
roleinmattersofmonetarypolicy.Adjustmenttotheexternaleconomiccycle
tookplacethroughthelevelofinternalproductiveactivity,monetaryoutflows
being backed by reserves of foreign exchange. In this context, customs tariffs
onlygeneratedfiscalincome,andwerenotconsideredamechanismforindu-
cing investment, that is, as a way of encouraging the installation of new
industrial plants or the development of new productive activities. The rate of
exchange,theinterestrate,andfiscal,monetaryandincomepolicyweresub-
ject to the requirements imposed by the need for equilibrium in the external
sector of the economy. In general terms it was the opportunity cost of
resources, bestowed by their international price, that controlled their alloca-
tionwithintheproductionapparatus.
It is also important to note that this model eliminates much of the discre-
tionary nature of government policy and puts policy makers in a position
where they have little freedom to pursue paths which are unexpected and
uncertain. From this point of view the near-automatic adjustment model of
thegoldstandardshouldbeseenasaregimeofmacroeconomicsignalswhich
arerelativelyclearandlargelyabletoeliminateuncertaintyforprivateagents
inrelationtothebehaviourofkeyprices.
Thecollapseofthegoldstandardandthedisappearanceoffreeconvertibility
tended to manifest themselves as crises of external finance which forced the
countries of the region to devalue their currencies and introduce quotas and
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
tariffandnon-tariffrestrictionsinordertoreducetherateofdomesticabsorpt-
ion of goods and services. The aim of these actions was an increase in expor-
table surplus and, at the same time, a limiting of the demand for foreign
exchange. It also encouraged domestic production of previously imported
goods, either from public companies or by encouraging private investment
throughtaxexemptions,subsidizedcreditorbankendorsementstoguarantee
theacquisitionofequipmentorinputsabroad.
Ofcourseallthisimpliedachangeintheregimeofmacroeconomicincent-
ives and the regulatory framework in which production activity was develop-
ing, andthe introduction ofa newgroup of signals which aimed at changing
thebehaviourofindividualeconomicagents.Wheretheinvisiblehandofthe
market had previously operated, there was now the visible hand of the state,
introducing customs tariffs, subsidized interest rates, investment endorse-
ments, foreign exchange permits and tariffs for public services which did not
covertheirownproductioncosts.Eachandeveryoneofthesemechanismswas
employedasaresponsetotheexternaldisequilibriumarisingfromthecollapse
of thegoldstandard, andineachcaseitwasnecessarytodevelopnewnorms
and agencies for applying them. This action in turn caused new adaptive
behavioursonthepartofindividualeconomicagents,andthustheregradually
evolved a new interactive organizational culture within the community. It is
obvious that relations between the state and civil society began to take shape
throughlinksofdirectinterdependencywhichwentfarbeyondthatinvolved
inthefreeplayofthepricemechanism.Weshouldexpectapriorithatuncer-
tainty, transaction costs and opportunistic behaviour would be greater in this
contextthaninthenear-automaticsettingofthegoldstandard.
As this sequence of events was taking shape in the various countries of the
region, institutions adapted and new government bodies took charge of the
day-to-dayrunningofthenewregimeofpublicpolicy.Thusnewgovernment
officeswerecreated,responsibleforallocatingimportpermits,bankingdepart-
ments in charge of granting investment endorsements, central banks taking
chargeoftheimplementationofactivemonetarypolicies,developmentbank-
inginchargeofprovidingindustrieswithsubsidizedcreditandmade-to-meas-
urepromotionalsystemsforthevarioussectorsofindustry.Paripassuwithall
this, there was also an increase in chambers of commerce, trade unions, and
otherorganizationswithincivilsociety.
The institutional framework of the period also fed on new modes of beha-
viour derived from the period of war and the high degree of nationalism
reigning in the main countries of the region, and this helps us to understand
the reason behind the nationalization of natural resources, of foreign com-
merceandoftheprincipalpublicservicestransport,energy,telecommunica-
tions which frequently accompanied ISI strategies. The creation of
technological knowledge passed into the hands of the large R&D laboratories
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
created within state firms, such as YPF or YCF (Argentina), Petrobras (Brazil),
Pemex(Mexico).Developmentbankingalsoremainedunderstatecontrol,the
principalagenciesfulfillingthisfunctionbeingNafinsa(Mexico),BNDE(Banco
NacionaldoDesenvolvimentoEconomico)(Brazil),theInstitutoArgentinode
Promocio n Industrial and the Banco Nacional de Desarrollo (Argentina), and
Corfo(Chile).
Further,thebeginningsofan industrialurbanproletariatanditsmanifesta-
tionintheformofunionsforeverybranchofactivityandtheappearanceofa
newgroupofinstitutionsinthelabourmarketmakesupanotherphenomenon
typicaloftheperiod.
There are indisputably differences in how the different countries of the
region went about constructing this vast institutional apparatus, in the way
inwhicheachsocietydevelopednormsandbehaviourpatterns,newroutines
ofsocialinteractionandsoon,overthecourseofthedecadesstudied.Without
questioneachcountryshouldbeseenaspossessingspecificcharacteristics,but
alsomovingforwardinsettingupanewregimeofmacroeconomicincentives
andregulatoryframeworkswithmanyfeaturesincommon.Ifwearetounder-
standwhattookplacewecannotignorefactorswefindthroughoutimperfect
informationandinexperienceofpublicagentsresponsibleforcarryingoutthe
transitiontothenewrulesofthegame,theirinabilitytoimposeclearperform-
ance criteria, the prevalent nationalistic ideology which resulted in the pre-
dominance in the state bureaucracy of the defence sectors and the armed
forces,featureswhichgavetheLatinAmericansubstitutionmodelsomeofits
mostmarkedinstitutionalcharacteristics.
AllthisgivestheinstitutionalmodelforISIveryspecialpolitical-ideological
characteristics. We could study what happened during the period from the
contemporary point of view of such writers as Coase and Williamson and
find in transaction costs the central explanation for much of the economic
direction and behavioural patterns developed over the years by individual
production agents and policy makers. However, we believe this interpretative
framework to be inadequate if we want to understand satisfactorily the char-
acteristicsofamodelofsocialorganizationwhichaimedatnothinglessthana
modification of the rules of appropriation of the surplus generated by the
primarysectorofsociety,andachannellingofthistofinancethedevelopment
of manufacturing. In our opinion this requires a vision of political economy
whichmoderninstitutionalisttheoriesdonotproperlyconvey.
Conclusion
Intheprecedingpageswehaveexaminedtwoissueswhichneo-classicalliter-
ature has more or less left to one side. The first is the development of local
technological capabilities in peripheral countries, the second the appearance
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
andstrengtheningofahugegroupofnewinstitutionsduringtheyearsfollow-
ingthewar.Onbothsubjectswehaveattemptedtocoverwhatisspecifictothe
organizational structure of Latin America following the collapse of the near-
automatic gold-standard regime, in order to understand the idiosyncratic
paths by which the countries of the region rejected it en masse in their
effort to construct a new development model which had internal demand
as its impetus. The idea was to recover the rate of growth of previous decades
in a world with little access to external financing, without currency convert-
ibility and clearly bearing the mark of protectionism and the climate of the
ColdWar.
SomecountriessuchasBrazilandMexico,forexample,whoseannualrates
of expansion in the long term were not significantly different from those of
KoreaorTaiwanfounditeasierthanotherstoadaptsuccessfullytothenew
conditions. At thesametime itis hard toaccept thegeneral picture of failure
describedintheneo-classicalwritingofthe1970sand1980sonLatinAmerica's
postwar industrial development. In our opinion this picture comes from the
very particularlensesthroughwhich themainstreamof theprofessionclaims
toreadtherealityoftheperiod.
Notes
1 See,forexample,Boyer(1993),orCoriatandDosi(1995).
2 AswellasSchumpeter'sbook(1942),akeylandmarkinthethinkingofthisstreamof
authorsistheseminalworkofNelsonandWinter(1982).
3 In this sense it would be instructive to examine the debate between Krugman and
Stiglitz at the Annual Conference on Development Economics, held by the World
BankinWashingtonin1993(Krugman,1993,andStiglitz,1993,pp.1549).
4 The term ìnstitutions' is used in contemporary writing in at least three different
ways which are worth explaining at the outset. First, it can be taken to mean
those norms and rules which govern the behaviour of social players. In this
sense, patent law, for example, is a regulatory institution. Secondly, the term
can refer to habits of behaviour of economic agents. In this sense David (1994)
usestheexampleofthecustomofextendingone'sopenhandtogreetastranger,a
habit which originally showed that one's approach was friendly and therefore
unarmed, but which has since become a universally-accepted convention.
Thirdly, theterm ìnstitution' canbeusedforagenciesorbodies publicorprivate
which intervene in the running of a community's day-to-day life, such as banks,
tradeunions,businessfirms,universities,thecentralbankandsoon.Onthissubject
see David (1994); Granovetter (1985); North (1996); Williamson (1985); Freeman
(1995).
5 The `stylized' image of the firm was originally conceived as an instrument for
studying income distribution in a world of constant returns to scale and factors of
productionpaidforbytheirmarginalproductivity.ItwasEuler'stheoremandnotthe
complexitiesoftheworldofproductionandindustrialorganizationthatinspiredthe
`stylized' conception of the neo-classical production scenario. Over the last two
decadestherewereverysignificantconceptualcontributionsintheareaofìndustrial
organization'basedonanacceptanceoftheexistenceofimperfectmarkets,andona
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
reconsideration of the theory of firms and state intervention. However these con-
tributions continued to assume the optimizing behaviour of firms, and do not
includetheideaofboundedrationality(Tirole,1988;SchmalenseeandWillig,1989).
6 See,forexample,Krueger(1974).
7 See the World Bank study (1993). A critical view of this work can be found in Lall
(1993).
8 Theideathattechnological knowledgeis incompletely `specified' thatis, itis not
completely documented has been developed by Nelson and Winter in various
recentworks.SeeNelsonandWinter(1982).
9 It is interesting to note that military men such as Park and Pero n served as
youngofficersintheJapaneseandItalianarmedforces,respectively,beforereturning
to Korea and Argentina, being reintegrated and eventually becoming presidents
of the respective Republic. During their period of military training they not
only received training in warfare but also developed a world vision they would
have the opportunity to implement, and a view of the bond between civil society
andmilitary societywhichtheywouldeventuallyimposeontheirrespectivecoun-
tries.
10 Itisinterestingtonotethatthisproductiveplayerdoesnotappearinthesameform
inthecontextofKoreaandTaiwanalthoughitdoesyearslaterinSingaporeand
thatitmakesasignificantdifferencetosocialinteractionandthemodelofindustrial
organizationprevalentinthesesocieties.
11 Ford, for example, had been established in Argentina since 1919, but only really
begantoproducecarslocallyin1962.Thecompanyhadpreviouslybeenimporting
units which were nearly finished, to which were added locally produced tyres and
batteries;althoughthereweretariffs,thesewereforfiscalreasonsandnotasincent-
ivestomanufacturingproductionanddomesticverticalintegrationaswouldbethe
caselateron.
12 The case of Siam Di Tella, the Argentine automotive firm, and its inability to
compete with the large transnational firms in the sector, clearly illustrates the
process referred to in the text. On this subject see Bisang, Burachik and Katz
(1995),pp.2526.
References
Bisang,R.,Burachik,G.andKatz,J.(1995)Haciaunnuevomodelodeorganizacionindustrial.
ElsectormanufactureroArgentinoenlosanos90,BuenosAires:AlianzaEditorial.
Bisang, R. and Kosacoff, B. (1993) `Las Exportaciones Industriales en una Econom a
en Transformacio n: Las Sorpresas del Caso Argentino 197490, El desaf o de la
competitividad. La industria argentina en transformacion, Buenos Aires: CEPAL/Alianza
Editorial.
Boyer,R.(1993)ÀbouttheRoleandEfficiencyofMarkets:History,TheoryandPolicyin
theLightoftheNineties',Monograf a9320,August,Paris:CEPREMAP/CNS.
Coriat,B.andDosi,G.(1995)`TheInstitutionalEmbeddednessofEconomicChange.An
Appraisal of the Èvolutionary' and `Regulationist' Research Programmes', Interna-
tional Institute for Applied Systems Analysis (IIASA), WP (working paper)-95117,
Luxembourg.
David,P.(1994)`WhyareInstitutionsthe`CarriersofHistory'?PathDependenceandthe
Evolution of Conventions, Organizations and Institutions', Structural change and eco-
nomicdynamics,vol.5,no.2.
(1975)TechnologicalChoice,InnovationandEconomicGrowth,Cambridge:Cambridge
UniversityPress.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Dom nguez,L.andBrown,F.(1995a)`LaNaturalezaMicroyMacroecono micadelCam-
bioTecnico:UnCasodelaIndustrialAlimentaria',mimeo,UniversidadAuto nomade
Mexico,November.
(1995b)`LaNaturalezaMicroyMacroecono micadelCambioTecnico:UnCasodela
IndustriaMetalmecanica',mimeo,UniversidadAuto nomadeMexicoAugust.
Dunning,J.H.(1988)ExplainingInternationalProduction,London:UnwinHyman.
Freeman, C. (1995) `The National ``System of Innovation'' in Historical Perspective',
Cambridge Journal of Economics, vol. 19, no. 1, February, Cambridge: Cambridge Uni-
versityPress.
Granovetter, M. (1985) Èconomic Action and Social Structure: The Problem of Embed-
dedness',StructuralChangeandEconomicDynamics,vol.91,no.3,November.
Hollander S. (1965) The Source of Increased Efficiency, a Study of Dupont Rayon Plants,
Cambridge,Mass.:MITPress.
Katz, J. (1987) Technology Generation in Latin American Manufacturing Industries, Hong
Kong:Macmillan.
(1986) Desarrollo y Crisis de la Capacidad Tecnologica Latinoamericana. El caso de la
industriametalmecanica,BuenosAires:ECLAC.
(1976)ImportaciondeTecnolog a,AprendizajeeIndustrializacionDependiente, Mexico:
FondodeCulturaEcono mica.
Katz, J. and Vera, H. (1995) `Historia Evolutiva de una Planta Metalmecanica Chilena:
RelacionesMicro-Macroydesarrollotecnolo gico',mimeo,October.
Katz, J., and Vera, H. (1997) `Historia Evolutiva de una Planta Metalmecanica Chilena',
ECLACJournal,vol.63,December.
Kosacoff,B. (1993)Eldesaf o de lacompetitividad: laindustriaArgentinaentransformacion,
BuenosAires:AlianzaEditorial.
Krueger,A.(1974)`ThePoliticalEconomyoftheRent-seekingSociety',AmericanEconomic
Review,vol.64no.3.
Krugman,P.(1993)`TowardaCounter-CounterrevolutioninDevelopmentTheory',Pro-
ceedings of the World Bank Annual Conference on Development Economics, Washington,
DC:TheWorldBank.
Lall,S.(1993)TheEastAsianMiracleStudy:DoestheBellTollforIndustrialStrategy?,working
paper no. 67, International Development Centre, Queen Elizabeth House, Oxford,
October.
(1992)`TechnologicalCapabilitiesandIndustrialization',WorldDevelopment,vol.20
(2),pp.16586.
Nelson,R.andWinter,S.(1982)AnEvolutionaryTheoryofEconomicChange,Boston,Mass.:
TheBelknapPressofHarvardUniversity.
North, D. (1996) Èconomic Performance through Time: The Limits to Knowledge',
mimeo,August,WashingtonUniversity.
Schmalensee, R. and Willig, R. (eds) 1989) Handbook of Industrial Organization, Amster-
dam;London:North-Holland.
Schumpeter,J.(1942)Capitalismo,SocialismoyDemocracia,NewYork:HarperTorchbooks.
Stiglitz, J.E. (1993) `Comment on ``Toward a Counter-Counter revolution in Develop-
mentTheory''',ProceedingsoftheWorldBankAnnualConferenceonDevelopmentEconom-
ics,Washington,DC:TheWorldBank.
Stiglitz, J.E. and Atkinson, A.B. (1969) À New View of Technical Change', Economic
Journal.
Solow,R.(1988)`GrowthTheoryandAfter',TheAmericanEconomicReview,vol.78,no.3.
Teitel, S. (1984) `Technology Creationin Semi-Industrial Economies', Journal of Develop-
mentEconomics,vol.16,pp.3961.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Tirole,J.(1988)TheTheoryofIndustrialOrganization,Cambridge,Mass.:MITPress.
Williamson,O.E.(1985)TheEconomicInsitutuionsofCapitalism:Firms,Markets,Relational
Contracting,NewYork:TheFreePress.
WorldBank,1993)TheEastAsianMiracle:EconomicGrowthandPublicPolicy,Washington,
DC:PolicyResearchDepartment,WorldBank.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
3
ECLAandtheTheoryofImport
SubstitutingIndustrializationinLatin
America
E.V.K.FitzGerald
Introduction
Import-substituting industrialization (ISI) has undoubtedly played a central
role in the economic development of Latin America in the present century.
Nonetheless, the effectiveness of this process as a basis for sustainable output
growth, rising living standards and social modernization is still strongly con-
tested. The critique of ISI is thus not only a question of understanding a
particular period of economic history, it is also a matter of evaluating the
current economic strategy of the region, based on increasing integration into
worldmarketsandlessstateinterventioninindustry.Thisisexplicitlydefined
incontrasttowhatisperceivedasthepreviousISIstrategy(Edwards,1995).
Import substitution has taken place naturally at least since the conquista
whenever transport costs and market size made it profitable, but the genesis
of ISI as a strategy is generally regarded as originating as a response to the
externalshocks
1
oftheGreatDepressionandtheSecondWorldWar(Ground,
1988).Despitealongtraditionofdebateonindustrializationandtariffsinthe
region, and exposure to similar debates in Central Europe between the wars,
formaleconomictheorizationaboutISIonlybeganinLatinAmericaafterthe
HavanaConferencein1947(FitzGerald,1994).
This theorization was initiated by the UN Economic Commission for Latin
America (ECLA)
2
and subsequently generated a robust critical response from
neo-classicaltradetheorists,broadeningintoaworldwidedebateontheappro-
priate development strategy for the `Third World' emerging from the decolo-
nizationprocess:
3
Inthiscentury,andespeciallysincetheSecondWorldWar,industrialization
indevelopingcountrieshastypicallymeantimportsubstitution.Industries
havebeensetuptoproducegoodsthatwerepreviouslyimported,andthese
goods have mainly been sold in the home market. Governments have
58
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 59
ensured the profitability of these industries by protecting them against
competingimportsthroughtariffsandcontrols.
(Little,ScitovskyandScott,1970,p.1)
Indeed import substitution and public investment remained at the centre of
acceptedindustrializationstrategiesalbeitwithconcernabouttheefficiency
oftheresultingproductivestructureuntiltheendofthe1970s(WorldBank,
1979).
InLatinAmerica,ECLAtheoryformedthecoreof`structuralism'
4
andhada
profound influence not only in academic circles but also among economic
policy-makers throughout the region. Indeed, during the process of regime
shift during the 1980s towards the subsequent desarrollo hacia fuera, or out-
ward-looking development, not only have most of the economic problems of
previousdecadesbeenattributedtoISIbutalsothecepalinoshavebeenblamed
forthegovernmentpoliciesofdesarrollohaciadentro,orinward-lookingdevel-
opment, which made this strategy possible.
5
However, there appears to have
been little attempt to formalize ECLA theory, and thus work out the wider
implications of specific insights into market-clearing or institutional beha-
viour.Thisinturnmadeitmoredifficulttoengageinacoherentdebatewith
mainstreameconomicsindeedthemetamorphosisofstructuralisteconomics
into dependency theory by the 1970s appears to have reflected a positive
distrust of formal economic theorization in favour of essentially sociological
interpretationsof`worldsystems'.
6
This chapter attempts to restate the structuralist theory of ISI elaborated by
the ECLA, in terms of comparable economic theory, rather than address the
experienceofISIassuchwhichistheconcernofthecountrychaptersinthis
volume. In the next section the structuralist model of capital accumulation
basedonISIissetoutintermsofopen-economymacroeconomics,theexternal
constraint on growth and the employment problem. The contemporary neo-
classical critique of the ISI model is then discussed, the main elements being
the welfare loss from abandoning comparative advantage and the sectoral
imbalance brought about by protection; although it is also shown that ECLA
itselfwashighlycriticaloftheISIasitdevelopedinpractice.Wethenexamine
theissueoftechnologicalchangeinECLAtheory,andsuggestthatitmayhave
anticipated elementsofsubsequent theories ofintra-industry trade andendo-
genous growth, even if total factor productivity growth under ISI was rather
low.Thisisfollowedbyanexaminationofthe`politicaleconomy'explanations
oftheproblemsofimportsubstitutionintermsofthelackofastrongindustrial
class, rent-seeking from government intervention, and populist macroeco-
nomic policy all of which are supposed to be related closely to ISI. Finally,
in the final section we conclude that the structuralist theory of import-sub-
stitutingindustrializationisbestseenasamodelofstate-ledindustrialization,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
60 ECLAandtheTheoryofISI
anditisthefiscalcontradictionscontainedinthismodelwhichhavedeterm-
inateconsequencesforfiscalanddistributiveimbalances.
Thetheoryofimportsubstitutionasamodelofaccumulation
Theoriginsofthemodel
Thestructuralisttheoryofimport-substitutingindustrializationisessentiallya
modelofaccumulationinthebroadsenseofbeingconcernedwithlong-run
growth, investment, employment and income distribution rather than with
static efficiency as such. It also has clear historical foundations. The explicit
root is an interpretation of the development of the Latin American economy
duringtheGreatDepressionandtheSecondWorldWar
7
whichsawtheexperi-
enceofimportsubstitutionduringtheprevioustwodecadesasa`spontaneous
process' of both governments and firms responding to external shock (Rodr -
guez, 1980). Rather than a consciously directed strategy of industrialization
andmacroeconomicdemandmanagement,thisprocesswasunderstoodtobe
the outcome of manufacturers utilizing existing plant as competitive imports
fellandgovernments strugglingtomaintainexpenditureastradetaxrevenue
declined.ECLAtheorizationwasthusconcernedwithunderstandingthelogic
of this process, with the aim of rationalizing it rather than designing a new
industrializationstrategy.
The implicit root of this model is to be found in the region's experience of
reintegration to the world economy in the 1940s. The role of US suppliers
during the postwar period, active participation in the birth of the UN, and
Washington'sinitialforeignpolicyallgavetheimpressionthatLatinAmerica
wouldjointheless-advancedindustrialcountriessuchasSouthernEurope
andbaseitscontinuedindustrializationontheexpansionofprimaryexportsto
theUSA(andeventuallytoareconstructedEurope)atreasonableprices.Man-
ufacturing expansion was to be based on domestic markets, to be eventually
followed by exports of labour-intensive products such as textiles. However,
when the creation of the ITO was blocked and US protective barriers raised
after the Havana Conference of 1947, it seemed that this export-oriented
industrialization strategy was no longer a strategic option for Latin America
(FitzGerald,1994).
Three main justifications for ISI were put forward in the initial ECLA form-
ulation during the late 1940s and through the 1950s. First, the external con-
straint on growth was attributed to the declining terms of trade for primary
products(thatis,price)andlackofmarketaccess(thatis,volume)forprimary
exports,requiringsomedomesticsourceofgrowth.Second,therewasasocial
needtorapidlyincreaseemploymentinordertoabsorbthegrowingworkforce
andofferbetteropportunitiestounderemployedlabourinpeasantagriculture,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 61
which the primary export sectors could not achieve. Third, the only way to
generate rapid technological progress was through industrialization, because
thebenefitsofanyprogressinprimaryexportproductivityaccruetoimporters
ratherthanexporters.This`model'ineffectcanberegardedasalogicalexten-
sion of modifying three key assumptions of the standard neo-classical trade
model:thatacountryfacesaninfinitelyelasticexportdemandatgivenworld
prices; that there is full employment and intersectoral mobility of labour and
capital;andthattherearenosignificantexternalitiesfrominvestment.
AsaninstitutiontheECLAlogicallyadoptedaspecificallyregionalviewpoint,
as opposed to that of a single country (Rodr guez, 1980). This meant that
regional coordination in import substitution was taken for granted, and thus
the limitations on plant scale imposed by country size could be overcome by
local market integration. Further, it implied that the familiar small-country
assumption of standard trade models could not be applied, not only because
thedemandcurvewouldbedownwardslopingforLatinAmericaexportersasa
whole,butalsobecauseexportmarketgainsbyonecountryattheexpenseof
another could not be considered a welfare gain from the point of view of the
regionasawhole.
Thetwo-sectoropeneconomymodelofISI
Thecepalino model ofISIlinked todeteriorating terms of tradewas setout by
Prebisch in a two-sector form with endogenous price formation (Prebisch,
1959; discussed in Rodr guez, 1980, ch. 5). There are two sectors: primary
exports and domestic industry; with free competition, mobile labour and
declining returns. Any expansion of exports reduces marginal incomes to the
primary sector due to both inelastic world markets and increasing unit-cost
productionconditions;andasthereturnoncapitalisgiven,thismainlyaffects
wages in both sectors, which equalize through the labour market. Marginal
incomesinthetwosectorsarethusequalizedatequilibrium.Industrialexpan-
sion requires lower real wages to combat increasing costs and maintain com-
petitivenesswithimportsfromtheworldmarket.Theseassumptionsgeneratea
market-clearing equilibrium model for response to external shock
8
which is
comparableinformtothewell-knownneo-classicalmodel(forexample,Cor-
den,1984)butfromwhichratherdifferentresultsarederiveddueprimarilyto
theendogeneityofpriceintheexportfunction.
Prebisch's own exposition is rather opaque and inconsistent in parts par-
ticularlyonthelabourmarketsohereweshallsetoutthemodelformallyin
terms of the adjustment of the capital stock (K), with labour not acting as a
constraint on output. Consider an economy with two sectors (i), primary
exports (1) and industry (2), each of which has a production function of the
form:
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Q
i
A
i
K
i
gi
, 0 >g
i
>1 3:1
and where a given capital stock (K) can be freely allocated
9
between the two
sectorsbyproducers:
K
1
K
2
K 3:2
The price of industrial output (P
2
) is determined exogenously on world mar-
kets,buttheworldpriceofexports(P
1
)dependsonexportsupply
P
1
p Q
1
b
, b<0 3:3
Underthesecircumstancesinvestorswillusecapitaluptoapointwhereatthe
going cost of capital (r) they can just make a profit, yielding the familiar
equilibriumcondition:
@Q
1
@Q
2
P
1
rP
2
3:4
@K
1
@K
2
Differentiating equation(3.2)andsubstitutingintoequation (3.4)thenyields
thefamiliarconditionfortheequilibriumallocationofproduction:
@Q
2
P
1
3:5
@Q
1
P
2
However, fromthe point ofview of theeconomy asa whole, itis aggregate
profits(R)andthuswelfarewhichshouldbemaximized,
R P
1
Q
1
P
2
Q
2
K r 3:6
theconditionforwhichis
@Q
2
P
1
1b 3:7
@Q
1
P
2
This optimal equilibrium in (3.7) thus leads to a higher level of industrial
output (Q
2
) and a lower level of export (Q
1
) than the market would set in
(3.5). In other words the maximization of profits
10
requires an appropriate
relative price adjustment specifically an import duty based on the export
elasticity (b) and this constitutes the comparative-statics case for protection.
NotethatthePrebischmodelisunliketheneo-classicalopen-economymodel
as there is surplus labour and a downward sloping demand curve for the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 63
exports, rather than infinite elasticity for the small country and full employ-
ment. Nonetheless, the Prebisch model still lies within the mainstream tradi-
tionwheremarketsclear,butnowyieldasuboptimalresult.
The effect of exogenous price shocks in the Prebisch model is asymmetric,
because incomes are measured in industrial prices, so exporters are affected
more by an external price fall than industry. At equilibrium, output and
employmentinthetwosectorsaredefined,andtheexchangerateisendogen-
ousbeingthatratewhichbalancestheexternalaccount.Deteriorationofthe
international terms of trade will then automatically trigger a devaluation
which will stimulate export supply, and thus further oversupply and so on.
Evenifexportdemandriseswithgrowthatthecentre
11
oftheworldeconomy,
theresulting income rise on theperiphery generateshigherimport demand
due to the high import coefficients in industry and balance of payments
problems.Theconsequentdepreciationoftheexchangerateagainshiftslabour
towardsexportsandthetermsoftradedeterioratefurtherasexportsrise.Non-
tradedindustrialoutputexpandsasdomesticpricesriseandrealwagesfall;but
thisgeneratesevenmoreimportsduetothetechnologicaldifferencebetween
centreandperipheryandtheexternaldisequilibriumisperpetuated.
Thekeyassumptionforthisperverseoutcomeistheelasticityofdemandfor
exportsboththelow-incomeelasticityandtheslopeofthepricecurveand
the technological import-dependence of industry. In this sort of ùnequal
exchange' the level of national income is less than what it would be under
either acollectively negotiated primary export price or a deliberate process of
industrialization. Market clearing does not lead to an optimal solution in
Prebisch's model because marginal products in the two sectors are not equal-
ized.Althoughmarginalreturnsareequalizedtomarginalcostinbothsectors
intheconventionalmanner,intheexportsectormarginalrevenueisnotequal
to the (world) price. An optimal solution at least from the region's point of
viewwouldtakeadvantageofmonopolypower,leadingtofewerexports(and
higher export income) and more industrial output. Prebisch concludes that
without such a collusion between exporters, a second-best equivalent of this
outcomewithcompetitivecapitalandlabourmarketscanonlybeobtainedby
altering relative domestic prices through appropriate adjustment of import
dutiesorexporttaxes.
Growth,thetermsoftradeandimportsubstitution
In its first formulation, the structuralist model of ISI was derived from the
supposed secular deterioration in the terms of trade.
12
It was argued that the
only way to raise the rate of growth in Latin America is to reduce the import
contentofsupply,becauseforgiventermsoftrade(T),importsintotheCentre
(M
1
)areafunctionofthelevelofincomeintheCentre(Y
1
),andthetermsof
trade(T),andimportsintothePeripherydependonPeripheryincome(Y
2
)and
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
thetermsoftrade;butCentre-Peripherytrademustalsobalancebydefinition,
so
3:8 M
1
m
1
Y
1
a
1
T
b
1
M
2
m
2
Y
2
a
2
T
b
2
3:9
M
1
T M
2
3:10
whichsolvesforSouthernoutputasafunctionofNorthernoutput

a
2
Y
2
m
2
Y
a
1
T
1b
1
b
2
3:11
m
1
Forgivenprice(b)andtechnical(m)coefficients,thenbecausefromEngel's
Lawtheincomeelasticityofdemandforimports(a)inthecentreislowerthan
that for the periphery that is, a
2
>a
1
then growth at the periphery will
alwaysbelessthanatthecentre.Inaddition,thedegreeofincomeconcentra-
tion in the periphery also means that there is a disproportionate demand for
imported(`luxury')productsandacorrespondinglackofdomesticdemandfor
therelativelysimpleproductsofdomesticmanufacturing.Inconsequence:
@Y
2
@Y
1
< 3:12
@t @t
Assuming that the centre coefficients are exogenous,
13
it is clearly only by
reducingthetwocoefficients(b,m)intheperipheryimportdemandequation
(3.11) that growth there can be speeded up and this can be achieved by
import substitution.
14
In other words, the argument for protection did not
derive from static efficiency considerations, but rather from dynamic growth
considerationsandasymmetricmarketstructure.
Two distinct stages of ISI were subsequently distinguished (Furtado, 1970;
Fajnzylber, 1983). The first initial stage was the substitution of mass non-
durable consumer goods and basic inputs behind tariff protection, with con-
tinued supply of consumer durables, complex inputs and capital goods from
abroad relying on an assured domestic market, ample credit and investment
andfamiliartechnology.Thesecondadvancedstageinvolvedsubstitution
for consumer durables, complex inputs and capital goods. This would face a
more demanding market formed by primary exporters, existing consumer-
goodsbranchesandeliteconsumersrequiringheavytechnologicalinvestment,
extensivegovernmentsupportandqualitycompetitionwithforeignsuppliers.
TheassumptionsoftheoriginalECLAmodelcanthusbemademorerealistic
bydistinguishingbetweenthesetwostagesofISIintermsofstructuralshiftsin
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 65
theincomeandpriceelasticitiesofimportsastheresultoftheindustrialization
process itself (Villareal, 1988, pp. 1447). During the initial stage imports
constrain growth because the income elasticity of demand is greater than
unity;whilelowpriceelasticitymeansthatdirectinterventionisnecessaryto
reduce the import coefficient. Once domestic industry produces previously
imported goods in the advanced stage, the price elasticity of demand for
imports will rise; while the saturation of domestic markets will reduce the
income elasticity; so industrialization strategy can place greater reliance on
exports rather than domestic demand, and exchange rate policy will become
more effective in containing imports. The schematic representation of this
processissetoutbyVillarealas
Initial Advanced
Importdemand:
incomeelasticity(a) >1 <1
priceelasticity(b) <1 >1
Thereisnoreason,however,whyonestageshouldfollowautomaticallyfrom
the other: this depends on the size of the domestic market permitting the
scale economies required for capital goods production and upon the estab-
lished ìnitial' industries being willing to forgo their privileged imports of
inputsandmachinery.
A central element of this model was the position that savings as such were
notaconstraintongrowth.ThiswasnotaKeynesianpositionastothelevelof
effectivedemand,butratheralogicalextensionoftheECLAargumentthatthe
short-run constraint on domestic output expansion was the availability of
foreignexchange.Investmentdemanditselfwasfelttobetoolowtogenerate
fullemploymentduetothelackofinvestor willingnessonthepartofprivate
firms in other words, savings were low because investment was low, rather
thantheotherwayaround.Theexternalconstraintandinvestorhesitationcan
be seen as Keynesian in the broader sense of liquidity preference and animal
spirits,respectively.
The low level of industrial investment despite the relatively high level of
profitability ensured by protection was attributed to a number of internal
factors (Prebisch, 1950). First, the historical recipients of export rents that
is, large landowners and mining corporations were unwilling to invest in
manufacturing;second,thescaleofmodernindustrialprojectswasbeyondthe
financial and managerial capacity of domestic entrepreneurs; third, the inad-
equate economic infrastructure energy, transport, and so on and a skilled
workforce; and fourth, the lack of technological expertise that was monopol-
izedbyforeigncompanies.Theresolutionofthesefourproblemswasregarded
byECLAasrequiringlarge- scalepublicinvestment,considerableintervention
ofthestateinproductionandindicativeplanningoftheeconomyasawhole.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Nonetheless, this strategic role for the state was always seen to be support for
the private sector and a hopefully temporary substitution for the lack of
large domestic entrepreneurs. This view
15
did not represent support for a
process of socialization ofthe means ofproduction, therefore;norcould itbe
described as `populist' in the sense of being designed to prejudice business
intereststothebenefitoftheworkingclassorsmallproducers.
Theexternalconstraint
TheECLAmodelthusdependedontherebeingarealèxternalconstraint'on
thelevelofoutput,aswellastheavailableproductivecapacity.Exportmarkets
werebelievedtobelimitedbyprotectionistbarrierstoimportstothecentrefor
manufactures,andtheoversupplyofrawmaterialsfromtheperiphery.
16
This,
ofcourse,wasalimitationonperipheralexportersasagroup;butneithercould
individual countries adjust their balance of payments to the desired growth
ratebyusingtherealexchangerate.
If exports and imports are at all elastic to the real exchange rate (e), then it
might appear that trade balance can be achieved at any level of aggregate
demand (Y) by setting e at the appropriate level, and thus that the external
sector cannot be considered as a constraint on growth. The ECLA argument,
however, does not depend on maintaining that the elasticities are effectively
zero;butratherthattherequiredleveloftherealexchangeratecannotinfact
beachievedduebothtotheeffectofthenominalexchangerateondomestic
pricesandinflation,andthesociallimitsontherealwagerate.
This point can be clarified by adapting the standard open-economy macro-
economic model
17
of domestic price formation. The real exchange rate
depends on the ratio between traded and non-traded prices, reflected in the
relationship between the nominal exchange rate (E), world prices (P
w
) and
domesticprices(P
d
)
EP
w
e 3:13
P
d
Domestic prices depend in turn on domestic costs, made up of the nominal
wage(W),theunitlabourinput(u),theunitimportinput(m)andworldprices,
adjustedbyaprofitmark-up(r)
P
d
1ruWmEP
w
3:14
Thismeansthatanynominaldevaluationthatis,increaseinEintendedto
reduce the trade gap will be `passed through' to domestic prices to the extent
that: (i) the import coefficient (m) is large and unresponsive to the exchange
rateintheshortrun;and(ii)thenominalwage(W)itselfadjuststothechange
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 67
in domestic prices. It is easily shown from equation (3.14) that for a given
elasticity of nominal wages to domestic prices (f) and share of wages in costs
(s)thentheelasticityofthesepricestothenominalexchangerate(e)isgiven
by:
1s
e 3:15
1fs
So that if nominal wages are fixed (f0) then the elasticity of domestic
inflation in response to devaluation will be less than unity (e1s), and
the real exchange rate adjusts correspondingly. However, if wages are fully
indexed to inflation ( 1) then the nominal devaluation will be ineffective
becausedomesticpriceswillrisebythesameamountasthenominalexchange
rate (e), the real exchange rate (e) will not change and thus the trade gap will
notadjust.
Even if wages are incompletely indexed, the structuralist approach argues
thattherewillbeasociallyacceptable`floor'totherealwage.Ifwedefinethe
realwage(w) as
W
w 3:16
P
d
andthusfromequation3.14

1 1
e wu 3:17
m 1r
thenitisclearthatthereisadirectrelationshipbetweentherealexchangerate
andtherealwagerate.Ifthereisalowerlimittotherealwage(w
)thenthereis
anupperlimittotherealexchangerate(e
)

1 1
e w
u 3:18
m 1r
In this case, because the trade gap increases with the rate of growth but
decreases with the real exchange rate, there does indeed exist an external
constraintonthemaximumrateofgrowthachievable.
This approach also forms a major component of the structuralist theory of
inflation where an inflationary spiral is brought about by large devaluations
causing equivalent wage and price rises, which in turn force increases in the
monetizedfiscaldeficitbecauseoftheimpactongovernmentexpenditureand
dollar-denominated debt service (Rodr guez, 1980).
18
This was seen as the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
justificationfor`heterodox'stabilizationprogrammesbasedondirectwageand
pricecontrolsasanalternativetoexchangerateadjustment.Inadditionpublic
action was felt to be required in order to break `bottlenecks' in domestic
supplies of wagegoods particularly foodstuffs and key industrial inputs
(Thorp,1971).
19
EmploymentandISI
The other aspect of income distribution that is central to the ECLA model is
thatofemployment.Fundamentally,itisassumedthatindustrialor`modern
sector'employmentgrowthatarategreaterthanthatoftheworkforceisthe
main means of eliminating poverty by absorbing labour from the backward
low-productivitysectorsofpeasantagriculture,pettyservicesandsoon.Thisin
turn requires that output and the capital stock must grow faster than the
sumofproductivitygrowthpluspopulationgrowth.Inotherwordsthatthere
is no substitution of labour for capital as there is in the neo-classical model
(Rodr guez,1980,pp.71ff.)oratleastthatthemarket-clearingwagewouldbe
toolowtobesociallysustainable.Thecepalinotheoryofstructuralunemploy-
ment is formally derived from a Leontiev-type employment function, where
employment(L)dependsonoutput(Q)andtechnologicallydeterminedlabour
productivityparameters(B,b)
LBQ
b
3:19
this,withacapitaloutputfunctionofthefamiliarform
Q AK
a
3:20
defines the `capital density' (K/L), which rises over time with technical pro-
20
gress. This progress results from the rise of both capital and labour produc-
tivity that is, A rising and B declining over time due to the nature of
technological progress at the labour-scarce centre, which is then transferred
to the periphery through imports of plant and equipment. Substituting
equation (3.20) into equation (3.19) and differentiating with respect to time
yieldsthefollowingrelationshipbetweentherateofgrowthofemployment(l)
andtherateofgrowthofthecapitalstock(k),mediatedbytheratesofchange
inthetwoproductivitycoefficients(a,b):
lbbaabk 3:21
Fromthisitcanbesimplyshownthattherateofaccumulation(k
)requiredto
reduce the unemployment rate or maintain full employment depends on
theseparametersandtherateofgrowthofthelabourforce(n):
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 69
nbba
k
3:22
ab
Inotherwords,anincreaseintherateofchangeofcapitalproductivity(a)will
reduce the rate of accumulation (k
) required to reduce unemployment.

Because the embodiment of new technologies in the capital stock is achieved
through new investment that is, by lowering the average `vintage' of the
capitalstocktheadoptionofnewcapital-intensivetechnologiesisjustifiedas
a `dynamic' solution to the employment problem. This dynamic approach
contrastswiththe`static'solutionbasedontheadoptionof`labour-intensive'
techniquesorastrategybasedonlabour-intensivesectorssuchasagriculture
thatis,adeliberateupwardshiftinB.
21
Eventually,theECLAdevelopeditsmodelofstructuralheterogeneityinthe
labourmarket.Aseparate`backward'sectorhadinitiallybeendefinedascom-
plementarytothatinthe`modern'thatis,K-usingsector,implicitlybased
ontheLewis-typeformulationofsurpluslabourasaresidualheldintherural
peasant or urban informal sectors (Prebisch, 1959). Subsequently, the model
was substantially altered to reflect the notion that increased output in the
modern sector would displace output in the backward sector, and as the
latter is more labour-intensive than the former, then total employment
could actually fall with growth (Rodr guez, 1980, pp. 98ff). The analysis of
wage determination also moved away from the initial concept of surplus
labour in the backward sector determining the supply price of labour for the
modern sector (Prebisch, 1950) towards nominal wage negotiation in
themodernsector,producingthemodelwehavediscussedabove.
Insum,fullemploymentandthusrisingincomesforthoseabsorbedfrom
the backward sector in ECLA theory is determined by the rate of accumula-
tion inthe modern sectorandincorporated technology on the onehandand
demographic dynamics on the other. State-led industrialization based on
importsubstitutionwasthusseenasameansofovercomingstructuralpoverty
inthelongrun.
Thecritiqueoftheimport-substitutionmodel
ThecritiqueofISItheory canbeconvenientlysubdividedbetweenthreelines
ofargument.
22
First,thereisthecritiquefromthestandpointoforthodoxtrade
theory, which argues that ECLA theory neglects the basic principles of com-
parativeadvantage.Second,thereisthecritiquefromthestandpointofsectoral
imbalance, which argues that protection leads to bias against agriculture
andthus to bothlimitedexports andby extension,GDP growth and little
employment creation. Third, there is the argument from the standpoint
of institutional economics, which argues that state intervention and the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
rent-seekingitgeneratesbothleadtolargeinefficiencies.Thefirstandsecond
lines of critique are considered here, and the third in a separate discussion of
politicaleconomybelow(p.80ff).
Neglectofcomparativeadvantage
Most of the initial theoretical critique of the ECLA model was, in effect, a
reassertion of the axioms of the HeckscherOhlinSamuelson (HOS) model
which include a common production-possibility frontier, full employment of
labour and capital, and the absence of monopoly power. In effect, the ECLA
model and Prebisch in particular questioned the validity of these assump-
tions and derived the implications of other axioms on growth and employ-
ment. However, all the critics appear to have targeted the rather general
argumentinPrebisch(1950)andSinger(1950)theso-called`PrebischSinger
thesis' on the terms of trade and did not take into account the full ECLA
analysisofdomesticeconomicstructures.ThefailureofECLAtosetoutitsown
theory rigorously, moreover, meant that critics could freely rely, to consider-
able effect, upon the long tradition of the free-trade critique of protection on
thebasisoftheprincipleofcomparativeadvantage(Balassa,1980).
Flanders, (1964) appears to be the only neo-classical economist to directly
addressthePrebisch (1959)modeldiscussed above,clearlyidentifying itsreli-
anceontheabandonmentofthestandardsmall-countryandfull-employment
axioms. While pointing out that Prebisch is much less àutarchic' than many
criticssuppose,Flandersstressesthatinthemodelthepositiveeffectofprotec-
tiondependsuponconcertedactionbyenoughperipherycountriestobeable
to exercise their relative monopolistic power as exporters. He also reveals
Prebisch'serroneousapplicationofthefactor-priceequalizationtheorem(Pre-
bisch,1950):thefailureofwagestoequalizebetweencentreandperipherydoes
not necessarily indicate a failure to distribute technical progress, because the
return on capital must also be taken into account.
23
Flanders then goes on to
arguethattheECLAassumptionofsurpluslabour,whilereasonableinitself
and undermining a key assumption of the HOS model is not really an
argument for protected industrialization as such, but rather for expansion of
non-traded sectors such as transport infrastructure, education or peasant
agricultureinordertocreatemoreemployment.
Theinterestintradeliberalizationhasfocusedonthebenefitsofòpenness',
in explicit contrast to ISI. Nonetheless, attempts to define openness and by
extension `closedness' in terms of the ratio of external trade to GDP are
meaninglessbecausethesedependontheeconomicstructureoftheeconomy
in general and factor endowments in particular (Edwards, 1993). The absence
ofpricedistortionsthatis,deviationsbetweenlocalandworldpricesisthus
taken to be equivalent to openness and vice versa, whatever the trade ratios,
because what matters is that markets should be contested internationally so
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 71
that local production for the domestic market is efficient. This approach
reflects the earlier notion that any deviation from the optimal import/supply
ratioattheworldpricesetdefinedbythegeneralequilibriummodelshould
bedefinedasimportsubstitution(Desai,1969).However,thisdefinitionseems
tosufferfrom obvioustheoreticalobjections asitdependson theassumption
that world prices are in some sense both stable and welfare optimizing in
practice,quiteapartfromthepracticaldifficultyofestimatingthepricesets.
Sectoralimbalance
The study by Little and associates for the OECD (Little, Scitovsky and Scott,
1970) represents a major critique of ISI and a prospectus for export-oriented
policies, although curiously it shares much common ground with the ECLA.
Thecaseforèncouragingindustry'isadmitted,butitisarguedthatprotection
isaninefficientandineffectivewayofachievingthisaim:
Themainthesesinthisbookarethatindustryhasbeenover-encouragedin
relation to agriculture, and that, although there are arguments for giving
special encouragement to industry, this encouragement could be provided
in forms which would not, as present policies do, discourage exports,
including agricultural exports; which would promote greater efficiency in
theuse of resources; andwhich wouldcreate aless unequal distribution of
incomeandhigherlevelsofemploymentinbothindustryandagriculture.
(1970,p.1)
Itis arguedthatprotectionwillobviously create someindustrialemployment
directly,butthatitwillalsostimulateinternalmigrationandurbanunemploy-
ment.Abiasiscreatedagainstlabour-intensiveexportssuchastextilesdue
tothepresenceofcheapcapitalgoodsimportsandsociallegislationdesigned
to protect the interests of those who are already employed in industry. In
addition, employment-intensive agriculture suffers from a lack of credit,
excessivetaxesandanovervaluedexchangerate.
Sectoral imbalance also arises because of excessive capital goods imports
causedbyfactorpricedistortionsandrisingdemandforintermediateandraw
material imports (even though manufactured consumer goods imports fall).
Placing more emphasis on saving than on earning foreign exchange creates a
bias against exports, which is manifested in shortages of imported inputs,
exchangerateovervaluationandhighdomesticcostlevels.
Little,ScitovskyandScott(1970)calculatethepartofGDPgrowthwhichis
explained by the growth in manufacturing value-added during the chosen
period without taking into account any externalities or dynamic effects.
However, the degree of protection may also have increased during the period
inquestion,andthustheincreasedpricedistortionduringtheperiodshouldbe
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
nettedout.TheresultingestimatesforthethreeLatinAmericaneconomiesin
theirOECDstudy(p.75)indicatethatfortheperiodfrom195052to196466,
only one-third of GDP growth in Argentina and Brazil was contributed by
manufacturing, and one-quarter in Mexico. In other words, manufacturing
was not the major source of growth, even during the `golden age' of ISI.
However, this approach ignores the dynamic effects of and externalities from
industrialization
They suggest that a much better strategy would be to subsidize the use of
labour, provide public training in skills, channel savings towards industry
through development banks, and promote exports vigorously. Interestingly,
notonlyaremostofthesesuggestionstobefoundinearlierECLAdocuments,
but also the Little proposals admit that industrialization is not generated
spontaneouslybythemarketandrequiresspecificgovernmentintervention.
24
AssessingthecritiqueofISItheory
Insum,thecontemporarycritiqueofISItheoryasdistinctfromISIpractice
canbesummedupas:
1 the relative price distortion due to import tariffs in defiance of the prin-
ciples of comparativeadvantage means abias against primary exports, and
thus a lower rate of growth and welfare levels than could otherwise be
achieved;
2 protection of consumer goods and industrial wage support leads to inap-
propriatechoiceoftechnique,andthuslessemploymentandmoreimports
thanwouldobtainatinternationalprices;
3 an overvalued exchange rate to keep industrial costs down leads to discri-
mination against both primary and manufacturing exports, and promotes
non-tradedsectors,aggravatingthebalanceofpaymentsproblem;
4 governments encourage excess public expenditure, industrial capitalists
engage in rent-seeking behaviour and industrial workers force up wages
underISIgeneratingchronicmacroeconomicdisequilibrium.
Assessing these criticisms thus presents a serious methodological problem
because it is not always clear whether what is being criticized is a particular
economic theory, a specific set of policy recommendations, or the historical
experienceofaparticularcountry.
25
Thesecanbecomedangerouslyconflated
intheleyendanegratotheeffectthat:(i)ECLAregardedISIasthebestmodelof
economicdevelopment;(ii)ECLArecommendedISIasapolicytoLatinAmer-
ican governments; (iii) Latin American governments implemented ISI due to
ECLAinfluence.
However,aswehaveseen,proposition(i)isnotstrictlytrueECLAsawISIas
a process forced by external shock, a process which could be rationalized by
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 73
state intervention but which was essentially a second-best solution. Proposi-
tion(ii)istrueinalimitedsense,althoughtheECLAwasclearlyawareearlyon
of the shortcomings of ISI in practice. Above all, proposition (iii) is clearly
historically inaccurate, even if economic doctrine in Latin America was
strongly influenced by structuralism. Indeed, some support for the ECLA
assumptionsonimperfectmarketsmayberegardedasprecursorsofsubsequent
developmentsinmainstreameconomictheory.
ECLA thought undoubtedly did have a considerable influence on govern-
menteconomists andacademicsthroughout theregionduring the1950sand
1960s.Becausestatemanagerswererelativelyautonomousfrompressurefrom
sectoralinterestgroupsandinternationalfinancialinstitutions,theywereopen
to an economic doctrine which appeared to justify state intervention and
would strengthen the nation internationally. Similar economic doctrines
were typical of late industrialization in Europe and Japan (Gerschenkron,
1969).Theindustrializationdoctrinewasparticularlyattractivetoeconomists
and senior administrators in ministries of planning or industry, and to devel-
opmentbanks;eventhoughmanyofficialsbutincentralbanksandministries
of finance continued to maintain their traditional views on the virtues of
soundmoneyandfreetrade.
26
Incontrast,thereislittleevidencethatECLAtheoryasawholewasespoused
by private industrializts either, despite their constant justification of tariff
protection in defence of particular interests (Cardoso, 1971). It may be that
theimpactofstructuralism was`lessevidentthanisoftenthought'(Ocampo,
1990, p. 169) because it could not be made part of effective policy. After the
1970s, structuralism and interventionism began to lose ground due a world-
wideprocessessuchasthecollapseofsocializmandthespreadofmultinational
corporations: `the `ìnternationalisation of capital'' does much to explain
changes in the ideological direction experienced in the developed countries'
(op. cit., p. 170). In addition, Latin American business had become more con-
fident in its ability to survive without protection, although it strongly sup-
portedinterventioninformssuchasmonetarystabilizationasameansofwage
control.
Technology,tradeandindustrialstrategy
Technologicalprogressandindustrialstrategy
The problem of technological progress, and the transfer of technology from
CentretoPeriphery,wasakeyelementoftheECLAtheoryofindustrialization
ingeneralandISIinparticular.Technologicalheterogeneityplayedakeypart
inECLAtheoryoftheinternationaltermsoftradeandemploymentaswellas
of industrialization itself. As we have seen, the ECLA regarded ISI as initially
beingtheresultofaspontaneousindustrializationprocesspromotedbyimport
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
restrictions deriving fromexternal trade shocks andlong term changes in the
world economy. Methodologically the ECLA model is thus better read as a
critique of the notion that market forces alone will generate economic devel-
opmentinLatinAmerican,
27
ratherthananeconomicpolicymodelassuch.
PolicyforECLAwasessentiallyaboutrationalizinganddirectingthisprocess
once it was triggered by external shocks. Once underway, it was necessary to
overcomethetendencyofthemarkettogenerateexternalconstraints,unem-
ployment and inappropriate technology. The logic of the argument is thus as
follows:(i)industryisnecessaryforperipheraldevelopment;(ii)industrializa-
tionontheperipherycanonlyproceedthroughimportsubstitutionbecauseof
problemsofquality,marketsize,tradebarriersandstart-upcost;(iii)thismeans
changing import composition deliberately (Prebisch, 1950). From relatively
early on it was clear that export promotion of both primary and industrial
productswouldbenecessaryforsustainedeconomicgrowth,butthiswould
have to be accompanied by import liberalization on the part of the industria-
lizedcountriestobeatalleffective(Prebisch,1964).
Thecepalinoargumentisthusnotaclassicalinfant-industryargumentbased
on astatictheoryoftheinternationaldivision oflabour wherein afree-trade
frameworkonlyprotectionofthosesectorswhichcanreducetheircostsinthe
long run is justified. The ECLA argument is based on a dynamic concept:
centre/periphery development takes place at different rates of productivity
growth, and as labour is not mobile internationally this condition leads to
the deterioration of the terms of trade and the concentration of the fruits of
progress at the centre.
28
Protection and deliberate industrialization constitute
theonlywaytocounteractthistrend;indeedprotectionshouldbemaintained
untilproductivityandwagedifferentialsareeliminated,ratherthanacceptthe
lowwagelevelsontheperipheryimpliedbylowerproductivitythere(Prebisch,
1959).
TheevidentproblemsofISIinpracticewerecentraltotheECLAanalysisby
the early 1960s, but structural constraints and technological backwardness
ratherthandomesticpricedistortionsorgovernmentinterventionwereseen
as the key problem in moving from the first to the second stage of ISI. More-
over,forlateindustrializerseventhefirststagewouldbeproblematic:ìmport
substitution took place only in countries
29
that had already completed the
initial stage of industrialization, that is, countries that already possessed a
significant nucleus of non-durable consumer goods industries' (Furtado,
1970, p. 90), because they could effectively combine installed capacity with
theexpansionofdomesticdemand.AsISIproceeded,inputcostsrosecreat-
ingmarketsforfurtherISI,but
industrializationinducedbyimportsubstitutionis,strictlyspeaking,aphe-
nomenon of the 1930s and the war period, that is, the period when the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 75
declineinthecapacitytoimportpermittedtheextensiveuseofanindustrial
nucleusformedinanearlierperiod.
(1970,p.92)
Countries undertaking ISI at a later stage
30
did not do so in isolation, but
rather the process was dominated by multinational corporations (MNCs)
leadingtohighimportcoefficientssothatìmportsubstitutingindustrialization
isrelativelyineffectiveasafactorforbringingaboutstructuralchangeincoun-
trieswhereithasbeenunderwayonlysincethefifties'(op.cit.,p.113).
Inthe1960sand1970sthesecondstageofISIwasregardedashamperedby
the intrinsic limitations of the first stage as discussed above. Moreover, the
imbalances caused by ISI made continued industrialization difficult: inflation
caused by the inelastic supply of agricultural products; shortage of skilled
labourduetoinadequateeducationalsystems;repressedfiscalsystemsandan
increasingdebtburden.Advancingtowardssustainedindustrialexportgrowth
washamperedbyhighinputcoststoexportersarisingfrompreviousindustrial
inefficiency,therelativeattractionofthedomesticmarket,andtheunwilling-
nessofmultinationalcorporationstoexporttocountrieswheretheyhadother
plantsortosubstituteforimportsfromtheirownplantsabroad.
In this second stage, moreover, capital goods sectors were regarded as still
dominated by MNC subsidiaries, so there was thought to be no incentive for
local technological development or exports. These affiliates import capital
goodstotheirownspecifications,anddomesticcompaniesneedsimilarequip-
ment to compete. Local investors avoid the capital goods sector as being too
competitive and risky. Although high technological and cost barriers to entry
existinthesesectors,theydonotactasnaturalprotectionbecauseoftheease
of entry for MNCs. In the ECLA approach, which is based on standard indus-
trial economics applied to imperfect markets, there is clear technological seg-
mentationbetweentypesoffirms.Foreignfirmsareoligopolisticleadersinthe
centreandtheperiphery,buttheyaremoreefficientintheformerandhavea
larger markets there, so research and development is not carried out in Latin
America(Fajnzylber,1983).
Insum,ECLAtheorywasnotbasedonasimplenotionofìnfantindustry'or
`domestic market exhaustion', but rather on quite specific propositions about
sectoral and firm-level behaviour. Moreover, the limitations of ISI in Latin
America were recognized as intrinsic to peripheral industrialization, rather
thanbeingduetopolicyerrorsbyparticulargovernments.Indeed,thesecond
stage of industrialization would require greater rather than less government
intervention,butreorientedtowardspromotinglocalresearch,labourskilling,
competitive exports and so on, along the lines of the East Asian experience,
whileresolvingmajorbottlenecksintheenergyandtransportsectorsbypublic
action.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
In the decades since the main structuralist and critical positions on import
substitution discussed above were established, there have been a number of
developments in economic theory which are relevant to the original debate.
Here we shall examine two: the exploration of market failure by `new trade
theory';andtheimplicationsofendogenousgrowththeory.
Newtradetheories
Recent developments in trade theory have moved beyond the familiar
HeckschesOhlinSamuelson (HOS) model to examine in more depth the
implications of trade in manufactures between industrial economies, where
trade is largely intrasectoral rather than intersectoral. Indeed, import
substitution theory had held that as output and trade moves away from nat-
ural-resource-based activities, the HOS assumptions that the production
function is the same in all countries breaks down so that the effectiveness
ofISIcouldbedefinedintermsofinducedproductivitygrowthoraccelerated
rates of capital accumulation rather than static comparative costs (Bruton,
1968).
In so-called `new' trade theories, technological change is a major dynamic
force,endowmentsofcapitalandskillsarebuiltupovertimeastheresultofthe
accumulationofphysicalandhumancapitalwithsubstantialscaleeconomies
andeconomicexternalities,whilepriceformationisbasedonvariousformsof
monopoly power (Grossman and Helpmann, 1991). At first sight, there are
intuitive similarities with some of the later ECLA theories discussed above,
even though the original cepalino models were based on trade between a
primary-exporting periphery and a centre exporting manufactures rather
than the intraindustry trade which is the concern of new trade theory. In
particular, the ECLA propositions as to the effects of barriers to entry and
firmsizeoncompetitionbetweenmultinationalsanddomesticfirmsdiscussed
aboveareclearprecursorsofamoreformaltheorization.
Thecumulative`buildingup'ofsectoralendowmentwhichiscentraltonew
trade theories endogenizes factor productivity in a dynamic model which
justifiesvariousformsofgovernmentinterventiontosupportinvestmentand
skilling.Incontrast,thefree-trade-in-every-periodneo-classicalmodelassumes
that productivity growth is the same in all activities possibly zero or that
productivitygrowthisrandom,oratleastthatproductivitygrowthisgreaterin
exportsthaninimport-competingsectors.Thefirsttwoassumptionsareclearly
implausible,andthethirdisempiricallycontingentsotheHOSmodelcannot
be generally valid. Moreover, modern theories of trade price formation as a
monopoly or oligopoly mark-up on costs, based on technological innova-
tion(newproducts)andbarrierstoentry,showthatmarketsdonotclearonthe
basisofmarginalcostsandfactorproductivity.This,too,isconsistentwiththe
earlierECLAmodel,aswehaveseen.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 77
However,itisnotatallclearthatnewtradetheorieswouldshowthattariff
protection is the best way of supporting industrialization (Krugman, 1989),
unlessthiscouldbeclearlyshowntopromotehigherratesofcapitalformation
orsomehow intrinsically lead togreater R&D activity ormore labour skilling.
ThismightoccurduringISIbutpresumablyastheresultofadditionalgovern-
mentpolicies,andaboveallcouldbeequallyappliedtoanexportpromotion
strategy.
31
Whatnewtradetheoriesdoshowisthatthewelfareeffectsoftrade
liberalization on employment and wages are technically indeterminate, and
thuspossiblynegative.ThestandardHOSassumptionisthatthelabourintens-
ity of import-competing industrial sectors is less than that of primary export
sectors; but where this is not true as is the case in most of Latin America
tradeliberalizationcanreduceemployment(FitzGeraldandPerosino,1999).
32
Byextension,theeffectsofISIonemploymentandincomedistributionmight
well be positive, but not necessarily so the result depending critically upon
therateofinvestment.
Indeed, seen from the perspective of the optimal solution to a dynamic
growth model in the presence of imperfectly competitive or `stagnant' export
markets, the intertemporal equilibrium solution requires managed import
volumes in other words, intertemporal quantity clearing rather than price
clearing in each year (Atkinson, 1969). This approach was widely adopted by
the larger industrializing countries in the 1960s, strongly influenced by the
Indianperspectiveplanningsystemwhichwasexplicitlybasedontheachieve-
mentofhighratesofaccumulation(RajandSen,1961).
Moreover,theconvergenceofgrowthratesacrosscountriesintheneo-class-
ical trade/growth model depends on the crucial assumption that trade occurs
simultaneously withtheinternationaldiffusion ofknowledge.If thisassump-
tionisdropped,andnodiffusionofknowledgetakesplace,thentradeingoods
canleadtoadivergenceofgrowthrates.Althoughtradeinintermediateinputs
or by multinational corporations which facilitate the diffusion of knowledge
canaffectthisresult,theoutcomeisstillindeterminate(Feenstra,1996).Again,
this modern theoretical development would seem to be congruent with the
intuitiveinsightintheoriginalECLAmodel.
Indeed,LonderoandTeitel(1996)findthatforanumberofLatinAmerican
countries manufactures originally developed for the domestic market during
theISIperiodweresubsequentlyexportedwithconsistentgrowthrates,attrib-
uting this to the external effects of industrialization, including technological
learning, which compensate for the low level of competition. This result is
inferredfromthefactthattherespectivebranch-levelinput-outputcoefficients
for manufactured exports indicate that the products are relatively capital or
skill-intensive, rather than reflecting `natural' comparative advantage from
unskilled labour or natural-resource endowments. They conclude that
`these results could be construed as implying that manufacturing production
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
incentives are successful when they promote activities relatively intensive in
thefactorswhichtheindustrializationprocesstendstomakemoreabundant'
(ibid.,p.598).
Endogenousgrowth theory
Productivity growth is clearly central to ISI theory, but ECLA found this diffi-
cult to formalize. Implicitly, the structuralist model has the following general
form:
Y AK
p
K
g
subjectto M Y
t
X
t
T
t
3:23
whereoutput(Y)isexplainedbychangesinthecapitalstockofboththepublic
(K
g
) andprivate (K
p
) sectors, with apositive effect of the former on the latter.
Anexternalconstraintlinkingimports(M)toexports(X)themselveslargelya
functionofworldtradevolumesandthecapitalstockitselfandthetermsof
trade (T). The scale factor (A) then represents the effect of policy regimes,
economicinstitutionsandsimilarenvironmentalfactors,aswellastheprocess
by which technology is embodied into investment through changes in total
factorproductivity.
33
In contrast, the standard neo-classical growth model has the familiar form
(Dennison,1967):
Y AK
a
L
b
3:24
whereoutputdependsoncapital(K)andlabour(L)supplyalone,withdeclin-
ingreturnstoscalewhichimplyaneventualexhaustionofthegrowthprocess.
Astrongpropositionofneo-classicalcriticsofstructuralismisthatunderthe
ISIregime,totalfactorproductivity(TFP)growthisbothlowanddecliningfor
two main reasons: first, that capital intensity is too great, leading to lower
growthforagivencapitalandlabourstockbecauseprotectionincreasesreturns
beyondtheoptimalpoint;andsecond,thatcompetitionislessthaninanopen
economy,sofirmsseelittleneedtoembodynewtechniquesinordertomain-
tainmarketshare.
However, modern endogenous growth theories set out to explain the long-
termpersistenceofgrowth,whichshouldnothappenatallintheneo-classical
model(BarroandSala-i-Martin,1995).Researchactivitywithinfirmsorina
separate`researchsector'takestheformofnewproducts,productqualityor
processcosts.Theresearchactivitycancontributetothestockofpublicknow-
ledge capital, which reduces the cost of future research to all firms. Similar
arguments can be applied to the externalities arising from the skilling of the
labour force or the provision of common infrastructure. What makes such
theories èndogenous' is the extent to which they model the way in which
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 79
themarketstructureitselfprovidesincentivesforeconomicagentstoinvestin
researchorskillsorinfrastructurewhichleadstogrowth,investmentwhich
is rewarded by patents(orother formsof rent). Themonopoly power granted
arises from a barrier to entry by competitors, which in turn generates the
monopoly rent. In other words, it is precisely forms of `market failure'
34
that
cangeneratepersistentgrowth:monopolisticcompetitionand/orexternalities
ontheonehand,andnon-decliningreturnstobroadcapitalphysical, orga-
nisational,financial,human,infrastructureandsoonontheother.
In other words, the optimal growth process does not derive from purely
competitive markets, and can be affected by government intervention. This
doesnotofcourse implythattariffprotectiontomanufacturing industrywill
necessarily engender an endogenous growth process by creating monopoly
rents because these may not be invested in research. Further, even if it is
reasonable to argue that capital investment has constant or even increasing
returnsitdoesnotnecessarilyfollowthatpublicinvestmentwillgeneratemore
externalities than are lost by its own internal inefficiency. Nonetheless, these
modern views do seem remarkably consistent with the earlier albeit
unformalizedviewsoftheECLA.
Bruton (1967) is an early attempt to estimate TFP growth,
35
comparing
Argentina, Brazil, Chile and Colombia for the 194064 period, and then con-
trasting these results with some industrial countries. He finds very low TFP
growth in LA `contrary to the frequently encountered notion that less de-
veloped countries with a somewhat primitive technological base may reap
large windfalls by exploiting recently developed knowledge' (p. 1102) and
thatcapitaloutputratiosaresimilarinLatinAmericaandindustrialcountries.
Hesuggeststhisisbecausethetechnologytransferredisdesignedforindustrial
economiesandthatmonopolypositionsofISIindustriesmakesthisprofitable;
solowTFPgrowthintheregioncanbeattributedtothefactthat`whateverthe
opportunities for exploitation of technology existed, there did not exist a
capacity to adapt and modify this technology to use it effectively in Latin
America'(p.1105).Moreover,hefindsthatTFPgrowthisnotpositivelycorrel-
ated to factor input growth, unlike in industrial countries where there is a
coefficient of near unity implying that there is not a strong process of learn-
ing-by-doingandthattheeffectsarepurelyadditive.
The Chenery (1986) estimates part of a much larger World Bank study of
developing countries as a whole of TFP growth show a somewhat different
picture. TFP growth still falls in Argentina and Brazil during the ISI
period, but actually rises in Chile, Colombia and Mexico; it is much higher
in Korea throughout, and actually rises in the second stage of ISI. Chenery
tests for the èmbodiment coefficient' in TFP for the group of LDCs as a
whole,
36
andalsofindsathatwhilethisissignificantforindustrialcountries,
itis not fordeveloping ones. Inother words, firms in industrialized countries
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
do seem to benefit from the technological externalities arising from high
national investment rates, but developing countries experience neither this
nor curiously enough the convergence (`catching up') benefits of techno-
logytransfer.
TheserathernegativefindingsarenotinconsistentwiththeECLAmodel.If
weacceptaroughestimateforobservedTFPgrowthof1percentperannumin
LatinAmericaduringthe195070period,andcorrectfortheaveragedeteriora-
tionofthetermsoftradeof2percentperannum(fromTable3.1below),then
theformulationinequation(3.23)wouldimplyatruetotalfactorproductivity
growth rate of 3 per cent per annum, which is quite respectable by interna-
tionalstandards.
37
Nonetheless,Solimano(1996)mayberighttoarguethat
thepost-wargrowthexperienceofLatinAmerica,sayfromthe1940stothe
onsetofthedebtcrisisintheearly1980s,reliedstronglyonfactoraccumu-
lation,withadecliningcontributionoftotalfactorproductivitygrowthover
time(asortofèxtensive'growthpattern).
Heattributesthisrecord,however,tothe`reducedcontributionoftotalfactor
productivitygrowthassociatedwithapolicyregimebasedonimportsubstitu-
tionandstatedirigismthatbroughtaboutadistortivestructureofincentivesin
theeconomy'(p.2).Tothisissuewenowturn.
ThepoliticaleconomyofISI
This chapter is concerned with the economic theory of import substitution,
and not with the historical experience of ISI as such. Nonetheless, the two
cannot be easily separated, if only because economic ideas evolve in
responsetoreal-worldproblems.Inconsequence,thereisaconsiderable`polit-
ical economy' content implicit to the economics of ISI, as we have seen, and
thustheoryisrootedinaparticularinterpretationofhistory.AsaUNinstitu-
tion,theECLAitselfwasnaturallycircumspectonthepoliticaleconomyissues;
but Prebisch, (1950) clearly regarded landlords as opponents of industrializa-
tion, and industrialists as in need of government support. Other structuralist
authors saw foreign firms as key actors in both the initiation and eventual
frustrationofISI.
Thelackofindustrialists
In his classic essay on the political economy of import substitution in Latin
America,Hirschmanalreadydetects
38
`considerabledisenchantmentwiththis
solutionof[sic]thedevelopmentproblem'(Hirschman,1968).Hearguesthatit
took from 1929 to 1949 `before the end of export-propelled growth became
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 81
official Latin American doctrine'. Then came desarrollo hacia adentro `which
flourished briefly in both theory and practice during the fifties and was
pronounced either dead or dud in the sixties'. He suggests that there were
`four motivations for ISI balance of payments difficulties, wars, gradual
growth of income, and deliberate development policy' and that in Latin
America `the ISI process is far smoother, less disruptive but also less learning-
intensivethanhasbeenthecaseforindustrializationinEurope,NorthAmerica
andJapan'.
Hirschman argues that thecentral issueis thatof the `sources of entrepren-
eurship'
39
whichdriveindustrialization.Onlywhenforeignexchangeisscarce
will local importers who are likely to be of immigrant origin and foreign
firms exporting into the economy turn to import substitution; if foreign
exchange income is expanding, industrialization will be confined to indigen-
ousentrepreneurs.ForeignersandminoritiesengagedinISIhavelittlepowerto
control economic policy that is, industrialists do not control the state.
Although the initial stage of ISI did lead to defensive monopolistic positions
being built up, the main problem was that it `caused delusions of success
among governments'. Subsequently, ISI gets stuck, as ìmport substituting
industry is affected by seemingly congenital inability to move into export
markets', there is inadequate generation of employment and a new depend-
encyonbranchesofmultinationalcorporationsemerges.Thepriorexistenceof
early ISI branches creates barriers to the establishment of suppliers and main-
tainslowdutiesonitsinputsunlessthenewbranchesaresetupbymembers
ofthesamefamily
40
leadingtotheconclusionthatìtmayinfactbeprefer-
able for the governments of late late-industrializing countries to be run by
tecnicos,bygroupsofplanner-technicians,ratherthanbythenewindustrialists
themselves'.
41
HirschmandisagreeswiththenotionthatISIproductsaretoocostlytoexport,
becausetheindustrialcountriesmanagetodosoandthehomemarketshould
developexperience.Ratherìtappearsthatthemuchadvertisednon-competi-
tiveness of Latin American industry may be rooted in the failure to modify
institutions rather than in any inability to bring down real costs', which he
attributes to the `distribution of power in Latin American societies', where
industrialistsneedpolicycertaintytoundertakemassiveinvestmentsrequired,
butìnnocountryofthecontinentdotheindustrialistsfeelsecurelyincontrol
ofvitaleconomicpoliciesaffectingthem'.Indeed:
policymakerspositivelycultivateunpredictabilityanddistancefrominter-
estgroups;atthesametimetheyarehighlymanipulative.Changesinfiscal,
monetary and foreign exchange policies are therefore frequent while com-
munication about these changes with the affected interest groups is infre-
quent. These are the socio-political traits which account, perhaps more
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
fundamentallythanthecost-pricestructureofnewindustries,fortheirpoor
exportperformance.
(1982,p.29)
In contrast, most Latin American authors do not attribute the `weakness of
the domestic bourgeoisie' to an inconsistent state or filial social mobility, but
rathertoitsdependentrelationshipwithexternalcapital(Cardoso,1967).They
alsosuggestthatthehighlevelsofprofitsinnon-tradedsectorssuchasbanking
and construction meant that no industrial investment would take place at all
withoutsimilarlevelsofprofitabilitywhichcouldonlybeprovidedbyprotec-
tion. External dependence also explains why such high and indiscriminate
levels of protection were afforded to industrial systems led by international
firms for whom the ìnfant industry' criterion could not logically be invoked
(Fajnzylber,1983).
Governmentaldistortion
Themostrecentlineofcritiqueofstate-ledindustrializationinLatinAmericais
directed not only at the effects of tariffs as such but also to any government
intervention in the attempt to promote industry beyond that induced
naturally by market forces. This theory of `state failure' argues that the
government is not capable of making better decisions than the private sector
in the sense of identifying externalities accurately, overcoming coordination
problems or foreseeing future comparative advantage. This is partly an
information problem, but also one of the government's own objective func-
tion, which is to expand its own expenditure and raise the income of its
members or at best to win the next election. This will not lead to welfare-
maximizing policy implementation, and this systemic distortion is aided and
abetted by the rent-seeking behaviour of those firms which benefit from pro-
tection. This state failure is seen as an intrinsic feature of ISI (Krueger, 1995;
WorldBank,1987).
In this tradition, Solimano (1996) identifies two distinct policy regimes:
`state dirigism-cum-import substitution, the development model of Latin
America between the 1940s and the 1980s, and market-based development'.
This first regime is characterized by capital-intensive development, where
publicinvestmentusedtoboostgrowthandavoidthepotentialcoordination
failures of decentralized markets, while private investment is stimulated
by subsidized credit, tax rebates and preferential exchange rates. At the
microeconomic level, this leads to inefficient allocation decisions and rent-
seeking where governments tend to install or continue protection in order
tobenefitparticularindustrialpressuregroups.Thewelfarecostofthisprotec-
tion is borne by consumers as a whole or the unemployed, neither of whom
constituteanorganizedpoliticallobby(Krueger,1995).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 83
This argument reaches its strongest form in Edwards (1995), where rent-
seekingbecomesacentralforceintechnologychoices:
The inward looking strategy generated rampant rent-seeking opportunities
and resulted in the use of highly capital-intensive techniques, which ham-
peredthecreationofemploymentthroughouttheregion.
(Edwards,1995,p.44)
whichinturnincreasedthecostofindustrializationandeventuallyrenderedit
unsustainable:
Althoughseveraldecadesofprotectionistpoliciesaccomplishedthegoalof
creating an industrial sector in Latin America, they did so at high cost.
Exports were generally discouraged, the exchange rate became overvalued,
employmentcreation lagged, andmassive amounts of resources, including
skilled human talent, were withdrawn from the productive sphere and
devoted to lobbying for an ever-favourable treatment of unproductive sec-
torsoftheeconomy.
(1995.,p.119)
At the macroeconomic level, it is argued that chronic inflation and
recurrent balance of payments crisis were the product of excess domestic
demand which derived directly from ISI. Popular pressure on governments to
increasepublicexpenditure,toraisewages,toreducelabourmarketflexibility
andtoprotectvulnerableindustrialsectorscombinedwithdistributivecon-
flictsoverincomesharesaretherootcauseofthesemacroeconomicdisequi-
libriainthisinterpretation(DornbuschandEdwards,1991).
Thestateandindustrialaccumulation
This chapter has attempted to reexamine the structuralist theory of import
substitution as the basis for state-led industrial accumulation. This theory
clearlyhasanintrinsicimportanceasasignificantinfluenceonthepatternof
industrializationthatdidactuallyemergeinLatinAmericaduringthepostwar
decades. We have seen how many of the neo-classical and political-economy
critiques of this theory were anticipated by cepalino authors themselves, or
based on misreading of the relevant texts. However, ECLA theory did not
establish a coherent economic argument for state intervention with the
corresponding fiscal doctrine which could have underpinned a long-term
policyenvironmentforprivateinvestment,exportexpansion andlabourskil-
ling. It is evident that a clear strategic concept did form an important part of
the rapid industrialization process in `late developers' such as Germany and
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Japan(Gerschenkron,1969),aswellasfortheEastAsiandynamiceconomiesin
themorerecentpast.
Thetheoreticalargumentsputforwardinthebodyofthischapterimplythat
thecentralissueinevaluatingISIisdynamicratherthanstaticefficiency.The
outcomeofprotectionandgovernmentinterventionshouldthusbemeasured
in terms of high rates of productive investment in industry. If state interven-
tion could ensure that rapid industrial accumulation by private firms would
take place, it did seem reasonable to expect that eventually scale economies
andlearning-by-doingwouldmakethemcapableofcompetinginternationally.
Indeed, the levels of tariff protection and price distortion by 1970 were com-
parablewiththoseinEastAsiaatthattime(Taylor,1996),whenthelimitations
of ISI were recognized by ECLA and the shift towards export promotion was
alreadyunderwayintheregion.
Theproblem is how toexplain why thegeneralized shifttowards industrial
exports as an engine for growth did not take place until the debt-related
dismantling of import protection in the late 1980s (Bulmer-Thomas, 1994).
Was it due to a combination of misguided policy and vested interests, or to
external shocks and crisis management? There is little in ECLA thought or
structuralist theory to explain this delay, which probably prevented Latin
Americanindustryfromgainingacompetitivefootholdinworldmarketsand
raising the rate of productive investment to levels capable of adsorbing the
labourforceandreducingpovertyonasustainablebasis.
A first problem in evaluating the theory of ISI is thus that of periodization.
The notion of a single uniform model being applied by Latin American
governmentsinresponsetoECLAdoctrineisquiteerroneous:periodsofliber-
alization in one country coincided with nationalizations in others; real
exchange rates and tariff levels rose and fell. Moreover, `by the mid-1970s
short-term factors began to dominate...to the point where it is no longer
relevant to speak of ``development strategy'' of any sort' (Rodr guez, 1980,
p.175).
An approach consistent with the original ECLA theory would be to look at
theperiodsoverwhichtheexternalconstrainteffectivelyboundfortheregion
as a whole Table 3.1 suggests that only in 1965 did the purchasing power of
exports returntoits 1928levelon asustainedbasis,andfromthenon itrises
steadily through 200 in the mid-1970s and 300 in the mid-1980s. When
population growth is taken into account, the real value of export income per
head only returned to its pre-Second World War level in the mid-1970s. The
import quantum recovers earlier, but this is due to foreign borrowing rather
than exports. We can assert that the external constraint on the region as a
whole was lifted de facto by 1975; when import substitution and industrial
protection ceased to be the main source of growth and export expansion
couldbeexpectedtotakeoverastheengineofgrowth.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table3.1 Theexternalconstraint accordingtoECLA
E. V.K.FitzGerald 85
Merchandise Export Purchasing Import
termsoftrade volume powerofexports volume
1928 100 100 100 100
1930 67 88 59 76
1935 66 106 70 56
1940 67 91 61 59
1945 69 111 77 66
1950 94 115 108 105
1955 62 131 82 129
1960 53 166 89 145
1965 49 211 104 150
1970 53 254 135 147
Source: ECLACdatabase,citedinGround(1988).
Theavailableempiricalstudiesallshowthatthemaindeterminantofgrowth
duringthepre-1975periodinLatinAmericawastherateofcapitalaccumula-
tion.Themostcomprehensivegrowth-accountingexerciseforLatinAmericain
thepostwarperiod(Hofman,1999)showsthatduringthe195073periodjoint
factor productivity grew reasonably rapidly; in fact all countries show more
rapid productivity growth in 195073 than in the 198094 period of liberal-
ization. However, once changes in the quality of labour and capital are taken
into account, TFP growth is reduced to quite low levels, implying that factor
accumulationisindeedthemainsourceofgrowthintheregion.
42
Moreover,as
Table3.2shows,thecapitaloutputratiorisesovertimeinallcasesexceptfor
Colombia which reflects the pattern of growth which ECLA theory anticip-
ated.Thestrongesttrendisthatincapitalperworker,whichsupportsboththe
structuralisthypothesisthatinvestmentisthemotorofgrowthandthestruc-
turalist explanation of unemployment and poverty. However, the question of
whyinvestmentratesremainedrelativelylowremainsunexplaineddespitethe
enormousefforttoraisethem.
The investment rate has been historically low in Latin America (Hofman,
1992,andalsoseeTable3.3)butwithoutevidenceonpublicandprivateinvest-
mentontheonehand,andbetweentradeandnon-tradedinvestmentonthe
other,itisdifficulttocometoafirmcausalconclusion.
43
Therateofaggregate
investmentinArgentinaandBrazilactuallyfellintheearlyISIperiod194059
relativetothepreviousdecadesofexogenousshock,althoughitthenrosein
thelaterISIperiod(196079).InChileandMexicotherateroseinearlyISIand
remainedatthatlevel;whileinColombiatheeffectofISIoninvestmentdoes
notseemtohavebeenverysignificant.However,inallfivecasestheinvestment
rate appears to have fallen in the subsequent liberalization period (198089),
exceptinChilepossiblebecausestructuraladjustmentstartedearlierthere.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table3.2 Aggregatecapital,labourandoutput
ratiosinLatinAmerica,195094
Output/ Capital/ Output/
capital labour labour
Argentina
1950 1.03 3.3 3.4
1973 0.86 7.2 6.1
1994 0.72 10.4 7.4
Brazil
1950 1.35 1.1 1.5
1973 0.79 4.6 3.6
1994 0.43 10.8 4.7
Chile
1950 0.79 3.4 2.7
1973 0.65 8.0 5.3
1994 0.68 9.5 6.5
Colombia
1950 0.52 3.3 1.7
1973 0.69 5.0 3.4
1994 0.63 7.8 4.9
Mexico
1950 1.05 2.5 2.7
1973 0.81 8.5 6.9
1994 0.60 12.9 7.6
Units: OutputandcapitalinUS$millionsat1980ICP
prices;labourinhoursworked.
Source: Hofman(1999).
Table3.3 Ratesofaccumulation(GFCF/GDP)
Argentina Brazil Chile Colombia Mexico
192029 19.2 17.7 18.3 20.0 10.5
193039 18.5 10.3 23.0 15.4 8.3
194049 14.3 11.9 20.2 14.8 11.7
195059 15.3 18.3 18.5 14.9 15.5
196069 22.0 18.1 21.3 16.0 19.0
197079 23.5 21.7 18.7 16.4 22.2
198089 20.0 20.1 19.8 16.0 21.0
Source: CalculatedfromdatainHofman(1992).
Standard microeconomic theory now recognizes that uncertainty rather
than the difference between expected profit and interest rates is the major
disincentive for firms to undertake long-term fixed investments (Dixit and
Pindyck, 1994). Modern macroeconomic theory also stresses the importance
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 87
ofstabledemandgrowthinexplainingaggregateprivateinvestment,aswellas
theprovisionofoverheadcapitalintheformofskillsandinfrastructure(Fitz-
Gerald, Jansen and Vos, 1994). Tariff protection and public investment might
be regarded as a means of providing this market stability. For the subsequent
switch to export-led industrial growth, business confidence in a stable eco-
nomic regime and `rules of the game' are essential too as the East Asian
experience indicates. Latin America, in contrast, experienced a succession of
external shocks and frequent shifts in economic policy which undoubtedly
underminedtheeffectivenessoftheindustrializationstrategy.
In retrospect, it is clear that one of the root causes of the severity of these
externalshocksandthefrequentchangesineconomicpolicyduringthepost-
war period was fiscal in nature. Persistent public sector deficits generated
inflationary pressures and excessive debt levels, which subsequently made
macroeconomic management extremely difficult and inevitably increased
business uncertainty. What is more, the weakness of the state arising from
the chronic fiscal crisis probably made it impossible for the Latin American
statetoforcetheprivatesectorintoexportcompetitivenessalongthelinesof
theEastAsia
44
orindeedthoseofGermanyandJapan.
Contemporarystructuralisttheorysuggestedthatfiscalimbalancesweretoa
considerable extent endogenous due to historically weak tax systems, the
expenditure requirements of economic development, and the cost of debt
service (ECLA, 1973). In contrast, the subsequent critique attributes growing
fiscal deficits during this period to populist pressures for greater welfare
expenditure and unrealistic levels of wages and employment in combination
withthedesireofbureaucratstoincreasethesizeofthestate(Edwards,1995).
The key question is to what extent this `fiscal crisis' was intrinsic to the ISI
model.One-quarterofcurrentgovernmentexpenditureinLatinAmericadur-
ing the 1960s was devoted to production subsidies and services for private
enterprise; the remainder being divided almost equally between defence and
security, health and education, and general administration (ECLA, 1973). As
health and education provision was mainly oriented towards modern sector
employees, and the armed forces towards securing property rights; it seems
reasonable to suggest that a considerable part of current expenditure in Latin
Americacouldbeattributedtothepromotionofprivateinvestment.Thelevel
ofpublicinvestmentintheISIperiodreflectedachangeincompositionaway
fromirrigationandruralroads,towardsurbaninfrastructureandheavyindus-
try (Baer, 1974) which was also designed to promote private investment.
45
ComparisonofTables3.3and3.4indicatesthatpublicinvestmentrangedfrom
a third to a half of total capital formation with a clear tendency for this
proportiontorisebetween1950and1975.Inconsequence,theaveragelevelof
public expenditure in Latin America rose by over one-half, from about 17 per
cent of GDP in 1950 to some 28 per cent in 1975, and it seems reasonable to
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
8
8
Table3.4 ThefiscalstructureinLatinAmerica
Argentina Brazil Chile Colombia Mexico LA
1950 1970 1950 1970 1950 1970 1950 1970 1950 1970 1970
Publicsector(%ofGDP)
Generalgovt.expend. 14.7 17.3 13.3 24.3 13.7 25.6 11.5 10.6 6.5 6.0 18.4
Publicfixedinvestment 7.8 7.9 4.2 9.0 3.3 9.0 1.9 6.7 6.1 7.5 7.3
Totalpublicexpenditure
a
22.5 25.2 17.5 33.3 17.0 34.6 13.4 17.3 12.6 13.5 25.7
FiscalIncome 17.0 15.0 16.0 27.0 16.7 21.8 12.0 13.4 7.5 7.1 17.2
Sharesoffiscalincome(%)
Incomeandpropertytax 37 32 26 28 40 34 28 35 46 51
Consumptiontaxes 57 52 63 65 15 46 47 47 36 36
Importandexportduty 6 16 11 7 45 30 25 18 18 13
Fiscaldeficit(%ofGDP)
Generalgovt.saving
b
2.3 2.3 2.7 2.7 3.0 3.6 0.5 2.8 1.0 1.1 1.2
Publicsectorborrowing
c
5.5 10.2 1.5 6.3 0.3 12.8 1.4 3.9 5.1 6.4 8.5
a
generalgovernmentexpenditurepluspublicfixedinvestment;
b
generalgovernmentexpenditurelessfiscalincome;
generalgovernmentsavinglesspublicinvestment.
Source: FitzGerald(1978)basedonECLAdata.
c
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 89
conclude that this shift was closely related to the state-led industrialization
strategy of which import substitution formed a part (FitzGerald, 1978). This
patternseemstohavebeencommontootherlateindustrializationprojects.
Fiscal revenue also increased as the Latin American states expanded, but by
considerably less than expenditure (see Table 3.4). Revenue rose from appro-
ximately12percentofGDPin1950to20percentin1975(FitzGerald,1978).
Thisincreaseof8percentagepointswasconsiderablylessthanthe11percent-
agepointincreaseinpublicexpenditure.Therewasasharpfallinthepropor-
tionoftaxrevenueraisedfrompropertyandincomes,particularlyinArgentina
and Chile, dueto the various investment incentives established aspart of the
industrialization drive. The share of revenue from duties on exports and
imports also appears to have declined in most cases due in part to export
promotionandalsotothefactthattariffrateswerehighestonthosecommod-
itycategorieswhichwerenotasaresultimported.Inconsequence,theshareof
taxationimposedonconsumption,theleastprogressivesourceofrevenue,rose
sharply throughout the region. ECLA estimates cited by Furtado (1970, p. 66)
and reproduced in Table 3.5 suggest that in Latin America as a whole, fiscal
incidence showed some progression between the rural and urban poor, but
werelowerfortheurbanmiddleclassandreachedlittlemorethanone-fifthof
disposableincomeforpropertyownersinthemid-1960s.
46
Hearguesthatthe
resistancetofurthertaxpressurecameoverwhelminglyfromtheupper-income
groupseventhoughthedesignofpublicexpenditureinthisperiodwassuchas
tosupportprivateinvestment.
Theinevitableresultofthesetwodivergenttrendswasasteadyexpansionof
thestructuralfiscaldeficitthroughouttheregion,despitefrequentattemptsat
macroeconomic stabilization. As capital markets were limited or non-existent
in the region, only two outcomes were possible either monetization of the
deficit with consequent inflationary pressures, or public borrowing abroad
with consequent sovereign debt accumulation.
47
Inflationary pressures were
Table3.5 FiscalincidenceinLatinAmerica
a
Incomegroup I II III IV Total
Shareofpopulation 50% 45% 3% 2% 100%
Averageincome(US$in1960) $120 $400 $1750 $3500 $365
Taxationandsocialsecurityas:
Percentofdisposableincome 13.0 20.0 16.5 21.0 18.4
PercentofGDP 1.7 8.4 1.9 3.2 15.3
Source: Furtado(1970,p.66)onthebasisofECLAdata.
a
Furtado,loccit.,describesthesefourincomegroups`themassofruralworkers'(I),the
`massofurbanworkers'(II),`highersalaryearnersandsmallentrepreneurs'(III),and
`largelandownersandentrepreneurs'(IV)respectively.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
counteredbyrepeatedattemptsatmacroeconomicstabilization,whichtended
to depress private investment through the accelerator effect. The build-up of
debt which rose tenfold from US$3 billion for the five largest economies in
1955toUS$31billionin1975ledinexorablytowardsthesubsequentvulner-
ability to world interest rates, and had strongly negative effects for private
investment in terms of both further fiscal deficits and increased macroeco-
nomicuncertainty.
Insum,itseemsthatthemainshortcomingoftheISImodelinLatinAmerica
was essentially fiscal in nature, leading to low rates of capital accumulation
ratherthanacrisisofgrowthorproductivityassuch.Atthelevelofeconomic
doctrine, this lack of positive fiscal theory was a central weakness in the
structuralist model of industrialization. ECLA had a clear view of the need to
force the pace of investment but not of how to finance it without creating
macroeconomicinstability,whichinturnreducestherateofaccumulation.
Inconclusion,themainpointmadeinthischapterhasbeenthatthestruc-
turalisttheoryofimportsubstitution,despiteitsfaults,wasmuchmorecomplex
and profound than its critics were prepared to gave it credit for. The cepalinos
understood that import substitution started as an endogenous response to an
exogenous shock, generating a process which could be rationalized under the
continued restrictions on export growth in the postwar decades. As we have
seen, ECLA theory can be read as an early attempt at adapting mainstream
economics to conditions of systemic market failure themes that have been
taken up subsequently in new theories of institutional economics, industrial
tradeandendogenousgrowth.Finally,thetheoryofimport-substitutingindus-
trialization undoubtedly influenced economic doctrine in the region and
beyond,butbythemid-1970sthisdoctrinewasclearlydueforreplacementby
anewapproachtothereintegrationofLatinAmericatoarapidlygrowingworld
economy. Unfortunately, economic policy since then appears to have been
conditionedbyreactiveresponsetoanotherexogenousshockthedebtcrisis
ratherthananewtheoryofLatinAmericanindustrialization.
Notes
1 Particularly the collapse of payments and trading systems, declining export prices
anddemand,andcapitaltransferreversals.
2 Throughout this chapter the source for ECLA economic theory is Rodr guez (1980)
unlessotherwiseindicated.ForabriefaccountofECLAasaninstitutionalactor,see
FitzGerald(1993).Asweareherereferringtothedevelopmentofcepalinothoughtin
thethirdquarterofthetwentiethcentury,wehavereferredthroughoutthischapter
to ECLA (rather than ECLAC, the name adopted in 1984 to reflect the growing
numberofmemberstatesfromthenon-Spanish-speakingCaribbean).
3 TheinfluenceofotherstrandsofdevelopmenteconomicsonECLAthought,andvice
versa, is a largely unexplored subject but see FitzGerald (1994) for a preliminary
assessment.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 91
4 For a definition of structuralism as a general theory of institutional economics
appliedtosemi-industrializedcountries,seeFitzGeraldandVos(1989,ch.1).
5 Atypicalrecentstatementofthiskindis:`[ThedISI]policiesdevelopedinresponse
to external shock between 1929 and 1950) did not come as the result of new
economic theory, but rather the force of circumstance in extreme economic crisis.
This was to change, however: starting in the 1950s the unorthodox policy experi-
mentswhichhadsoftenedtheblowofthedepressionandthewarhadbeenclothed
withthetrappingsofaneweconomictheory,propoundedbynowasa``model''of
economicdevelopmentbynewschoolsofthought(theDependencyschoolandthe
Structuralists), and spread with heightened respectability to other countries within
the region and beyond' (Taylor, 1996, p. 6). Such is the perceived influence of
peripheraleconomicdoctrineviewedfromtheCentre.
6 Sincethemid-1980sanewschoolof`neo-structuralist'and`neo-Keynesian'econom-
istsofconsiderabletechnicalmerithasemergedinLatinAmerica,butunfortunately
theirworkfallsoutsidethescopeofthischapter.
7 AcentralthemeinthefirsttwoECLAInformesofwhichonlythesecondwaswritten
by Prebisch himself. The critical approach was not quite as unusual for United
Nations agencies as it might seem today: the League of Nations had established a
similartradition,takenupbyUNinNewYorkinitspost-SecondWorldWarEconomic
SurveysonwhichKalecki,LewisandSingerallworked.
8 Curiously,thismodelappearsnevertohavebeenpublishedinSpanishatleastitis
notlistedintheofficialECLAbibliography.
9 Infact,itwouldbemorerealistictoallowonlynewinvestmenttobereallocatedin
responsetorelativeprices,whichwouldbringaboutasignificantadjustmentlagand
thusdynamicbehaviour(FitzGeraldandPerosino,1999).
10 Orsocialincomeifwehadincludedwageswithemploymentdeterminedbyoutput
andrealwagesbythesupplypriceoflabourfromtheìnformal'sectoralongLewis-
typelines.
11 SeeFitzGerald(1994)foradiscussionoftheoriginsanduseoftheconceptof`centre'
and`periphery'bytheECLA.
12 Itisnotmyintentionheretoanalysethedebatesonthetermsoftradeandunequal
exchange,onwhichseeFloto(1989).Itshouldbenoted,however,thattheoriginal
cepalinothesis(ECLA,1949)wasbasedonasymmetricpriceformation:theperiphery
basing agricultural export prices onlow costs from excesslabour supplyin the way
explainedaboveandthecentreapplyingamonopolisticmark-uptoindustrialexports
to benefit its own firms and labour force. This `neo-Ricardian' model was soon
replaced by the version based on differential income elasticities set out in Prebisch
(1950) and Singer (1950). The initial neo-classical critique (for example, by Viner,
onwhichseeFitzGerald,1994)oftheECLAmodelwasalmostexclusivelydirectedat
this second terms-of-trade argument (rather than ISI as such), from the point of
view of the orthodox trade model (of classically Ricardian origin) which holds
that primary sector prices should rise relative to manufactures over time, due to
increasing industrial productivity on the one hand and natural resource scarcity
on the other. Recent empirical research seems to indicate that not only is the
original PrebischSinger thesis valid for the primary commodity terms of trade in
thelongrun(Sapsfordetal.,1992),butalsomuchthesameproblemseemstooccur
withlabour-intensivemanufacturedexportsfromdevelopingcountriesaswell(Sarkar
etal.,1991).
13 Though susceptible to international economic action through improved market
access to and coordinated growth in the G7 economies which became a central
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
14.
14.
themeinECLAreportsandsubsequentlyinthoseoftheUNCTAD,towhichPrebisch
movedasSecretary-Generalin1964.
14 In consequence, the standard ECLA measure of effective import substitution
(Villareal,1988)isthechangeintheratio(m)ofrealimports(M)tototalsupply(S)
in that product category, so import substitution is measured between two periods
(1,2) as
M
2
M
1
m
2
m
1

S
2
S
1
The aggregation of sectoral measures can have a misleading effect if the sectoral
balance is changing, or if relative prices change due to devaluation or productivity
trends. Nonetheless this is the measure most used in practice. Work on long-term
patterns of industrialization has also led to a redefinition of effective import sub-
stitution as the difference betweengrowthin outputwithno changein the import
ratio and the observed rate (Chenery, 1986). Supply is defined as the sum of final
domesticdemand(D),intermediatedemand(R)andexports(E)
SQM RDE
Foragivendomesticsupplyratio(uQ=S),importsubstitutionisthendefinedas
u
2
u
1
S
2
15 Thiswasalsotheviewoftheinternationalfinancialinstitutionsatthetimeandfora
furthertwodecadesseeforinstanceWorldBank(1979).Indeed,aconditionforthe
receiptofAlliancefor Progressaidfundsfrom the USAafter1961 wasthe constitu-
tionofaplanningministry!
16 Thistoowasawidespreadassumptionbydevelopmenteconomistsatthetimesee
forinstanceRajandSen(1961),Winston(1967)orAtkinson(1969).
17 This is not set out formally by the ECLA, nor in Rodr guez (1980); so here we have
employedthestandardframeworkestablishedbyDornbusch(1980).SeealsoTaylor
(1991)foranextendeddiscussionofstructuralisttheoriesofinflation.
18 Itshouldbenoted,however,thatPrebisch'owninflationtheoryremainedbasically
monetarist.
19 The subsequent theory of ìnertial' inflation established in the 1980s was based on
marketexpectationsoffuturepriceandwagerises,butfallsoutsidetheperiodofthis
study.
20 Infactseethefinalsectionbelowcapitaldensity(K/L)inLatinAmericahasrisen
since1950morerapidlythanlabourproductivity(Q/L)leadingtoasteadydeclinein
theoutputcapitalratio(Q/K).
21 Forexample,asproposedbyLittle(seenextsection).
22 We shall not consider here the radical critique of ISI from Marxist and dependista
authors,mainlybecausetheyhavelittleeconomiccontent.
23 This is, of course, a widespread error it is frequently suggested by labour market
reformersthatwage/productivityratiosmakeLatinAmericauncompetitive,without
takingintoaccountotherfactorsofproduction.
24 ApointalsomadeinWorldBank(1979).
25 ForausefulsurveyofalternativecritiquesofISIinpractice,seeFishlow(1987).
26 An interesting and significant parallel is the espousal of structural adjustment by
senior state managers since the debt crisis. Here, too, is a doctrine which justifies
radicalstateinterventionofteninoppositiontoprivatesectorgroupsalbeitina
differentdirectionfromstructuralism.
27 Inotherwords,whatisnowtermed`convergence'inendogenousgrowththeory.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 93
28 Rodr guez(1980)identifiesthisasathirdECLAtheoryoftheinternationaltermsof
trade,followingthefirst(Ricardian)andsecond(elasticities)theories.
29 BywhichhemeansArgentina,Mexico,BrazilandChile.
30 SuchasColombia,PeruorVenezuela.
31 AsappearstobethecaseintheEastAsian`miracleeconomies'.
32 Aneffectwhichcanbeexacerbatedbytheeffectofimportsfrompoorercountrieson
wagedispersionandthedemandforunskilledlabour.
33 Alternatively,theECLAmodelmightbeseenasastandardÀK'modelwithincreas-
ingreturnstoscale(Romer,1986):
Y AK
a
; a>1
34 Wewouldprefertocallthis`marketrealism',astheseareintrinsiccharacteristicsof
almostallmarkets.
35 Asinallstudiesofthiskind,asimpleCobbDouglasformulationisassumed,where
thelabourcoefficientcanbeinferredfromthewageshareinnational incomeb, so
thattheestimationequationforTFPgrowth(a)isderivedfromthegrowthinoutput
(y)capitalstock(k)andofthepopulation(n).
a y1bkb
n
36 TestingeconometricallyforthecorrelationbetweenTFPgrowthratesandthegrowth
offactorstock.
37 Supposeadeterminatetrend()inthetermsoftradeandaproportionateexpansion
inboththecapitalstockandexportvolume,thenthesolutionfortherateofgrowth
ofTFP(a)fromequation(3.23)is:
a ykt
38 Citing Prebisch, Furtado and Tavares writing in the mid-1960s as his sources;
although his own direct experience appears to have been mainly confined to
Colombia.
39 Hirschman offers no empirical evidence whatsoever for his argument; a weakness
whichheshareswiththelatertheoristsof`rent-seeking'.
40 Hesuggeststhatothercareeropportunitiesforindustrializts'sonsmaypreventthis
sooutsiders(geographicalorethnic)arenotabadthing.Hecontraststhesuccessof
ISIinMexico(Monterrey),Brazil(SaoPaulo)andColombia(Medell n)withfailurein
ChileandArgentina,withoutsuchsecondaryindustrialcentres.
41 There is a curious parallel here with the current view of international institutions
suchastheWorldBank,thatonlyindependenttecnicoscancarryoutimportliberal-
izationagainsttheoppositionofthedomesticbusinesssector.
42 In this context, it is interesting to note that the rate of growth of labour skills was
morerapidin195073thanin198094(Hofman,1992).
43 Taylor (1996) attempts to measure the causes of Latin American growth using a
properly specified econometric model for the subsequent 197089 period using
internationalpaneldata.Hisresultsòfferlittlesupportfortheclaimsthataprinciple
tradedistortionliketariffsretardsgrowth...'(p.17);hefoundthatinvestmentrates
increase GDP growth, while population growth retards it. The much lower rates of
investmentintheregioncomparedtoAsia(forbothfixedandhumancapital)may
thus explain most of the notorious difference in GDP growth rates rather than the
policyregimeassuch,althoughhighcostsofcapitalandlowfinancialdevelopment
bothinhibitgrossfixedcapitalinvestment(p.22).Thisappearstobebecausethose
economies with higher levels of protection also invested more: ÌSI distorted the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
economy,but,bypromotingmanufacturing,thegrowthimpactwasnotassevereas
itmighthavebeen'(p.23).
44 The East Asian economies also started their industrialization drives with much
more equitable wealth distributions often underpinned by land reform. This
inequalityhadalwaysbeenseenbyECLAasaconstraintongrowthinLatinAmerica
(Prebisch,1950).
45 Furtado has a somewhat more radical interpretation: `since public investment is
largely intended to create external economies for private investment and since the
latter is in the hands of the two per cent minority whose savings represent a sig-
nificantproportionofitsincome,itmaybededucedthatbothinthewayitfinances
its expenditure and in the way it allocates its expenditure, the action of the State
servesnotonlytoconsolidatetheexistingpatternofwealthandincomedistribution
buttofosteranevenmoreconcentratedone'(Furtado,1970,p.67).
46 Thisnecessarilybroadestimateseemstobesupportedbythemoredetailedevidence
forPeruandColombiacitedinFitzGerald(1978).
47 Only Mexico, during the desarrollo estabilizador period from 1952 to 1976, had
constructedanalternativefiscaldeficitswerecoveredbymarginalreserverequire-
mentsimposedonbanks.
References
Atkinson,A.B.(1969)ÌmportStrategyandGrowthunderConditionsofStagnantExport
Earnings',OxfordEconomicPapers,vol.21(3)pp.32539.
Baer,W.(1974)`TheRoleofGovernmentEnterprisesinLatinAmerica'sIndustrialisation',
inD.T.Geithman(ed.),FiscalPolicyforIndustrialisationandDevelopmentinLatinAmer-
ica,Gainsville,FA:UniversityofFloridaPress.
Balassa, B.A. (1980) The Process of International Development and Alternative Development
Strategies,Princeton:PrincetonUniversityPress.
Barro,R.J.andSala-i-Martin,X.(1995)EconomicGrowth,NewYork:McGraw-Hill.
Bruton, H.J. (1970) `The Import-Substitution Strategy of Economic Development: A
Survey',ThePakistanDevelopmentReview,vol.10(2),pp.12346.
(1968)ÌmportSubstitutionandProductivityGrowth',JournalofDevelopmentStudies,
vol.4(3)pp.30626.
(1967)`ProductivityGrowthinLatinAmerica',AmericanEconomicReview,vol.57(5)
pp.1099116.
Cardoso,F.H.(1971)Ideologiasdelaburgues aindustrialensociedadesdependientes,Mexico
City:SigloXXI.
(1967)`TheIndustrialElite',inS.M.LipsetandA.Solari(eds),ElitesinLatinAmerica,
NewYork:OxfordUniversityPress.
Cardoso, F.H. and Falletto, E. (1971) Dependencia y desarrollo en America Latina, Mexico
City:SigloXXI.
Chenery,H.(1960)`PatternsofIndustrialGrowth',AmericanEconomicReview,September.
Chenery, H., Robinson, S. and Syrquin, M. (1986) Industrialisation and Growth: A Com-
parativeStudy,NewYork:OxfordUniversityPress.
Corden, W. (1984) `Booming Sector and Dutch Disease Economics: A Survey and Con-
solidation',OxfordEconomicPapers,vol.36,pp.35980.
Desai,P.(1969)ÀlternativeMeasuresofImportSubstitution',OxfordEconomicPapers,vol.
21(3),pp.31224.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 95
Dennison,E.F.(1967)WhyGrowthRatesDiffer,Washington,DC:BrookingsInstitute.
D az-Alejandro, C.F. (1984a) `The 1940s in Latin America', in M. Syrquin, L. Taylor and
L.E. Westphal (eds), Economic Structure and Performance: Essays in Honor of Hollis B.
Chenery,Orlando,FA:AcademicPress.
(1984b)`LatinAmericainthe1930s',inR.Thorpop.cit.
Dixit, A.K. and Pindyck, R.S. (1994) Investment under Uncertainty, Princeton: Princeton
UniversityPress.
Dornbusch,R.(1980)OpenEconomyMacroeconomics,NewYork:BasicBooks.
Dornbusch, R. and Edwards, S. (eds) (1991) Macroeconomic Populism in Latin America,
Chicago:UniversityofChicagoPress.
ECLA(1973)Lastareasdelapol ticafiscalytributariaalaluzdelaproblematicadedesarrollo,
Santiago:UnitedNations.
Edwards, S. (1995) Crisis and Reform in Latin America: From Crisis to Hope, New York:
OxfordUniversityPress.
(1993)Òpenness,TradeLiberalisationandGrowthinDevelopingCountries',Journal
ofEconomicLiterature,vol.31,pp.135893.
Fajnzylber,F.(1983)LaindustrializaciontruncadeAmericaLatina.Mexico:NuevaImagen.
Feenstra,R.C.(1996)`TradeandUnevenGrowth',JournalofDevelopment Economics,vol.
49,pp.22956.
Fishlow,A.(1987)`TheStateofLatinAmericanEconomics',inEconomicandSocialProgress
inLatinAmerica,Washington,DC:Inter-AmericanDevelopmentBank.
FitzGerald, E.V.K. (1994) ÈCLA and the Formation of Latin American Economic Doc-
trine', in D. Rock (ed.), Latin America in the 1940s: War and Transitions, Los Angeles:
CaliforniaUniversityPress,pp.89108.
(1993) Èconomic Commission for Latin America and the Caribbean', in J.
Krieger (ed.), Oxford Companion to the Politics of the World, Oxford: Oxford University
Presspp.2556.
(1978)`TheFiscalCrisisoftheLatinAmericanState',inJ.F.J.Toye(ed.),Taxationand
EconomicDevelopment,London:Cass.
FitzGerald,E.V.K.andPerosino,G.(1999)`TradeLiberalization,EmploymentandWages:
ACriticalApproach',inR.Faini(ed.),LabourMarketsandPovertyinDevelopingCountries,
Oxford:OxfordUniversityPress.
FitzGerald, E.V.K, and Vos, R. (1989) Financing Economic Development: A Structuralist
Approach toMonetaryPolicy,Aldershot:Gower.
FitzGerald, E.V.K, Jansen, K. and Vos, R. (1994) Ìnternational Constraints on Private
Investment and Economic Adjustment in Developing Countries', in J.W. Gunning
andH.L.M.Kox(eds),Trade,AidandDevelopment,London:Macmillan,pp.185220.
Flanders, M.J. (1964) `Prebisch on Protectionism: an Evaluation', The Economic Journal,
Vol.24,No.294.
Floto,E.(1989)`TheCenterPeripherySystemandUnequalExchange',CepalReview,no.
39,pp.13554.
Furtado,C.(1985)AFantasiaOrganizada,RiodeJaniero:PazeTerra.
(1970) Economic Development of Latin America, Cambridge: Cambridge University
Press.
Gerschenkron, A. (1969) `History of Economic Doctrines and Economic Post-war His-
tory',AmericanEconomicReview,vol.59(2),pp.117.
(1962) Economic Backwardness in Historical Perspective, Cambridge, Mass.: Harvard
UniversityPress.
Grossman, G.M. and Helpman, E. (1991) Innovation and Growth in the Global Economy,
Cambridge,Mass.:MITPress.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Ground,R.L.(1988)`TheGenesisofImportSubstitutioninLatinAmerica',CepalReview,
no.36,pp.179203.
Hirschman,A.O.(1968)`ThePoliticalEconomyofImport-SubstitutingIndustrialisation
inLatinAmerica',QuarterlyJournalofEconomics,vol.82(1)pp.132.
(1961)ÌdeologiesofEconomicDevelopmentinLatinAmerica',inA.O.Hirschman
(ed.),LatinAmericanIssuesEssaysandComments,NewYork:TwentiethCenturyFund.
(1958)TheStrategyofEconomicDevelopment,NewHaven,Conn.:YaleUniversityPress.
Hofman,A.A.(1992)`CapitalAccumulationinLatinAmerica:ASix-CountryComparison
for195089',ReviewofIncomeandWealth,vol.38(4).
Hofman, A.A. (1999) `The Economic Development of Latin America: A Historial Perspective',
London:EdwardElgar.
Kafka, A. (1961) `The Theoretical Interpretation of Latin American Economic Develop-
ment',inH.S.Ellis(ed.),EconomicDevelopmentinLatinAmerica,NewYork:StMartin's
Press.
Kay, C. (1989) Latin American Theories of Development and Underdevelopment, London:
Routledge.
Krueger, A.O. (1995) Trade Policies and Developing Nations, Washington, DC: Brookings
Institution.
Krugman,P.(1989)`NewTradeTheoryandtheLessDevelopedCountries',inG.A.Calvo
andothers(eds),Debt,StabilisationandDevelopment,Oxford:Blackwell.
Little,I.,Scitovsky,T.andScott,M.(1970)IndustryandTradeinSomeDevelopingCountries:
AComparativeStudy,Oxford:OxfordUniversityPressfortheOECD.
Londero,E.andTeitel,S.(1996)ÌndustrialisationandtheFactorContentofLatinAmer-
icanExportsofManufactures',JournalofDevelopmentStudies,vol.32(4),pp.581601.
Ocampo,J.A.(1990)`NewEconomicThinkinginLatinAmerica',JournalofLatinAmerican
Studies,vol.22,pp.16982.
Pinto, A. (1965) `Concentracio n del progreso tecnico y de sus frutos en el desarrollo
latinoamericano',ElTrimestreEconomico,vol,35(1),no.137.
Prebisch, R. (1964) Nueva politica comercial para el desarrollo, Mexico: Fondo de Cultura
Econo mica.
(1959) `Commercial Policy in the Underdeveloped Countries', American Economic
Review,vol.44,pp.25173.
(1950)TheEconomicDevelopmentofLatinAmericaanditsPrincipalProblems,NewYork:
UnitedNations,EconomicCommissionforLatinAmerica.
Raj, K.N. and Sen, A.K. (1961) Àlternative Patterns of Growth under Conditions of
StagnantExportEarnings',OxfordEconomicPapers,vol.13(1),pp.4352.
Rodr guez,O.(1980)Lateor adelsubdesarrollodelaCepal,Mexico:SigloXXI.
Romer, P.M. (1986) Ìncreasing Returns and Long Run Growth', Journal of Political Econ-
omy,vol.99,pp.100237.
Sapsford, D., Sarkar, P. and Singer, H.W. (1992) `The PrebischSinger Terms of Trade
ControversyRevisited',JournalofInternationalDevelopment,vol.4(3).
Sarkar, P. and Singer, H.W. (1991) `Manufactured Exports of Developing Countries and
theirTermsofTradesince1965',WorldDevelopment,vol.19(4).
Sheahan, J. (1987) Patterns of Development in Latin America: Poverty, Repression and Eco-
nomicStrategy,Princeton:PrincetonUniversityPress.
Singer,H.(1950)`TheDistributionofGainsbetweenInvestingandBorrowingCountries'
AmericanEconomicReview,vol.40,pp.47385.
Solimano,A.(1996)ÈconomicGrowthunderAlternativeDevelopmentStrategies:Latin
America from the 1940s to the 1990s', in A. Solimano (ed.), Road Maps to Posterity:
EssaysonGrowth andDevelopment,AnnArbor:UniversityofMichiganPress.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
E.V.K.FitzGerald 97
Syrquin, M. and Chenery, H. (1985) Patternsof Development, 1950 to 1983, Washington,
DC:WorldBank.
Taylor, A.M. (1996) Òn the Costs of Inward-Looking Development: Historical Perspect-
ivesonPriceDistortions,GrowthandDivergenceinLatinAmericafromthe1930sto
the1980s',NBERWorkingPaperno.5432,Cambridge,Mass.:NBER.
Taylor, L. (1991) Income, Distribution, Inflation and Growth: Lectures on Structuralist Eco-
nomicTheory,Cambridge,Mass.:MITPress.
Thorp, R. (1984) Latin America in The 1930s: The Role of the Periphery in World Crisis,
London:Macmillan.
(1971)ÌnflationandtheFinancingofEconomicDevelopment',inK.B.Griffin(ed.),
FinancingDevelopmentinLatinAmerica,London:Macmillan.
Villareal, R. (1988) Industrializacion, deuda y disequilibrio externo en Mexico: un enfoque
neoestructuralista(192988),Mexico:FondodeCulturaEcono mica.
Winston,G.(1967)`NotesontheConceptofImportSubstitution',PakistanDevelopment
Review,Spring.
WorldBank(1987)WorldDevelopmentReport,1987,NewYork:OxfordUniversityPress.
(1979)WorldDevelopmentReport,1979,NewYork:OxfordUniversityPress.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
4
TheInfluenceofInternational
FinancialInstitutionsonISI
RichardWebb
Outline
Did international financial institutions (IFIs) influence import-substituting
industrialization (ISI) in Latin America? This chapter will examine the roles
played by the World Bank, the International Monetary Fund (IMF) and the
Inter-American Development Bank(IDB) overtheperiod fromthecreation of
theBretton Woods institutionsin1944 tothesecondoilcrisis in1979. These
institutions were minor players in the region during the 1950s, but their
lendingandinfluenceincreasedthroughthe1960sand1970s.TheIDBbegan
operationsonlyin1960,butsoonrivalledtheBankasasourceoffundswithin
theregion.ItwastheWorldBankaboveall,however,thatacquiredavoicewith
regardtodevelopmentpolicyissues.
ISIwillbeunderstoodherebroadlyasreferringtomeasuresandpoliciesthat
favour manufacturing activity in some preferential or discriminatory manner
and that, therefore, as a rule carry a first-round cost to non-manufacturing
economic activities. This rather wide definition of ISI covers, for instance,
import protection, industrial credits, public utility rates, exchange rates, eco-
nomic infrastructure, preferential tax treatment, supportive labour regimes,
vocational training, technology centres and other policies and measures to
achievewhattheEconomicCommissionforLatinAmericaandtheCaribbean
(ECLAC)calledàcceleratedindustrialdevelopment'.
To make the term ISI meaningful it would seem necessary to insist on the
choiceinvolved;ISIshouldbeunderstoodasapolicythatjustifiespreferential
treatment to import-substituting manufacturing rather than to exports
whether these are primary or manufactured to agriculture, or simply to
balancedgrowth.Thistrade-offisoftenblurredbyadvocatessincehardchoices
do not advance the cause of persuasion. The usual argument, instead, is that
farmers and other non-industrial members of society will eventually be
betteroffunderanISIregimethanunderregimesthatgivethemfirstplacein
98
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 99
policy treatment. In fact, the selling of ISI has been so successful that ìndus-
trialization', and by implication ISI policy, became synonyms for economic
development.Forclarity,IwilldistinguishbetweenISI,bywhichIwillmeana
policyregime,andindustrialgrowth,whichisanoutcome.
The actual achievement of industrial growth may of course be the con-
sequence notonly of ISI butalso of policies and measures thatdo not setout
to favour industry orthat do nothave industrialization as their primary goal.
This is usually the case, for instance, of exchange rate and tax policies and of
investment in infrastructure. Unintended ISI may also be the result of export
promotionwhenittakestheformofsubsidiesonmanufacturedexportsaimed
at generating foreign exchange. Similarly, industrial projects aimed at serving
agriculture, such as investments in fertilizer production or in food-processing
capacity, contribute to ISI. For that reason the influence of IFIs on ISI can,
potentially, extend well beyond what could be called `deliberate ISI', that is,
beyond the lending, policy advice and conditionality by the IFIs directed
specifically at industrial development. However, the detection of unintended
ISIposesagreaterresearchproblem.Evidenceofanìndustrializationintent'is
sufficient to establish `deliberate ISI' but proof of ùnintended ISI' requires an
analyticalimputation.
PerhapsthemostnotablecaseofunintendedinfluenceonISIbytheIFIshas
beentheresultofexportpromotion.Thoughexportpromotionoftenseeksto
reduceimportprotection,andinthatwaycaninfactconflictwithISI,export
promotionisalsopursuedthroughdevaluation,industrialcredits,provisionof
infrastructureandtaxadvantagesthatcanworktofavourbothexportsandISI.
And, even less directly, the additional foreign exchange and the overall eco-
nomic growth made possible by additional exports both contribute to indus-
trialization.
Did the IFIs foster industrial growth in Latin America? Any answer must be
complex, for several reasons: an attribution problem is posed by unintended
ISI,threeinstitutionsareinvolved,institutionalpolicieshavevariedovertime,
andtheinstitutionshavenotbeeninternallyconsistentintheirpolicystance.
Thereisalsoaquestionofdurability.Physicalinvestmentsthatadvanceindus-
trialization,suchasasteelplantorpowergeneration,aregenerallylong-lived.
But tariffs and other trade restrictions are frequently modified. And, in an
inflationary environment, price policies that bear on ISI such as exchange
rates,interestratesandutilitytariffscanchangedirectionovernight.
BrettonWoods(1944)tothefirstoilcrisis(1972)
BrettonWoodswasaboutmonetaryandfinancialarrangements,butthesewere
understood as means to the restoration and expansion of world trade and
production. The Conference was driven by a vision of a postwar era of free
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
100 InternationalFinancialInstitutionsandISI
trade, free capital movements and stable currencies. The Fund and the Bank
werecreatedtoprovideaframeworkoffairness,stabilityandahelpinghandfor
a commercially and financially interdependent world. Article I of the IMF
Articles of Agreement requires the institution `to facilitate the expansion and
balanced growth of international trade', and similar language is contained in
ArticleIoftheBank'scharter.TheentireBrettonWoodsagreementwasdriven
byaconvictionthatanopenworldeconomywouldprovideinsuranceagainst
reneweddepressionandinternationalconflict.Indeed,anInternationalTrade
Organization, which came into being in `temporary' fashion as the General
AgreementonTradeandTariffs,wasanintegralpartofthebroaderschemeand
itspromptcreationwasenvisagedbytheconfereesatBrettonWoods.Accord-
ingtoWolfgangKonig,the`guidingprinciple'atBrettonWoodswastoestab-
lish àsuitable framework for the free operation ofcompetitive market forces'
(Konig, 1973, p. 119). A belief in free trade and an anti-protectionist stance
werethusimbeddedinthegeneticmakeupofboththeFundandtheBank.
Anti-protectionism, moreover, was not merely a distant objective, to be
brought about gradually; the immediate task at Bretton Woods, as conceived
by the USA, the dominant partner in this arrangement, was to prod and ease
thewayforareluctantUKintotradeliberalization.
The UShadaprofound conviction of themerits oftrade liberalisation and
an abhorrence of the bilateral trade and manipulated exchange rates that
hadbeenoperatedbyNationalSocialistGermany.TheUnitedKingdom,on
the other hand, wanted to find a mechanism to protect itself against the
immediateimpactofthetradeliberalisationrequiredbytheUS.
(James,1996,p.30)
Likemodernstructuraladjustmentloans,theBrettonWoodsagreementtraded
US funding for British liberalization. And when the Bank negotiated its first
loan,US$250millionforreconstructionpurposestoFranceinearly1947,trade
liberalizationwasacentralpolicyrequirement.
Yet,whenitcametoLatinAmericandevelopmentpoliciesoverthesucceed-
ing quarter decade, neither the Bretton Woods institutions nor the Inter-
AmericanDevelopmentBankusedtheirinfluencetoanysignificantextentto
counter ISI. It could be argued that their day-to-day financial business was at
several removes from trade policy, and that their institutional energies and
finite leverage were occupied by the need to ensure macroeconomic credit-
worthinessandgoodprojectimplementation.Still,operationalprioritieswere
not an obstacle to greater intervention on trade in the early 1980s, when all
threeinstitutionstookupthecauseoftradeliberalizationwithmuchenergy.It
seems likely, instead, that other circumstances explain the largely passive
stanceoftheIFIsonISIfromthelate1940sthroughthe1970s.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 101
TheBankandtheFund
Tobeginwith,theengagementoftheBankandtheFundwithLatinAmerican
development came about gradually. Delays were caused by the ratification of
theBrettonWoodsagreementandbystart-updifficulties.AndwhentheBank
finallybeganlendinginMay1947,itopenedwithitssightssetonwhatKeynes
had called its `primary duty' European reconstruction. Its lending to Latin
America remained small: from 1948 through 1960 it averaged US$85 million
peryear,barely1percentofannualimportsbytheregion(ECLAC,1964,p.40;
Kapur, Lewis and Webb, 1997), and consisted mostly of loans for energy and
transportationprojectsaweakplatformforinfluenceontradepolicy.
Infact,theBankandtheFundmighthavehadanevenlowerprofileinLatin
Americaduringthe1950sand1960shaditnotbeenfortheunexpectedonsetof
theColdWar.During1947and1948Europeanreconstructionbecamethemost
urgent objective of American national security and the paradoxical effect was
thatthe`BankforReconstruction'waselbowedoutofthereconstructionbusi-
nessbytheMarshallPlan.TheBankfoundareasonforbeingin`development',
atermwhichhadbeenlittlemorethananafterthoughtatBrettonWoods,and
acceleratedandintensifieditsactivitiesintheless-developedareas.But,atthe
same time, it continued to lend heavily to Europe and to European former
colonies,suchasAustralia,underthelabelof`developmentlending'.
Both of the Bretton Woods institutions were caught up by the emergency.
Economic disruption, shortages and hunger throughout Europe came to be
associated with communism, as cause or effect. `Political health in Europe
depended on economic medicine' (Packenham, 1973, p. 33). The Bretton
Woods vision a world of open trade, capital flows and monetary stability,
anchoredontheFundandencouragedbyWorldBankcreditsandguarantees
wasputasideandsubordinatedtotherequirementsofanintensiveprogramme
torehabilitateandintegrateEuropeundertheMarshallPlan.RichardGardner
commentedonthisperiod:
The new measures devised to deal with the post-war disequilibrium soon
overshadowed the financial institutions designed in the war. The normal
objectivesoftheInternationalMonetaryFundweregraduallysubordinated
to the immediate requirements of European recovery...[T]he Fund aban-
doned its conservative lending policy and extended US$610 million in aid
to member countries. Many of these loans could hardly be considered for
short-termstabilisationpurposes,buttheydidservetofillthegapuntilthe
newmeasuresofEuropeanaidwerepassedbytheAmericanCongress.
The operations of the International Bank also yielded priority to the new
programmeofreconstructionaid,[but]...inpractice,theWesternEuropean
countries preferred Marshall Aid to assistance from the Bank, since the
former came as grants or loans on easier credit terms than the Bank could
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
make available. Therefore, as the Marshall Plan gained momentum, the
Bank moved out of the reconstruction field. It turned instead, somewhat
modestlyatfirsttothejobofhelpingunder-developedcountries.
(R.Gardner,1980,pp.3034)
OneprobableeffectofthenewsecurityimportanceoftheIFIswastoincrease
US determination to control the institutions. US control had never been in
doubt, but international participation at Bretton Woods the US had refused
Keynes' preference for a mere, two-nation summit and the design of the
charters approved at the conference amounted to tentative concessions to
internationalism. In Bretton Woods' peacetime scenario those steps might
have grown into an acceptance of more independent and more truly multi-
national institutions. Now, heightened US influence, partly expressed in IFI
board-approvedpoliciesandlendingdecisionsbutalsointheirstaffing,almost
certainly gave an additional edge to the open-economy, trade-promoting
inclinationsoftheBankandIMFagainstmoreinterventionistandprotection-
istEuropeaninclinations.
FortheBrettonWoodsinstitutions,however,themorevisibleandprofound
effectoftheColdWarwasthatpointedoutabovebyGardner:thecreationofa
massive, new financial actor in the form of the Marshall Plan and, more
generally, of US bilateral aid. From 1948 and for at least two decades US
bilateral aid would dwarf the funding and mute the voice of the Bretton
Woodsinstitutions.Also,becauseUSaidwasalmostentirelyfocusedonEurope
andonafew`rim'countriesborderingtheUSSRandChina,boththeIMFand
the Bank turned their attentions to those countries that were not awash in
MarshallPlanorotherUSaid.TheIFIshadbeencreatedas`world'institutions,
but both now began an evolution that by the late 1960s would convert them
almostentirelyintodeveloping-country-orientedagencies.
For the most part, until the large-scale decolonization of the 1960s, that
`developing world' consisted of Latin America and South Asia. But it was in
Latin America especially where, during the 1950s, the institutions began to
engage with doctrinal issues of development policy. India and Pakistan were
consideredtobelessinneedofadviceorofhands-onproject-levelassistance,
thanks to their responsible macroeconomic policies and strong and sophist-
icated and Oxford-trained civil services. Also, policy leverage in India and
Pakistan was undermined by their importance in the Cold War. By contrast,
LatinAmericahadlittleColdWarleverageand,instead,appearedtosufferfrom
weakeconomicmanagementandweakpublicadministration.Theregionthus
calledforandevenprovokedagreaterpolicyinterestbytheBrettonWoods
institutions.
For the Bank, market orientation with respect to trade policy began as a
general principle inherited from the Bretton Woods vision. But the doctrine
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 103
was strongly reinforced by practical circumstances: through the 1950s and
1960stheBankhadtosurvivebysellingitselftoWallStreet.Itscapitalendow-
mentwasquicklyexhaustedinreconstructionloans,andalladditionallending
before the creation of the International Development Association (IDA) in
1960wasfundedinapostwar,skittishWallStreet,scepticalofforeignlending
and more so of government borrowers, and, moreover, market-oriented in its
convictions. The result was a highly conservative lending policy centred on
a concrete, productive, revenue-generating and often foreign-exchange-
generatingportfolio.
This wasespecially true ofLatin Americawhere loanshadtostand on their
own,withonlyweaksupportfromUSforeignpolicyinterests,andwhich,after
1960,didnotbenefitfromtherelaxationoflendingcriteriamadepossibleby
theIDA'sconcessionaltermsandnon-marketfunding.Thusinvestmentssuch
as roads and railroads, which directly served exportable resources, as well as
`self-liquidating' energy projects, became the staples of Bank lending to the
region. As late as June 1971 three-quarters of Bank lending to Latin America
consistedoftransportationandenergyprojects.
When it came to trade and industrial policy, the Bank's position was to
encourage export promotion and moderate import protection. But, in the
main, the Bank's voice on these issues was weak. Neither lending volume nor
reputationprovidedmuchclout.Itslendingwassmallinabsolutetermsandin
comparison with US bilateral and with Export-Import Bank funding. And, as
yet, it had little of the prestige as a developmental institution that it would
come to enjoy by the late 1970s. During the 1950s it downplayed economic
studies.Itwasonlyinthe1960sthatitbegantodevelopasubstantialin-house
research staff and publications, and to complement its lending work with
missionsthatcarriedoutmajoreconomicandsectorstudies.
ButitwasdisinclinationasmuchasweakmusclethatmutedtheBankwhen
it came to trade and industrial policies. This disinclination contradicted the
inhibitionsagainstISIthatwerebuiltintoitsinstitutionalgenesandreinforced
bycontinuingUSoversight.
For one, the Bank's staff did not lean to doctrines, a result, perhaps, of the
notable absence of economists at senior levels. Instead, senior officials had
been recruited from careers in finance or public administration. The Bank's
strong Executive Vice-President, Robert Garner, a former private banker and
businessman,downgradedeconomicresearch,placingmosteconomistsatthe
service of regional loan offices.
1
Rosenstein-Rodan, the Bank's most powerful
academiceconomist,resignedin1954afteronlyafewyears.Garnercomment-
edthat,ÌfranklyneverthoughtthattheBankwashisdishoftea...theBank
wasnottheplaceforthedevelopmentofbroadeconomicpoliciesorstudies'.
Moreover,economicdoctrineitselfbegantoprovidemixedmessages.Asthe
fledgling science of development economics zeroed in on the critical role of
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
manufacturing, establishment economists, without backtracking on their
belief in free trade as a general principle, were nonetheless increasingly
inclined to accept market interventions that favoured industrialization. A
1951 United Nations report stated that èconomic development has to be
thought of largely in terms of industrialization... [Though] due importance
should be attached to agriculture in national development, it is nevertheless
true that industrialization forms the decisive element of economic develop-
ment'.HollisChenery,whowouldbecometheBank'schiefeconomistin1970,
andwhowouldhelpMcNamararedirecttheBanktowardsagricultureandthe
poor,begana1955articleondevelopmentwiththesentence:Ìndustrialisation
is the main hope of most poor countries trying to increase their levels of
income'. The most influential statement, however, was provided by Arthur
Lewis, whose model gave short shrift to agricultural growth and put industri-
alizationattheheartofdevelopment.TheLewismodel,thoughitcamelatein
the day from the point of view of the ideas that influenced Bank lending
allocation during the 1950s, confirmed and strengthened the less rigorous,
common-sense notions of the bankers, lawyers, administrators and even eco-
nomistswhoseday-to-daydecisionswereshapingtheBank'slendingpolicyin
thatdecade.
However,intheenditwastheBank'sroleasalendermorethantheleanings
ofitsstaffthatpushedtheinstitutiontowardswhatthestaffthemselveswould
describe asa`pragmatic' position on development, andwhich came tomean,
in many cases, support for ISI. That role centred on the making of visibly
productive, `quality' project loans, and on ensuring macroeconomic credit-
worthiness.
Efforts to influence borrower policies focused on good debtor behaviour:
repaymentofprewardebtarrears,settlementofclaimsarisingfromtheexpro-
priation offoreignassets,and`sound'fiscalandmonetarypolicies.Asasmall
lender, in any case, the Bank had little policy leverage and, at the time, was
more likely to refuse to lend when creditworthiness conditions were not met
than to sustain and use the banking relationship to press for better policies.
ThishappenedinbothBrazilandGuatemalainthemid-1950sandineachcase
thebreaklastedforabouttenyears.TheBank'spotentialpolicyleverage,from
the standpoint of ISI, was further circumscribed by a strong inhibition with
respecttointerventioninthematterofexchangerates,consideredtobefirmly
anIMFbailiwick.
For these several reasons, then, it is less surprising that when it ventured
outside the area of strict, general-purpose infrastructure, as it did increasingly
in the 1960s, the Bank on occasion found itself lending for ISI-related
projects. Thus the Bank supported a steel plant in Chile, as it was doing in
India, Turkey and Japan. It also extended to Latin America, beginning in
Colombia, its support for development finance companies which were
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 105
expectedtolendprincipallytoimport-substitutingindustry.Furthermore,the
private-sector-orientedInternationalFinanceCorporationoftheBank,created
in 1956 to lend to the private sector, was not averse to demanding higher
importtariffsforthesectorsinwhichitwasinvesting.
Apro-ISIbiaswascreatedbytwochangesintheBank'sprojectselectionand
procurement procedures. Thus, protection of local capital goods industries
became institutionalized in 1962 when the Bank adopted a 15 per cent pre-
ference rule for local suppliers of Bank projects: when bids were compared,
local companies were allowed to charge up to 15 percent more than interna-
tional suppliers.
2
A 1969 proposal to raise that preference to 25 per cent,
however,wasrejected.Thesecondchangeboreonprojectappraisalmethods,
which during the 1950s began to move from an almost exclusive reliance on
financial rates of return to an experimentation with, and application of, eco-
nomic rates of return. The most important modifications were the use of
shadow exchange ratesand shadow wages, bothof which served to justifyISI
projects.
Thisgrowingpragmatismattheoperationallevelcoexistedwithameasureof
continuing, underlying, pro-openness ideology. This was done, in part, by
caricaturing and satanizing Prebisch and ECLAC. A former Vice-President for
Latin America recalled that: `we were very skeptical...[and] critical some-
timesverycriticalofthedoctrinesespousedbyRau lPrebischinthedirection
ofautarchyandisolation'(SeeKnapp,1975).Atthesametimeitrationalizedits
owndoctrinalflexibilitywithtermssuchasàppropriateISI'andèfficientISI',
and by accompanying its ISI project support with boiler-plate lectures that
consistently insisted on openness and the desirability of export promotion.
This contrast between action and rhetoric perhaps explains the contradictory
recollectionsofstaffmembersaskedtodescribetheBank'spositionvis-a-visISI
in the 1950s and 1960s. Some were certain that the Bank had steadfastly
opposed ISI, but others a majority of those interviewed remembered that
theBankhad`tolerated'ISI.
3
TheInter-AmericanDevelopmentBank
TheInter-AmericanDevelopmentBankwassparedthesedoctrinalambiguities:
thetimeandcircumstancesofitsbirthallowedittoopenwithandmaintaina
moresympatheticstancewithregardtoISI.Yet,despitethatlackofinhibition,
throughthe1960sand1970stheIDBwasforthemostpartapassiveactorwith
respecttoISI:lessthan15percentofitslendingwasallocatedtoindustryand
eschewedpolicystatementsonISIpolicies.
UndersecretaryofStateDouglasDillon'sannouncementatameetingofthe
Inter-American Economic and Social Council in August 1958, that the USA
wouldsupportaregionaldevelopmentbank,wasinpartaconcessiontoalong-
standingdemandbyLatinAmericangovernmentsandbyECLAC,butmostlyit
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
was a response to a new phase of the Cold War. Initially located along well-
defined fronts in Europe and then Asia, the war had become global. A major
reasonwasdecolonization;communistsubversionnowseemedpossibleinany
developingcountry,andinthisnewColdWareconomicaidcountedasmuch
asmilitaryaid.Seekinginstitutionaltools,theUSAlookedapprovinglyonthe
model of effective, businesslike aid that had been developed by the 1950s
WorldBank,increasingtheresourcesofthatinstitutionwiththeInternational
DevelopmentAssociation,andcreatingtheIDB.TheIDBannouncementcame
onlythreemonthsafterVice-PresidentRichardNixon'sunhappytriptoLatin
America, but the trip only added conviction and urgency to the new policy.
Thus,theIDBdecisionwasfollowedadaylaterbyasimilarofferfortheMiddle
East, while, months before, the European Economic Community had estab-
lishedadevelopmentfundforAfrica.
Moreover, the threat in America's backyard quickly and hugely escalated
with Fidel Castro's 1959 revolution and its sequel of guerrilla movements
throughouttheregion.Respondingtothatescalation,theUSArapidlyboosted
IDB lending capacity with the creation of a US$500 million Social Progress
TrustFundin1961,andinstructedtheinstitutiontolendforsocialobjectives,
such as health, education, housing and land reform projects. The broader
contextforIDBpolicieswasprovidedbytheAllianceforProgress,announced
in1960byPresidentJohnKennedy:theAlliancewasaboveallacallforsocial
reform and social spending, but it encouraged government activism more
generallyanddownplayedearlierUSfreemarketrhetoric.
The tolerance of ISI was not all due to the security emergency. As noted
above,theemergenceofdevelopmenteconomicsduringthe1950shadopened
thedoorforacceptanceofinterventioninfavourofindustrialization,aconsent
thatwasrationalizedinorthodoxcircleswiththetermèfficient'ISI.Agrowing
tolerationinUSacademicandpolicycirclesoftheparticipationbydeveloping-
country governments in productive activities also smoothed the way for IDB
support of ISI. By contrast, the World Bank had shut itself out of industrial
lending by its refusal to finance government-owned productive enterprises
whileborrowergovernments,inturn,weregenerallyunwillingtoprovidethe
required guarantees on Bank loans to private investments. These terms of
reference for the new IDB, provided partly in its charter and partly by the
climate of opinion of the times, had mixed effects on the way in which the
BankwouldapproachISI.
On the one hand they meant that the IDB was not saddled with an anti-
protectionistinhibition.IncreatingtheBanktheUSAwentbeyondamereaid
budget decision; the institution was a diplomatic gesture to Latin demands,
which had been most forcefully articulated by ECLAC, ISI's champion in
the region. Allowing Latins to occupy the presidency and most of its senior
management positions, the USA was consenting to a degree of beneficiary
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 107
òwnership' and to an institution that would, over close to three decades,
diplomatically refrainfromattemptingtoinfluenceborrowerpolicies.Rather,
theIDBwouldnecessarilytakeasympatheticoratleastuncriticallineregard-
ingtheprotectionistpoliciesofmanyLatinAmericangovernments.Themove
wasaradicalturnforaUSAthat,someyearsbefore,hadattemptedtoabolish
ECLACandhadstonewalledcallsforaidwithlecturesonthevirtuesofprivate
foreigninvestmentandfreetrade.
Ontheotherhand,thesecurityemergencyandits`sociallending'corollary
tendedtocrowdoutindustriallending.From1961through1972,46percent
ofIDBlendingwasallocatedtoagriculture,sanitation,urbandevelopmentand
education.Thus,overtheyears196168theIDBservedastheregion'sprincipal
funding source for external financing in agriculture, lending US$643 million
for that purpose, while the World Bank provided US$350 million and the
Agencyfor InternationalDevelopment (AID), US$297 million(Accame, 1970,
p. 40). The preference given to social lending outlived the political circum-
stances that had triggered its creation. The sense of a regional security emer-
gency had largely evaporated by the mid-1960s, but the new institutions and
theirsocially-orientedendowmentslivedonoutlivingtherhetoricandsmall
secretariatoftheAllianceforProgressespeciallytheWorldBank'sIDAandthe
IDB's Social Progress Trust Fund. Moreover, social lending was given new
impetus during the 1970s as public opinion in the donor countries began to
question the efficacy of `trickle down' and to insist on a link between aid
appropriations andpoverty targeting. Thus, theIDB'sAnnual Report for1974
referredtothattrickle-downcritique.Reaffirmingits`social'mandate,itdrew
attentiontothe`consciouspreference'itgave`tocertaineconomicandsocial
sectorswhichhavetendedtobebypassedasLatinAmerica'sdevelopmenthas
progressedinrecentyears':
Thus,throughoutits15yearhistorytheBankhasgivenspecialandprefer-
entialattentiontotheintegraldevelopmentofLatinAmerica'scountryside.
Agriculture has traditionally been the Bank's leading lending sector,
accountingcurrentlyfor23percentoftheBank'stotallending.
Butbeyondthat,withintheBank'ssectoralclassification,largesupportfor
the development of the rural sector is masked in the electric power, trans-
portation,andsanitationsectors,tonameafew.
(Inter-AmericanDevelopmentBank,1974,pp.313)
Despitetherhetoric,theconcentrationonsocialsectorsdeclinedslightlydur-
ing the 1970s. Nonetheless, `social' loans accounted for 40.6 per cent of the
Bank'stotalportfolioin1981.
4
Inadditiontothecompetitionfromthisself-consciouspreferenceforsocial
lending, the industrial sector was also squeezed by what in quiet practice was
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
thelargestclaimantonIDBresources,namelyeconomicinfrastructure.Enjoy-
ingbothlenderandborrowerpreferencesaswasalsothecasewiththeWorld
Banklargeenergyandtransportationprojectsabsorbed41.3percentofBank
resourcesover196181.
5
LoanstoindustrypeakedduringtheBank'sfirstyears.Theyamountedto19
per cent of all lending over 196168, but subsequently fell a few percentage
points.Overthelongerperiod196181theyaccountedforonly14percentof
Banklending.
6
OnereasonforthistailingoffwasagrowingquestioningofISI
withintheregion,evenbyitsmostoutspokenadvocatesinECLAC.IftheIDB's
window of opportunity for an energetic supportive response to ISI had been
blocked off during the 1960s by a social mandate, by the 1970s the window
itself was closing. Thus, reviewing the Bank's lending policies in 1970, an
official pointed to the `loss of dynamism' by Latin America's manufacturing
sector,notingthat:
By the end of the decade, while recognising the importance of continuing
efforts toward import substitution, it became evident that Latin American
industryhadtobemodernisedandmademoreefficientifitweretocontinue
its dynamic growth. In order to accomplish this it would be necessary to
lowertariffbarriers,atleastwithintheregion,soastowidenmarketpossi-
bilities, permit the establishment of larger production units and increase
competitionwhichwouldleadtoincreasedefficiencyandlowercosts.
Inthedecadeofthe1970s...itwillbefoundnecessarytobeginagradual
evolutionaryloweringoftariffswithrespecttotherestoftheworldsothat
healthy competition may force necessary adjustments in efficiency and
lowerprofitmargins.
(Lynn,1970,pp.269)
AsimilarnotewassoundedbytheBank's1977AnnualReport:
In its industrial loans, the Bank seeks to emphasise the objectives of eco-
nomic efficiency and competitiveness in both the domestic and interna-
tionalmarkets.Inrecentyears,particularemphasishasbeenplacedonlong-
termefficiency...
(IDB,1977,p.17)
Inprinciple,thisshifttoèfficientISI'neednothaveshutoutlendingforindus-
try, but the Bank was ahead of its borrowers in this respect, and continuing
protectionisminmuchoftheregionmeantthattheBanknowbegantosuffer
theinhibitionsthathadpreviouslyrestrainedWorldBanklendingforindustry.
But overt lending for industry was not the whole story. Support for ISI was
also provided by the IDB, first through the margin of preference allowed
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 109
domestic or other Latin American suppliers in procurement, and second
throughanearmarkedlendingwindowthatwasdescribedasanèxportfinan-
cing programme' but that, in fact, was directed at financing the regional
manufactureofcapitalgoods.
The Bank early on adopted a policy of granting domestic suppliers a 15 per
cent margin of preference equivalent to a tariff in that amount when
comparing bids for procurement for IDB projects. This policy followed World
Bank practice, mentioned above, but carried an interesting modification:
borrowers were allowed to extend the preference to suppliers from any
countryintheregion.Inpractice,however,borrowersrarelyusedtheregional
preferenceallowance,preferringtokeepthemarginfordomesticsuppliersonly
or, alternatively, to seek the lowest-price bids in international markets (IDB,
1992,p.10).
Theèxportfinance'programmewasanIDBinnovationtosupportISIinthe
form of regional production of capital goods destined for intraregional trade.
The programme, with US agreement, finessed potential objections to explicit
ISIbyfundingtheexporterratherthantheimporterineachtransaction.And,
more generally, it was legitimized by the IDB's mandate to support regional
integration. It was launched in September 1963 but its impact was limited
during the decade by insufficient funding and other restrictions. Efforts to
loosen and expand the programme were blocked in the board, principally by
the USA, until 1975 when Venezuela used oil boom revenues to circumvent
thoseobjectionsbycreatingagenerousUS$500millionVenezuelaTrustFund
tosupplementtheregionalexportfinancefacility.Yet,by1979,totalfinancing
under the programme over the period 196179 amounted to only US$299
million(IDB,1979,p.22).
Transition,197282
In the World Bank, the growing `tolerance' for ISI during the 1960s and early
1970s was partly a matter of doctrinal evolution and partly an effect of the
broadeningofitslendingawayfromanarrowconcentrationongeneralinfra-
structure. Thus, lending for education began in the 1960s and was especially
targeted at industrial vocational training and work-related skills. Lending to
development-finance companies to support industrial development became a
majorprojectcategory.But,moreimportantly,theBankwasacquiringagreater
capacityforandwillingnesstoadviseongeneraldevelopmentpolicies,andits
messagewithregardtoISIreliedheavilyontheconceptofèfficientISI'though
itcontinuedtoplacerhetoricalpriorityonexportpromotion.
ThecollapseofBrettonWoodsin1971andthefirstoilshockin1972greatly
increased the Bank's preoccupation with the balance-of-payments problems
of its borrowers. Though it had worked to avoid programme loans to its
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
less-developedborrowers,insistingontheprojectformat,theBankhadinfact,
throughthe1950sand1960s,continuallylooseneditsdefinitionofwhatwas
meantbya`project'toaccommodateshort-runforeignexchangeemergencies.
This accommodation was taken further during the 1970s as the Bank greatly
expanded its volume of lending and sought to alleviate balance-of-payments
problems. One effect was to act against devaluation and thus ISI since
balance-of-paymentssupportwasineffectasubstituteforbalance-of-payments
adjustment.
Inadditiontoresortingtoquasi-programmelendingthroughitsprojects,the
Bankalsoincreasedits`tolerance'forISIprojectsasawaytorespondtoforeign
exchangeneeds.Thiswasespeciallythecaseafterthefirstoilshockwhenthe
foreign exchange needs of oil importers presented themselves as structural
ratherthancyclicalproblems.
TheseeventscoincidedwithamoreresponsivesetofofficialsintheBank.As
President,RobertMcNamarawaslessinthegripofthefreemarkettenetsthat
hadshapedBrettonWoodsandthatnotwithstandingtheBank'spragmatism
and `tolerance' had been a major part of the mindset of previous presidents
andseniorofficials.McNamara,forinstance,droppedthebanonloanstostate-
owned enterprises, opening the door to an expanded volume of lending for
industrial projects. Also, the Bank's chief economist during the 1970s, Hollis
Chenery, had stressed the problem of foreign exchange scarcity in previous
writingsonthe`two-gap'model,andurgedtheBanktowardsamoreinterven-
tionistapproachtobalance-of-paymentsneeds.Onesignalofthisattitudewas
a deliberate change, towards respect and cooperation, in the research depart-
ment'srelationswithECLAC.
During the 1970s, then, the frequency of ISI-supportive investments and
advice continued to increase, with a notable expansion in development-
finance-company lending, in IFC operations, and in loans to the industrial
sector. Also, Bank procurement was made even more favourable to local sup-
pliersamountingtotheacceptanceofahigherimplicittariff.
The IMF's influence on trade and industry policy was at all times marginal.
LiketheBank,itwascaughtinacontradictionbetweenaninherentlypro-trade
and anti-protectionist primary purpose, and the occasional and expedient
appeal ofinterventionist measures. Duringthe 1940sand 1950s, forinstance,
it justified the application of advance deposits on imports on monetary
grounds (Konig, 1973, p. 157). More generally, the priority attached to
exchange rate stability by the IMF led it to propose various fiscal and tariff
measuresas`temporary'substitutesfordevaluation,measureswhichnaturally
tendedtobecomepermanent.Asfiscalimbalancetookonanincreasingweight
in IMF programmes, `temporary' import surcharges became a favourite pro-
posal. In principle, the surcharges were justified as being `flat', and even as
beinganti-protectionistbecausetheyreducedtherelativedifferencesbetween
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 111
import tariffs on different goods and so reduced effective protection. But in
practice `temporary' flat surcharges tended to become permanent, creating
more headroom for subsequent, minor changes in nominal rates that could
translateintoverylargeincreasesineffectiveprotection.
Seekingtoexplainandjustifyandconvertotherstowhathadbecomeitsde
factooperationalposition,itwasin1979,immediatelybeforetheBank'sturnto
a highly active anti-protection position, that it published its most thorough
andwidely-circulateddefenceofISI.ThishappenedintheWorldDevelopment
Report (WDR) of that year, which stated that `governments have played a
crucial role in initiating and supporting the early stages of industrialization'
(WorldBank,1979,p.62).TheWDRmentionedtraditionalarguments,suchas
theèconomiesofscaleinherent[in]large-scaleprojectsthatwerebeyondthe
capabilities of the local private sector', the òpportunities for local industrial
expansion' created by government infrastructure investments, and govern-
ment èfforts to marshal skills and financial resources and direct them to
industry'. It stated that in some developing countries, including Mexico, `the
majorgovernment-supporteddevelopmentbankshavebecomepivotalinstitu-
tions in industrial development'. It noted that public industrial enterprises in
India, Korea and Turkey had `made pivotal contributions', `played akey role',
or`facilitatedtheintroductionofnewmanufacturingtechnologies'.Itthought
that publicprivate industrial competition in Mexico `makes both more effi-
cient' (World Bank, 1979, pp. 627). Noting the many needs of industrializa-
tion, it stated that: `government action can help to alleviate some of these
problems',andthat:
All of the currently industrialising countries, with the sole exception of
HongKong,haveprotecteddomesticproductiontosomeextent...Initially,
however, despite the inefficiency of the protective devices used in many
countries, the encouragement of import-substituting industries has gener-
ally secured a rapid expansion in manufacturing output. In the 1950s and
1960s Mexican manufacturing, for example, grew at annual rates above 7
per cent; Brazil, the Philippines and Turkey, among other countries, have
alsoachievedconsiderableratesofgrowthinmanufacturingthroughimport
substitution. The development of the manufacturing sector has in turn
helped to create and spread industrial and entrepreneurial skills, and in
some countries, such as Brazil and India, has resulted in the emergence of
substantialdomestictechnologicalcapacity.
(WorldBank,1979,p.67)
Atthesametime,inresearchandcountryadvicetheBankwasplacingincreas-
ing stress on export promotion, and research was begun on the potential for
manufactured exports. But the more important intellectual development was
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the research and advocacy of Bela Balassa who pointed out the inefficiency
associatedwithhighimportprotectionusingtheconceptofeffectiveprotec-
tionwhichhehelpedtodevelopandtheimportanceofexportexpansion.
Thesecondoilshockin1979pushedtheBankevenmoredecidedlytowards
aconcernwithbalance-of-paymentsproblems,someofwhichwerebeginning
to threaten the Bank's own portfolio. Now, however, the response placed
particular emphasis on new and especially manufactured exports, and to that
end it was argued on effective protection lines that the way to manufactured
exportsrequiredarollbackofprotection.Thisargumentpro-industrialization
but anti-protection became the core position underlying the proposal for
what came to be called structural adjustment lending that dominated Bank
policyadviceduringthe1980s.Thepersuasivenessofthatargumenthadlessto
do with economic science, however, than with `guilt by association' evidence
provided by the recession and macroeconomic instability of the decade, and
also with its political attraction following a coincidental turn-of-the-decade
rightwardshiftinG-7politics.
Notes
1 Rist saw this change às a deliberate intention to humiliate people' (Rist, 1961).
However, many economists reassigned to lending departments went on to become
senior officials. Gerald Alter, for instance, became Director of the Western Hemi-
sphere Department (oral history (conversation) between George Woods and Robert
Oliver, 13 July 1985). By downgrading research and putting economists to work in
operations,Garnerineffectincreasedtheroleofeconomicanalysis.
2 Whentariffsexistedthegeneralcasetherulewasthattheexistingtariffor15per
cent,whicheverwaslower,wouldbeaccepted.
3 RecollectionsobtainedfromaquestionnairesurveydesignedandcarriedoutbyJohn
Williamson and returned by 51 acting and retired Bank officials during 1994, and
kindlylenttotheauthorforthepurposesofthisarticle.
4 Includeseducation,health,water,sanitationandurbandevelopment.
5 AlllendingfiguresarefromIDBAnnualReportsfordifferentyears.
6 ThesefiguresareslightlyexaggeratedbecausepublishedIDBstatisticsincludeloans
totheminingsectorwiththoseforindustry.
References
Accame,F.(1970)inTheIDB'sFirstDecadeandPerspectivesfortheFuture,inInter-American
DevelopmentBank,RoundTable,PuntadelEste,April.
Chenery,H.(1955)`TheRoleofIndustrialisationinDevelopmentProgrammes',American
EconomicReview,May.
ECLAC(1964)TheEconomicDevelopmentofLatinAmericainthePost-warPeriod,NewYork:
UnitedNations.
Gardner,R.(1969)SterlingDollarDiplomacy:TheOriginsandProspectsofOurInternational
EconomicOrder,NewYork:McGraw-Hill.
Inter-AmericanDevelopmentBank(1992)ReporttotheBoardofExecutiveDirectors,The
ProcurementPolicyoftheInter-AmericanDevelopmentBank,GP-11811,23March.
AnnualReports(1974,1977,1979),Washington,DC:IADB.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RichardWebb 113
James,H.(1996)InternationalMonetaryCooperationSinceBrettonWoods,Washington,DC:
OxfordUniversityPress.
Kapur, D., Lewis, J. and Webb, R. (1997) A History of the World Bank, Washington, DC:
BrookingsInstitution.
Knapp,B.(1975)Ìnterview',WorldBankOralHistoryProgramme,24and30July.
Konig,W.(1973)ÌnternationalFinancialInstitutionsandLatinAmericanDevelopment',
in V. Urquidi and R. Thorp (eds), Latin America in the International Economy, Toronto:
JohnWiley&Sons.
Lynn,J.(1970)inTheIDB'sFirstDecadeandPerspectivesfortheFuture,inInter-American
DevelopmentBank,RoundTable,PuntadelEste,April.
Packenham,R.(1973)LiberalAmericaandtheThirdWorld,Princeton:PrincetonUniversity
Press.
Rist,L.(1961)InterviewforWorldBankOralHistoryProgram,July,WorldBankArchives.
UnitedNations(1951)MeasuresfortheEconomicDevelopmentofUnder-DevelopedCountries,
NewYork:UN.
WorldBank(1979)WorldDevelopmentReport,Washington,DC.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
5
TheIndustrializationofChileduring
*
Protectionism,194082
RicardoFfrench-Davis,OscarMunoz,Jose MiguelBenaventeand
GustavoCrespi
The objective of this chapter is to present an evaluation of the process of
industrializationofChileduringtheperiod194082.Theconceptofindustrial-
ization is understoodin thebroadest sense, including industrial development
as well as institutional modernization and social transformation. During the
four decades which elapsed from the beginning of the 1940s, the Chilean
economy completed a cycle of its economic development and began a new
phase.Thecycleoftheso-called`developmentfromwithin',whichhadbegun
withthe1930scrisis,culminatedin197374withthebreakdownofthedemo-
craticregime.
1
Throughoutthedecadeofthe1940stheindustrialgrowthrate
reached an average of 8 per cent a year but thereafter declined perceptibly
(Munoz,1995).Fortheentireperiod195071,anannualgrowthrateof5.4per
centhasbeenestimated,wellbelowtheLatinAmericanaverageforthisperiod
whichreached6.7percent(Benaventeetal.,1996,table4).
In the mid-1970s the development strategy took a radical turn, modifying
the objectives and policies in order to structure an open, privatized economy
free from state intervention. The previous model had been characterized by
progressive state intervention, varying in intensity according to the nature of
thegovernment.Itwas,however,alwaysbasedonamajorityopinionthatthe
stateshouldplayacentralroleinmanagingtheprocessofgrowthandinvest-
ment.Thestructuraladjustmentinitiatedinthe1970scausedaperiodofgreat
economic instability, aggravated by the international events which followed
theoilcrisisandsubsequentdisruptionoftheinternationalcapitalmarket.The
average annual growth rate fell to 1.8 percent, greatly below the 3.4 percent
forLatinAmericain197480,withaslightrecoveryto2.6 percentinthe1980s
* Originally prepared for the project on Ìndustrialization, the State and Economic Development in
the post-Second WorldWar period', sponsored by theIDB and coordinated by Rosemary Thorp. We
acknowledgetheguidanceofRosemaryThorpandtheobservationsreceivedintheworkshopswhich
took place in Oxford in September 1996 and in Paipa in May 1997. This work draws extensively on
previousarticlesbytheauthors.
114
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
RicardoFfrench-Davis,OscarMunoz,Jose MiguelBenaventeandGustavoCrespi 115
while the region as a whole sank into stagnation. Only from the end of the
1980s did the Chilean economy succeed in consolidating the new model of
developmentwhichhasmanifestedanunprecedenteddynamisminthe1990s
anannualgrowthofGDPintheorderof7percentin199095.
Thisworkisorganizedinfiveparts,thefirstofwhichputsforwardevidence
on global and industrial economic growth and the increase in productivity
from1940onwards.Thesecondpartdiscussestheroleofthestateandindus-
trialization;itbeginswithahistoricalreviewofthebackgroundofstateinter-
ventionfromthe1920stoemphasizethereaftertheeffectofthecreationofthe
`Corporation for the Development of Production' (CORFO) in 1939. Also
included is a brief review of the development of technological capacity. In
the third part the principal commercial and exchange rate policies applied
duringtheperiodareconsidered.Thediscussionofthesocialandagricultural
aspectsarethesubjectofthefourth part.Finally,thefifthpartcontainssome
finalreflectionsonthelegaciesoftheprocess.
Economicgrowth andproductivity
Globalgrowth,investmentandeconomicinstability
TheclassicindicatorsofeconomicgrowthshowthattheChileaneconomydid
not achieve an outstanding performance in the postwar years; Chile lost its
position in relation to other Latin American countries and the developing
world. While the GDP of the region grew at a rate of 5.5 per cent per year in
the first two decades of the postwar period, in Chile on average this rate
reached 4.3percentin 195071, anddropped to2.6 percent in197489 (see
Table5.1).Withthedemographicexpansionof2.3percentinthefirstperiod
and1.6 percentinthelast,thepercapitagrowthratesfellto2and1percent,
respectively. These averages hide enormous instability including years with
positivevariationsoftheGDPofupto10percent,andnegativevaluesupto
15percent.
Table5.1 Chile:rateofgrowthofGDP(averageannualpercentage)
Period ActualGDP
195061 4.0
196171 4.6
197174 0.3
197481 2.6
198189 2.6
198995 6.7
195095 3.8
Sources: Ffrench-DavisandMunoz(1990),andCentralBankofChile.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
116 TheIndustrializationofChile,194082
Table5.2 Chile:Rateofgrossfixedinvestment(averageper
centshareofGDP,1977prices)
Period ShareI/GDP
195060 19.4
196170 20.2
197173 15.9
197481 15.7
198289 15.3
199095 20.8
195095 18.2
These results are associated with a relatively stable rate of formation of fixed
capital throughout the greater part of the years in question up to 1970. With
short-termfluctuations,itremainedwithintherangeof1920percentofGDP,
to fall in the 1970s and 1980s to an average level of 16 per cent (Table 5.2).
Investment is a determining factor in the increase of potential production
capacity.
2
This expanded in a relatively stable manner in the 1950s and
1960s, at rates estimated at between 3.3 and 4.6 per cent. Demand, however,
was more unstable in these decades in response to external events and to
internal policies. The evolution of the use of production capacity became
much more unstable in the 1970s and 1980s. With this, inevitably, there was
a significant gap between actual and potential GDP, which obviously dimin-
ishedproductivityofinvestment(Table5.3).
Thesubperiodofthemostrapidriseinproductioncapacitycorrespondsvery
closelytothepresentdecade.Afterthis,themostdynamicperiodwas196271,
followedbythe1950s.Inthedecadeof1970to1980thegrowthinproduction
capacitywassignificantlylessthaninthetwoprecedingdecades.
Table5.3 Chile:productivityofcapital
Period Marginaloutput- ActualGDP/ Actualoutput-
capitalratio
a
potentialGDP
b
capitalratio
c
195061 0.27 0.96 0.26
196271 0.35 0.98 0.34
197274 0.11 0.96 0.11
197581 0.34 0.90 0.31
198289 0.36 0.88 0.32
a
Themarginalproduct/capitalcoefficientiscalculatedineach
subperiodwiththepeaksadjustedrespectively.
b
RatioofaveragesofactualandpotentialGDP.
c
Theproductof(
a
)times(
b
).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Another permanent feature of Chilean postwar economic development was
itshighmonetaryinstability:averageannualinflationrateswereoftheorderof
2035 per cent, and there were two periods of sharp acceleration in the
middle of the 1950s and at the beginning of the 1970s when there was a
strong populist bias; and three episodes of instability were provoked by the
balance-of-paymentscrisesin1962,1975and1982.
Theinstabilityofthelevelofeconomicactivityhadnegativerepercussionsin
twosenses.Ontheonehand,theunderutilizationofcapacity,whichwasmuch
greater in the two latter decades than in the first two, tended to reduce the
profitabilitybothsocialandmarketofcapital,diminishedinvestmentfund-
ingandworsenedthefinancialsituationonbusinesses.Inthemacroeconomic
sense,besidesdiscouraginginvestment,itreduceditseffectiveproductivity.
InTable5.3twoeffectsarebrokendown:oneonthepotentialproductivity
ofcapitalandanotheronactualproductivityquantifiedinthepeaksandin
the whole sub period respectively.
3
The data would suggest that potential
productivity was relatively similar in 198289, 196271 and 197581, being
slightly higher in the first period. How can this be reconciled with the image
that in the two most recent decades productivity should have been much
greater, due to the greater efficiency derived from the opening up and the
liberalization of the markets? The answer must be that surviving businesses
are more productive, butthat the mortality rate of businesses is greater. Since
weareinterestedinmeasuringtheproductivityofthewholeeconomyandnot
onlyasegmentofit,whenweconsiderthewholeeconomy,potentialproduct-
ivity was only slightly superior in 198289 than in 196271.
4
In the 1980s it
shouldhavebeensomewhatsuperiorto197581duetothedestructionresult-
ing from the liberation of imports and the fact that the exchange rate was
overvaluedinthe1970s(Ffrench-Davis,1986;Vergara,1980).
In the 1960s, excessive protection and administrative obstacles, in spite of
the advances in rationalizing policies, protected the inefficiencies. However,
the greater macroeconomic stability which was achieved contributed to
improving efficiency, to a lessening of the destruction of businesses, and to a
greater concentration of efforts in creating businesses rather than transfers of
existingassets.Italsogeneratedprofilesofdemand whichweremorepredict-
ableand,paradoxically,relativelymorestablepricesthaninthefollowingtwo
decades. Likewise, development was more integrated, which had the effect of
offeringmoreproductiveopportunitiestowidersectionsofsociety.Thiscom-
binationofelementsexplainstheimprovementintheperformanceofproduc-
tivityincomparisonwiththe1950s,andtherelativeclosenesstothepotential
productivityofthe1980s.
On the other hand, greater stability in the use of production capacity was
recordedinthe1960s.Theaverageleveloftherateofusewasalsoatarelatively
higher level. By contrast, the years of great underutilization of capacity have
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
been associated with deliberate or involuntary adjustments immediately after
expansions with fiscal, monetary or balance-of-payments imbalances. This
happened, for example, in the periods 195456, 1959, 197273, 197577 and
198285. Underutilization also intensified when the stabilizing policies relied
on only one or two of the instruments of stabilization policy, sacrificing the
equilibrium of other relevant variables for the sustainability of a comprehen-
sivemacroeconomicequilibrium,forexamplein195962andin197982.
Theindustrialsector
From an industrial perspective, the last 50 years can be divided into two
distinctphrasesofgrowthduringwhichradicallydifferentprojectswerecarried
out.Inthefirstofthesephases,whichbegantowardstheendofthe1930sand
ended with the breakdown of democracy in 1973, the major emphasis was
placed on the development of a new industrial framework, which had been
non-existentuntilthen.Thesecondphasewascharacterizedbytheabandon-
ment of the industrialization plan as such, and the setting up of an open
economy based on incentives for the private sector resulting in a market
whichwouldbefreefrompublicsectormanipulation.
Inthefirststage,firstlyunderthe`natural'protectionprovidedbythebreak-
downofthecommercialordercausedbytheSecondWorldWarandthen,asa
resultofanexplicitindustrialpolicy,averitableproliferationofsmallindustrial
plantsoccurred,concernedprincipallywiththeproductionofbasicconsumer
goodsanddurables;theseweretechnologicallysomewhatbackwardcompared
withbestinternationalpractices.Inspiteofallthis,itwastogiverisetoanovel
developmentwhichwouldinvolveadaptiveengineering:namely,adaptingthe
importedtechnologytothescaleofthenationalmarketandtheavailabilityof
localinputs.
One characteristic of this first phase was the high degree of vertical
integration of production processes, vis-a-vis best international practice,
which was the consequence of the weakness of the industrial set-up into
whichnewfirmsentered.Thisbiasedlearningtowardstheresolutionofbottle-
necks in process engineering, creating an evolutionary path which was com-
pletely different fromtheinternational experience.Inspite of therestrictions
caused by this industrial model, it is important to recognize that this also
reduced commercial and technological dependency on the exterior, allowed
the development of a new industrial culture, until that time non-existent,
createdlinkswithsectorssupplyingrawmaterials,andallowedforthecreation
of a basis for the process of accumulation of human capital in the national
economy.
Incontrast,inthephaseofreformwhichbeganin1974,nationaleconomic
policy was oriented principally towards ensuring price stability and freedom
of markets. Industrial policy, as well as commercial policy, were subordinated
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table5.4 Chile:sectoralgrowthofGDPandshareofindustry
a
Period Agriculture Mining Industry Total IndustrialGDP/
TotalGDP
194046 2.9 2.3 8.4 5.0 0.20
194653 0.9 1.8 6.8 3.3 0.22
195362 3.8 0.9 5.3 4.1 0.25
196271 3.5 4.1 5.4 4.5 0.26
197174 1.3 3.2 1.2 0.2 0.27
197481 1.8 3.4 1.0 2.6 0.22
198189 5.8 3.8 2.8 2.6 0.21
194071 2.3 1.4 6.2 4.2 0.25
197189 4.1 3.7 1.3 2.8 0.22
a
ThesubperiodshavebeenselectedaccordingtothemethodologyofFfrench-Davis
andMunoz(1990),inordertoidentifythe`peaks'intheeconomiccyclewhich
correspondtoafullemploymentofcapacity.
Sources: LenizandRozas(1974);MarcelandMeller(1986).Informationgenerated
fromProgramadeAnalisisdelaDinamicaIndustrial(PADI),EconomicCommissionfor
LatinAmericaandtheCaribbean(ECLAC).
to the objective of macroeconomic òrder'. Consequently, the processes of
commercial opening up and deregulation of the markets of intermediate and
final industrial goods, amongst other measures, were at the heart of the neo-
liberal experiment. This change in policies had a strikingly different impact
depending on the productive sector in question. During the first phase
(194071) the role assumed by the industrial sector as a motor of growth is
clear;ratesofexpansionwereonaverage50percenthigherthanthoseofthe
rest of the economy; industry continued to maintain a leading position until
1971(Table5.4).
Withtheneo-liberalexperience,theindustrialsectorceasedtobethemotor
of growth, experiencing a sharp deterioration particularly during the first
period of the neo-liberal regime during which industrial expansion was nil.
AccordingtoMarcelandMeller (1986,table4), industrialproduction in1980
was6 percentlessthanin1974,whentherecoveryresultedinapeakafterthe
drastic trade liberalization of 197479 and the recession of 1975. In 1981,
aggregate industrial production began to fall again due to the revaluation of
theexchangerate,andagainin1982inresponsetoanewanddeeprecession.
In 1982, manufacturing production was 15 percent less than in 1980. In this
phase,theleadingsectorsofgrowthweretheserviceindustriesandthesectors
which were intensive in natural resources. Leadership appears to have been
concentratedinthefinancialsectorwhichgrew14percentperyearbetween
1974and1981andinimportcommerce16 percentperyear(Ffrench-Davis,
1982,table3).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
An important feature of this process was the apparent trade-off existing
between industrial development and the rest of the productive sectors. The
highgrowthratesintheindustrialsectorwenthand-in-handwithlowratesin
agriculture and mining in the first phase. This was not unrelated to policies
which on the one hand modified the price system for the benefit of the
industrialsector,generatingfierceprotectionofmanufacturing,whichatthat
time produced only for the internal market and on the other hand failed to
protect the exporting sector and, frequently, the substitution of agricultural
imports(seethefollowingsection).Specifically,theindustrialpolicyofthefirst
phase frequently resorted to exchange-rate appreciation to move the
surpluses to the industrial sector this practice wastemporarily suspended in
thesecondhalfofthe1960s.Duringtheneo-liberalphasethederegulationof
foreigntradedrasticallymodifiedrelativepricesandpermittedagreateraccu-
mulation in the sectors exporting natural resources, mining, and agricultural
produce.Inspiteoftheindustrialrecession,manufacturingexportsexpanded
vigorously between 1974 and 1978 (Ffrench-Davis, 1979a). The divergence
between both indicators is explained because `manufacturing' exports were
concentrated in items of a very low industrial value added in comparison
with natural resource content. Two-thirds of manufacturing exports in 1978
corresponded tofishmeal,molybdenumoxide,semi-refinedcopper,cellulose,
bulktimberandpaper.Ontheotherhand,manufacturingexportsimmediately
startedtoslowdownandeventodeclinein1981.Theseriousovervaluationof
theexchangeratewhichoccurredbetween1978and1981wasthedetermining
factor.
In the 1980s, after the sharp drop in 1981 and 1982, the manufacturing
sector experienced a recovery along with GDP. Exports in this sector again
underwent a strong expansion, boosted by a real exchange rate which depre-
ciated130percentbetween1981and1988.Nevertheless,mostoftheexports
continued to be represented by items which were intensive in natural
resources. Towards 1989 copper concentrated, smelted and refined
accounted for 50 per cent of exports; fishmeal, wood, paper and cellulose
together accounted for 15 per cent. By contrast, other more highly manufac-
tured goods on average accounted for 11 per cent; they represented around 2
percentofGDPintheyearofreference.
The different priority which the industrial sector has had during the two
principalhistoricalphasesofChileanpostwardevelopmentwasevidentinthe
evolutionoftheparticipationofindustrialproductsinGDP.Fromaparticipa-
tionof22percentinthe1950s,risingtoamaximumlevelof27percentatthe
beginningofthe1970s,thiscoefficientdroppedthereafterto19percentwith
tradeliberalization,torecoverandslightlysurpass20percentinrecentyears.
Labour productivity in the industrial sector has suffered major fluctuations
overthelast50years.
5
Inspiteofthis,asharpcontrastcanbeseenbetweentwo
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
distinctphases:thephaseofimportsubstitution,whenindustrialproductivity
grew at 5 per cent per annum, mainly concentrated in the 1950s and 1960s;
andthephasecorrespondingtotheneo-liberalexperiment,whenproductivity
waspracticallyatastandstill.Withregardtolabourincome,thesametrendis
reflected, in so far as real average wages between 1970 and 1990 did not
increase,thoughthereweremajorfluctuationsduringthisperiod.
Itispossibletooutlinesomehypothesesonthegrowthoflabourproductiv-
ityduringthemodelofgrowthorientedtowardsthedomesticmarket.Firstly,
there was an acceleration of the learning processes which had begun before.
Secondly,therewasanintensifieduseofcapital,togetherwithgrowingratesof
investment.Ineffect,whilstintheperiod194053therehadbeenagrossrate
ofinvestmentof16 percentofGDPandarateofinvestmentinmachineryand
equipmentof4.1percent,inthefollowingtenyearsthesefigureshadreached
19percentand6 percentrespectively(Hofman,1992).Thirdly,therewasthe
strong performance of internal demand which demonstrated greater stability
thanks to better macroeconomic management. On average, as mentioned
above, effective demand was closer to the production frontier which made
wayforgreaterlevelsofuseofplantcapacity.
Finally, a fourth factor in the growth of value added per worker during this
phaseisrelatedtosuccessivechangesintheproductionstructure.Mostofthe
industrial growth of the 1940s was biased towards industries which were
labour-intensive textiles, footwear, clothing and printing and with small-
scale plants. Subsequently, there was a greater expansion in sectors linked to
thetransportandlight-engineeringindustries,whoseplantswerecharacterized
by their intensive use of design capacity and product engineering, as well as
alsohavingaworkforceskilledinproductionprocessesandproductassembly.
Another sector which expanded greatly at this time was industrial commod-
ities, such as chemicals and petrochemicals, the iron and steel industries and
copperrefining,amongstothers.Theywerecharacterizedbytheirintensiveuse
of raw materials and process engineering, with relatively large plants which
generally served external demand. In brief, physical capital per worker
increased,andtheaverageworkeracquiredagreaterdegreeofhumancapital.
This expansion coincided with important institutional support for the
adaptation of technology and accumulation of human capital. At this time
severalstateinstituteswerecreatedwiththeaimsofadapting,improvingand
spreading new technological knowledge relevant to the different industrial
processes and products, especially in sectors related to raw materials
processing(CIMM,INTEC,INFOR,IFOP).
6
Inadditiontothis,highereducation
technology centres were created with the aim of providing industry with a
growing contingent of better-skilled workers, especially engineers and techni-
ciansfromtheuniversitiesofSantaMar a,Concepcio n,TecnicadelEstadoand
INACAP.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Towards the end of the 1960s the vigour in the growth of industrial value
addedperworkerdeclined.Inpart,thiscanbeassociatedwithaslowerrateof
expansion in internal demand almost one percentage point less and a
reduced use of plant capacity, while the rate of investment in machinery and
equipmentwasmaintainedat5.5percent.Asfortheinternalstructureofthe
sector,therewasasignificantslow-downintherateofgrowthofvalueadded
per worker in industrial activity. The two main sectors affected, and those
whichformedmostoftheindustrialsector,werefoodandindustrialcommod-
itiesparticularlytextiles.Incontrast,therewasagoodrelativeperformancein
thesectorsoftransportandlightengineering.
From1973,Chileunderwentaseriesofradicalchangesinthemanagement
of economic policy, which had a great impact on the production structure.
Market deregulation and greater openness to foreign competition redirected
industry towards specialization in the production of goods intensive in raw
materials and unskilled labour. One way of evaluating the magnitude of the
structural change that these reforms triggered is to consider the trend in the
tradebalancebysector.
Figures 5.1 and 5.2 show the high degree of autarchy before the process of
opening up. There was also a sizeable negative trade balance, which becomes
practically non-existent if smelted and refined copper are included; these
goods represented a large part of Chilean exports during the period. Even
though the net balance was maintained at a low positive level, except during
theyearsofhighgrowthinimports(1981),withtheopeningupoftradeboth
exports and imports grew extensively. In any case, the change in the relative
balances within industry was of great importance. At the beginning of the
1970s the commodities sector stood out as the surplus sector, and one which
8000
6000
4000
2000
0
2000
4000
6000
1970 1973 1976 1979 1982 1985 1988 1991 1994
B
i
l
l
i
o
n

U
S
$

8
5

Exports (exc. copper)
Imports (exc. copper)
Net (exc. copper)
Years
Figure5.1 Chileanbalanceoftradepattern(excludesexportsderivedfromcopper)
Source: InformationgeneratedfromProgramadeAnalisisdelaDinamicaIndustrial(PADI),ECLAC.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
B
i
l
l
i
o
n

U
S
$

8
5

1000
500
0
500
1000
1500
2000
2500
3000
I
II
III
IV
V
1972 1970 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994
Years
Figure 5.2 Chile: net trade balance by subsections (group IVexcludes products derived
fromcopper)
Source: InformationgeneratedfromProgramadeAnalisisdelaDinamicaIndustrial(PADI),ECLAC.
maintainedthischaracteristicalthoughwithoutintensifyingit.Foodreversed
itsbalancefromadeficittoasurplus,whereastheremainingsectors,especially
those associated with light-engineering and transport, increased their trade
deficits.
Withregardtothetrendinproductivity,theearly1970sshowed,overall,an
industrial sectoratastandstillin thevalueadded per worker whichlastedfor
more than ten years, and notable falls in the cases of transport and light
engineering.Theonlysectorwithapositivetrendwasindustrialcommodities,
whichincreasinglycametoincludesectorsinvolvedinprocessingrawmateri-
als, especially copper smelting, fishmeal and pulp for paper. Furthermore, in
thisnewsceneofgreaterexposuretointernationalcompetition,thedifferences
in the relative costsvis-a-visthe rest of the world were decisive factors (Tables
5.5and5.6).
Comparing the value added per worker in Chile with representative values
internationally,forexampleintheUSA,itcanbeseenthatalthoughasof1970
allindustrialsectorsseemedtobefallingbehind,therelativegapswerediffer-
ent. Whilst the commodities sector was at a level of product per worker of
nearly 50 per cent of best practice, the automobile and light engineering
industries only reached 10 per cent. It was not, therefore, surprising that
thesesectorswereliterallysweptawaybygreaterinternationalcompetition.
In addition to the force of competition, there were deplorable macroeco-
nomic conditions at this time, with the lowest rates of growth in GDP on
record and a rate of investment which throughout the whole period only
reached 15.5 per cent. This period also coincided with the disbanding of the
networkofinstitutionalsupportforindustry,andthefreezingofitsbudgets.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table5.5 Chile:averagegrowthrateofvalueaddedper
industrialworker,194089
Sector
a
Total
Period I II III IV V
194046
b
2.9 2.9 2.2 8.4 5.1 4.3
194653 2.5 0.1 0.3 6.7 2.2 0.1
195362 9.2 8.8 8.4 10.6 8.6 9.0
196266 8.7 7.7 4.0 2.5 3.1 3.8
197174 0.7 6.6 3.9 1.2 4.2 2.1
197481 4.8 8.1 7.2 1.4 10.5 4.4
198189 2.7 8.2 4.9 3.4 5.1 2.1
a
Definitionofsectors:
Ilightengineeringindustry:381,382,383,385
IItransport:384
IIIfood,drinkandtobacco:311,313,314
IVindustrialcommodities:341,351,354,355,356,371,372
Vtraditionalindustries:321,322,323,324,331,332,352,361,362,
369,390
b
Intheperiod194046,thelightengineeringindustryfigureswere
incorporatedintoasinglesector,sothatsectorsIandIIcouldnotbe
differentiated.
Sources: LenizandRozas(1974)for194066;ECLACfor197089.
Table5.6 Chile:valueaddedperworkerrelativetotheUSA,197094
InitialRelative FinalRelative %variation
efficiency efficiency
Sector (197094)
I 12.7 10.7 15:9
II 11.8 8.2 30:6
III 16.1 10.3 36:2
IV 53.8 30.0 44:2
V 14.1 14.2 0.5
Total 20.9 16.2 22:5
Source: InformationgeneratedfromProgramadeAnalisisdelaDinamica
Industrial(PADI),EconomicCommissionforLatinAmericaandtheCaribbean
(ECLAC).
Theroleofthestateinindustrialization
Precedentsforstateinterventioninthe192030s
The increasing state presence in the Chilean economy began to be felt in the
1920s,althoughitwasafterthe1930crisisthatthemodeloftheintervention-
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ist state characterized Chilean development until the mid-1970s. The post-
crisis international atmosphere was favourable for greater state activism. In
the USA as much as in Europe the problem of unemployment and instability
arose,andnewapproachesandpoliciesbegantoemergewhichfavouredturn-
ing away from the laissez-faire system prevalent until then.
7
But the internal
situationbecamecriticalduetotheworldDepressionandthestatehadtotake
onaleadingroleinordertorelievesomewhattheeffectsofthecrisis.
However,itwouldbeverysimplistictoexplainthechangeinthestate'srole
onlyintermsoftheinfluenceofexternalfactors.WithinChile'seconomyand
society several factors occurred which determined the course taken. Firstly,
there was the so-called òligarchic crisis' which brought Arturo Alessandri to
power. His programme was based on a calling into question of the oligarchic
regimetheannouncementofanewPoliticalConstitutionin1925.A`meso-
cratic' alliance developed, based especially on middle-class sectors. This was a
period of expansion for popular and anti-establishment organizations: the
popular press developed, the left-wing parties and even the military demon-
stratedagainstoligarchicpower.
Secondly, the model of development in force favoured a sharp increase in
public revenue. Between 188185 and 192125 taxation increased 3.2 times
expressedinnominaldollars,thatisat4.8percentperyear,comparedwitha
growthinGDPof2.3percentandinexportsof2.7percentannually(Meller,
1996, tables 1.2 and 4.1). This made possible increases in expenditure on
infrastructure and on welfare benefits (Arellano, 1985). But the state's greater
economic importance was demonstrated also by institutional changes. The
mostimportantofthesewasthecreationoftheCentralBank,aftermanyfailed
attemptstoestablishamonetaryauthoritywhichwouldregulatethebanking
and monetary system. Various other relevant institutions were also created
during this period to modernize the state: the Taxation Board, the General
Audit Office of the Republic, the Banking Regulatory Body, and financial
agenciesforvarioussectors.
Thirdly, the state technocracy was strengthened, based especially on engi-
neers who were very active in promoting public works and in formulating
economic policies (Ibaez, 1983). Under the leadership of the Minister of n
Finance, Pablo Ram rez (192729), reforms were introduced giving the state
theroleofactivelyandpragmaticallydirectingnationaldevelopment.Agroup
ofengineerswhohadbeenprominentintheStateRailCompanybecamepart
ofthemachineryofstatemanagementandbegantodesignavisionforstrate-
gicnationaldevelopment.
8
Asaconsequenceofthe1930crisis,thestatewasforcedtoassumeaveryactive
role in stabilization. The Chilean economy was one of the most affected on a
worldwidelevelbytheimpactofthecrisis;thenitratesexportindustrycrumbled
andtherewasmassiveunemployment;taxrevenuesdropped,equilibriuminthe
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
balanceofpaymentsbecameunsustainableandtheCentralBankwasforcedto
suspendconvertibilityanddevaluethepeso.Thegovernmentpushedthrough
aninitiativetostimulateconstructioninordertocreatejobs,andexchange-rate
controltoadministerscarcesuppliesofforeigncurrency(Marfan,1984).
The private sector organized and established the Confederation of Produc-
tion and Trade, made up of the National Agriculture Board, the Society for
IndustrialDevelopmentandtheBoardofTrade.Oneofitsmaindemandswas
thecreationofaNationalEconomicAdvisoryBody,withprivateparticipation,
which would coordinate economic policy and production development. But
early in 1939, because of the earthquake which devastated the central area of
Chile, the government proposed the formation of CORFO Corporation of
Production Development with the objective of designing a programme
ofnationaleconomicdevelopment.Theprivateindustrialsectorreceivedthis
proposal with interest, although with some reservations as to the form of
finance fear of tax rises and the entrepreneurial role which the state
wouldhave.
CreationofCORFOandtheinstitutionalframeworkfortheproductive
sector
At the end of 1938 a centreleft coalition government was elected called the
PopularFront,whichmarkedthestartofanewerainChilebasedonthestate's
activeroleaspromoterofthecountry'sindustrialization.Themaininstrument
toimplementthisobjectivewasCORFO.
9
Itsfirstinitiativesweretheformula-
tionofplansfortheagriculture,industryandminingsectors.Externalfinance
was arranged with Eximbank, which had just been set up by the US govern-
ment, and this enabled the financing of public investment programmes and
medium-termloansfortheprivatesector.
Throughoutthe1940sand1950sCORFOfulfilledthreeimportantfunctions:
development of public and private investment, organization of new state
companies in strategic economic areas and the development of technological
research.
10
Also, great importance was given to the preparation of systematic
economic information which the government had lacked. Thus, the first
national accounts of Chile were recorded: a base of economic information
whichwouldpermitsomeattemptsatplanninginthe1960s(Munoz,1993).
By the early 1960s, and under the auspices of the Alliance for Progress
brought about by the government of President Kennedy, CORFO formulated
the first national development plan, designed for the decade of the 1960s.
Although this plan lacked practical application, the work involved in its pre-
parationandinformation-gatheringwasavaluablebyproductwhichledtoan
improvementinnationalaccountingandinthebalance-of-paymentsrecords,
and provided better support for various sectoral plans. In the 1960s CORFO's
workcontinuedwiththecreationofstate-controlledandmixedcompaniesin
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the new cutting-edge sectors, such as petrochemicals, telecommunications,
informationtechnologyandtourism.Thisperiodculminatedinthedesignof
a new industrial strategy, which sought to rectify many of the distortions in
industrial policy, and to further facilitate Chile's participation in the interna-
tionalsphere.
In 1968 the Vice-President of CORFO organized a wide debate between the
government authorities and businessmen and professionals, with the aim of
achieving a new design for industrial strategy (CORFO, 1969 and 1970). The
topicsunderdiscussionwerethemainweaknessesanddistortionsinprevious
policies,inparticularthedistortionsderivedfromthehighlevelofprotection-
ism,aswellastheinstitutionalweaknessesinthepublic-sectorcontrolofstate
companies.
11
Theantagonismbetweenthestateandtheprivatesector wasto
givewaytoavisionwheretheywouldcomplementeachother.Inthiswaythey
wouldtogetherreachtheobjectivewhichwasseenasacentralchallengetothe
Chilean economy: effective insertion into the international economy. In this
respectitwasrecognizedthatduetothesmallscaleofitseconomy,Chilecould
notcontinuewithanindustrializationpolicydirectedexclusivelytowardsthe
internal market. For this reason, the strategy proposed an intensification of
industrial exports.
12
However, it was not taken for granted that external
demand should be the only factor to stimulate industrial growth, rather that
internal demand should be stimulated by a more effective redistribution of
income. But this would imply that the supply of industrial goods would
becomemore accessibletomiddle andlow incomegroups.Itwasarguedthat
thehighrelativepriceofindustrialgoods,whichwereprotectedbytariffsand
internal monopolies, had had a detrimental effect on the demand for
those goods in low-income and rural sectors. Therefore, a greater degree of
rationalization in the industrial structure, a more accelerated rate of growth,
and greater competition due to a gradual process of opening-up inter-
nationally were to promote a lower price of consumer goods to be found in
the typical shopping basket. However, the new focus was not applied after
1970, since the new government, Popular Unity, switched the direction of
economicpolicytowardsanincreaseinstatecontrol.Industrialpolicybecame
far less important in the national political debate and in the debate on the
role of the state in the economy. The atmosphere was extremely polarized
and the future of the national sociopolitical system was in the balance. This
would determine daily life for Chilean people a subject which is dealt
withlater.
CORFO now assumed the task of leading and implementing the process of
nationalization of large companies, and thereafter of running what was
referredtoastheàreaofsocialproperty'.Itsimportancecanbeseenthrough
theparticipationofstate-controlledcompaniesinGDP.In1973thisroseto39
percent, compared to 14 percent in 1965 (Hachette and Luders, 1992). After
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
1973 and with the objective of economic freedom from the military govern-
ment,CORFOconverteditsroletothatofconductingtheprivatizationprocess
ofstatecompanies,whichfinishedattheendofthe1980s.
Innovations,learninganddevelopmentoftechnologicalcapacity
TheinstitutionalframeworkofsupportforResearchandDevelopment(R&D)
went through several stages. It began with a very slow pattern of change and
wasdominatedbytheuniversitiesforpracticallyonehundredyears.Fromthe
1950s, under the protection of the import-substituting model, until the early
1970s, there was a significant increase in the number of these institutions, as
muchfromwithintheuniversitiesasthroughthecreationofalargenumberof
public-sector centres related to applied research and experimental develop-
ment. In the 1970s and coinciding with the change towards a free market
model which favoured market allocation of resources, there was a noticeable
decreaseinthestate'srole,asmuchinthecreationofnewinstitutionsasinthe
public financing of this sector. It is worth mentioning that this decrease also
coincided with the privatization of the main state companies except
CODELCO, ENAMI and ENAP
13
and with this there was a loss of an instru-
ment of technological policy central to the promotion of the most technolo-
gically dynamic sectors. The new institutional phase which began from 1974
was accompanied by a drop in expenditure on R & D; during this period the
creation of new public bodies was halted and there were attempts to achieve
self-financingstatusintheexistingones.However,thesituationchangedagain
after the creation of competitive grants that became available from 1980
onwards.
The creation of CORFO in 1939 was a starting point in the development of
institutions. In 1952 the Service of Technical Cooperation (SERCOTEC) was
created, geared towards technological assistance for the small and medium-
sized segments of industry. This was followed by the Institute for the
DevelopmentofAgriculture(INDAP,1962),whichwasdesignedtofinancethe
adoption of new technologies in the agricultural sector, particularly for the
smallproducerswhowerebeneficiariesoftheAgrarianReform,theInstituteof
Fisheries Development (IFOP, 1963), the Institute of Agrarian Research (INIA,
1964) and the Forestry Institute (INFOR, 1965). From then on, efforts were
aimedatmeetingtheneedsofmoretechnologicallycomplexareasofproduc-
tion.Therefore,in1965theNuclearEnergyCommissionofChile(CChEN)was
created,theInstituteofTechnologicalResearch(INTEC)in1968,theCentrefor
MiningandMetallurgy(CIMM)in1970,andfinally,theNationalInstitutefor
Normalization(INN)in1973.Asregardsthecoordinationandfinanceofscien-
tificandtechnologicalactivities,theNationalBoardofScientificResearchand
Technology was created (CONICYT). This institution's most important pro-
gramme was the grants awarded to basic research projects, mainly selected
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table5.7 Chile:averagesharesofexpenditureonresearchanddevelopmentby
institutionalsectors,196592(percentages)
Years Universities Institutes Companies Funds Other R&D/GDP
196571 68.1 26.9 n/a 5.0 n/a 0.41
197278 48.3 51.7 n/a n/a n/a 0.42
197981 41.3 46.3 3.1 1.2 8.1 0.36
198289 43.4 33.1 7.7 7.3 8.5 0.50
199092 36.6 31.6 10.6 14.7 6.5 0.62
Source: CONICYT(1993).
fromamongsttheuniversities;becauseofthis,CONICYTwasfromthebegin-
ning more orientedtowards thedevelopment of basicscience rather thanthe
productive application of its knowledge. Thus, the institutional supply of
technologicalservicesatanationallevelwasbegun.
Thisexpansioninthenumberofinstitutionswasaccompaniedbyasizeable
growthinexpenditureonR&Datanationallevelexpenditurerosefrom0.07
percentofGDPin1961to0.41percentin196571(Table5.7).Asregardsits
composition, there was an increase in the participation of state institutions
risingfromlessthan20percentoftotalexpenditureinthemid-1960s,tomore
than 60 per cent by 1973 which was accompanied by a reduction in the
participationofuniversities.
Towardsthemid-1970s,thechangeofgovernmentandtheresultingsudden
changeintheeconomicsituationledtochangesinpublicspendingonR&D.
Thepromulgationofanewlawofuniversityself-financemeantthattheirfiscal
contribution was frozen, as well as the CONICYT fund; this in turn caused a
sharp fall in this sector's participation in R & D. Public institutes were also
subjecttothispolicyofself-finance,buttheyobtainednewsourcesoffinance
ofa`nonfiscal'origin(CPU,1987p.108onwards).
Tradepoliciesforindustrialization
Untilthemid-1960s,Chileantradepolicydidnotdifferinmanyrespectsfrom
those of many other countries in the region.
14
Effective protection was wide-
spread; but it was not systematic or deliberate it was largely a response to
restrictionswhicharosefrommomentsofcrisisinthebalanceofpaymentsand
then continued in the medium term. In addition to these restrictions, which
protected producers of import substitutes and discouraged exporters, there
were ad hoc protection measures promoted by the government or by private
interests.Partoftheimpactonexporterswascompensatedforbyliberalization
measures of the imported capital and intermediate goods used by the export
producers,especiallyafter1958.Theexchangeratepolicy,ingeneral,operated
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
withoneormultiplefixedrateswhichgraduallylostpurchasingpowerinthe
faceofadomesticinflationhigherthantheinternationalrates.Letusnowlook
atthisprocess.
Theinitialperiod
Around 1929 there was, generally, wide foreign trade freedom in Chile. Cus-
toms duties were low and their objective was to collect funds. However, the
mainsourceofincomeavailabletotheExchequer wasanothertaxonforeign
trade; this consisted of a duty on nitrate exports equivalent to 32 per cent of
theirvalue(Cabezo n,1971,p.4),therebyexploitingChile'smonopolyposition
intheworldmarket.
Following the crisis, controls on foreign exchange and imports began to be
widespread. Onefurther important aspect inthenew frameworkwasthat the
monetarydisciplinewhichthegoldstandardhadimposedhadnowgone,but
also exchange rate flexibility had arrived. The world crisis which deeply
affected Chile provoked the application of policies which, either consciously
or subconsciously, boosted import substitution. Nevertheless, even without
these policies, some import substitution occurred; between 1932 and 1935
the exchange rate, with regard to sterling, was devalued by 300 per cent and
importsfellfrom31percentofGDPin1929toonly7percentin1932(Munoz,
1968). Furthermore, imports of consumer goods fell 82 per cent, decreasing
from52percentoftotalimportsin1928to38percentby1932.Theimport-
substitutionpolicies,aresultoftheGreatDepression,werestrengthenedwith
therisetopowerofthePopularFrontcoalitionin1938,andreinforcedbythe
obstacles to foreign trade imposed by the world powers during the Second
WorldWar.
After 1945, foreign trade policy mainly responded to short-term crisis, in
a framework of rigid exchange rates and increasing inflation. The high level
of protection, granted indiscriminately to industry, was notoriously
inflexible downwards; its level was altered by a periodic tightening of
restrictions in the face of successive balance-of-payments crises. These restric-
tions were at least partially still in place after each difficulty had been over-
come.
In the 1950s, foreign trade was controlled by the Board of Foreign Trade
(CONDECOR), an autonomous institution which had been redesigned in
1950 and invested with full powers to ban imports or exports. The rate of
exchangetowhichimportsandexportsweresubjectwasfixedbytheMinistry
of Economy based on a Central Bank report and on CONDECOR, or, altern-
atively, by the `market'. In other words, this gave rise to a coexistence of free
exchangerates with asystemof multiplerates.By 1952there wasasystemof
fivemultipleexchangeratesandthreefreeexchangerates,andvariouscombi-
nations of both (Central Bank, 1952, pp. 3746). Import duties consisted
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
mainly of very heterogeneous tariffs and of an admission tax with three cat-
egoriesofrates.Infact,onlyone-thirdofthemoneycollectedinimporttaxes
correspondedtotariffs.Theaveragerateofmoneycollectedinimporttaxes,at
26 percent,waslessthanthetotalnominalrateduetonumerousexemptions:
public sector imports, bilateral trade agreements, and various private sector
imports.
Themaincharacteristicofforeigntradepolicyinthe195255periodwasthe
persistent instability of the rules faced by importers and exporters: a sizeable
fraction of trade was carried out under bilateral agreements, and there was a
general system of quotas and quantitative controls, completely lacking any
formofcoordination.Asaresultoftheprevalentappreciatedratesofexchange,
there was excessive demand in almost all areas of imports. As there was no
mechanismforfixingorallocatingquotas,thesysteminevitablyfellunderthe
influence of particular interests which made it a victim of favouritism and
privileges. Furthermore, the system was characterized by its notorious ineffi-
ciency,creatingfrequentshortagesinrawmaterialsandintermediategoodsfor
nationalindustry.
Byitsverynature,foreigntradepolicydidnotleadtoadeficitonthebalance
of payments nor to a large foreign debt. On the contrary, the effects of this
policyweremadevisiblemainlybyveryunstablecontributionsonthepartof
foreign trade to the internal supply of goods resulting in shortages and
speculationandbyadecreaseintheproportionofimportsofcapitalgoods.
The balance of payments tended to remain at a forced equilibrium, with a
variablelevelmainlydependingoncopperexports.
Giventheincoherenceofthepolicies,theycannotseriouslybecataloguedas
anISIstrategy.Theinstabilityoftradeprotectionmarginsandthelevelsofreal
exchange rates conspired against companies' ability to evaluate projects and
allocate investment resources. Nevertheless, there were sectors isolated from
that unstable situation, under absolute protection, which therefore had the
internalmarkettothemselves.However,macroeconomicinstability,associated
withthetransferoffrequentexternalshockstotheinternalmarket,generated
highlevelsofinstabilityofdemandforsuchimportsubstitutes.
Earlyattemptsatliberalization
During the 1950s the emerging theory of development began to strongly
emphasize the importance of capital formation in order to accelerate growth.
Anibal Pinto, in his well-known book on Chilean economic development,
sharply observed that in Chile a myth had been created of the `tremendous
effort'ofindustrialization:
Anyonewouldsay,inalltruth,thatthecountryhassurpassedthesacrifices
of the Russian people and the tenacious Nordic countries, who have
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
managed (byreason orforce) tosetasidequotaswhichin someyearshave
exceeded30percentofcurrentincome... Inshort,oneofthefundamental
weaknesses of policy has been its inability to extract from the population,
particularly from those sectors which have above-average incomes, a suffi-
cient quota of resources necessary to accelerate growth and create a more
solidandefficientbasefortheproductivesystem.
(Pinto,1973,pp.1878)
Forhispart,NicholasKaldorhadpointedtothelowrateofinvestmentasoneof
thelong-termproblemsintheeconomicdevelopmentofthecountry.Kaldor's
diagnosiswasthatthecauseofthelowrateofinvestmenthadtobeexplained
by the very unequal income distribution and by the low propensity amongst
the capitalist class to save. Unlike the mature industrial nations, in Chile the
business class designated an excessively high proportion of their income to
consumption which diverted resources from capital formation (a more recent
discussionofthisargumentcanbefoundinMarcelandPalma,1990).
Chilemadeanefforttoincreaseitsrateofinvestmentduringthe1940s.But
inthefollowingdecadethattendencycametoastandstillandevenfellback.
Variousfactorscoincidedtocreateanatmosphereunlikelytogenerateinvest-
ment.Ontheonehand,politicswastakingamorepopulistdirection,stimul-
ating a widespread expenditure policy, especially social, which was not
accompanied by finance. Arellano (1985) has shown that between 1935 and
1955 public social expenditure in Chile multiplied three times, in real terms.
This increase in social expenditure required an increase in public revenues,
which were mainly achieved by social security contributions. As a proportion
of GDP these contributions grew from 2.8 per cent in 1935 to 5.9 per cent in
1956 (Arellano, 1985). However, businessmen considered these increases in
taxation as a disincentive to saving and private investment. On the other
hand, inflationary pressures became more acute from 1953, reaching a max-
imumof84percentin1955.Thisatmospherewascongenialforspeculationto
thedetrimentofproductiveinvestment.Inaddition,uncertaintyincreasedina
more belligerent social climate, as a result of union demands claiming real
wagesincreasesinconditionsofrapidinflation.
There were two attempts at moderate trade reforms, one in 195658 and
anotherin195961(Ffrench-Davis,1973).Inthefirstcase,headedbyaforeign
technical mission (KleinSaks), a certain degree of normalization of foreign
trade was achieved and this meant a relative positive increase in imports of
capitalgoods.Quantitativerestrictionswerereplacedbypricerestrictions,such
as the exchange rate and import deposit accounts. However, the use of these
instrumentswasnotintensiveenoughtopreventasubstantialriseinimports.
Asaresult,theforeigntradedeficitledtothefirstsignsofwhatwastobecomea
large foreign debt. In contrast to previous experience, balance-of-payments
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
difficulties meant further indebtedness and new import restrictions. In 1957,
thecurrent accountdeficit representednearlyaquarterofall exports,absorb-
ingaccumulatedreservesinitsstride.Notuntil1958,andafterintensiveuseof
further restrictions import deposit accounts at prohibitive levels in many
cases was it eventually possible to reduce foreign buying to levels more
compatible with those of Chilean export potential. The import deposit
accounts were used in such a way that import restrictions once again became
irregularandunstableforalargepartofChileantrade.
A new episode in liberalization was marked by the arrival to power of the
Jorge Alessandri administration. The liberalization movement, which had
begun in April 1956 and had been set back by the balance-of-payments diffi-
cultiesin1957and1958,wasextensivelywidenedduring1959.However,there
wasasharpdifferenceinoneveryimportantaspect.Insofarastheprogramme
put in place by the KleinSaks Mission hoped eventually to achieve a freely
fluctuatingexchangerate,theAlessandriprogrammewasbuiltaroundanom-
inal fixed exchange rate, as the anchor of price stability. The banking and
capital markets for exchange transactions were amalgamated and access to
the exchange market was liberalized. Because the government trusted that
capital taken abroad in the past would now return, foreign currency bank
accountswereencouragedwithintheChileanbankingsystem.
The free import of practically all types ofgood was made law. Nevertheless,
mostofthemerchandisewhichhadpreviouslybeenbannedwasnowaffected
byapriorimportdepositaccountof5000percent.Thegovernmentauthorized
the setting-up of prior import deposits in currency, as well as in dollars and
short-termdollarbondsissuedbytheTreasury.Aswastobeexpected,withina
fewdaysmostoftheimportdepositaccountsincurrencywereconvertedinto
dollarbonds.
Between 1959 and 1961, the rates and rules governing import deposit
accounts underwent continual modifications, clearly oriented towards trade
liberalization.Theauthoritytoreducecustomsdutieswasalsousedintensively.
The demand for imports was rapidly growing, whereas exports came to a
standstill from 1960, after an initial rise. In 1961 the current account deficit
representedalmost70percentofexportsforthatyear;furthermore,therewas
intensivedollarizationwithasignificantriseinforeigncurrencydeposits.
Policies adopted during the period 195961 quickly relieved the import
restrictions so that Chile experienced greater freedom in international trade
thaninanyotherperiodsince1929.Theexchangeratewasgivenakeyrolein
generating expectations regarding price stability. Henceforth, the govern-
ment'sprogrammewouldbecloselylinkedtotheinvariabilityoftheexchange
rate. In January 1959 the rate of exchange was fixed at 1.05 escudos per US
dollar the highest real level in the whole period 195270. As a result of an
increaseof35percentininternalpricesin1959,therealvalueoftheexchange
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
rate nevertheless fell, to below any of the levels reached after the reform of
1956.Thenaturalconsequenceoftheapplicationofapolicywithadeclining
realexchangerate,greatreductionsinquantitativerestrictionsandfreeaccess
to the foreign exchange market, was a considerable rise in imports, massive
capitaloutflowsandalargebalance-of-paymentsdeficit.Thegovernmentwas
slow toreact totheadverseconditions, actingonly whenitsforeigncurrency
reservesranout,inDecember1961.
Afterthreeyearsofthisexperiment,the1959foreigntradepolicyresultedin
theexactoppositeofitsdeclaredobjectives.Theresultwastheexistenceoftwo
currency markets, control over access to trade transactions, numerous import
restrictions, resumption of earlier dollar bond deposits, and a proliferation of
bureaucraticcontrols.
Analternativeroute,196570
Intheperiodwhichfollowedthe1962crisis,theaimofmostpolicydesignwas
toconfronttheintensecrisiscausedbythepolicyadoptedintheyears1959
61. A large part of the problem was caused by red-tape restrictions and the
higherimportcostspaidforwithdollarbonddeposits.ThefirstpartoftheFrei
administration, 196567, was characterized by a gradual liberalization of
imports of intermediate products and capital goods and the end of red tape.
Thesecondpart,196870,ontheotherhand,wascharacterizedbyapersistent
rationalizationofthepackageofforeigntraderestrictions,andbyattemptsto
harmonize the foreign sector with industrial development policy. In spite of
this split, there was a new exchange rate policy throughout the 1960s. The
`programmed or crawling peg exchange rate' policy was the most noticeable
featureoftheforeignsector.
TheprogrammedexchangeratepolicystartedinApril1965andlasteduntil
1970 (Ffrench-Davis, 1964 and Carrasco and Ffrench-Davis, 1969). The new
policy covered both the financial and trading rates, both of which were peri-
odically fixed by the Central Bank which tried to maintain real levels which
were more or less stable depending on the medium-term outlook forecast for
Chilean foreign trade. The new policy tried to do away with the frequent
alterations in the price relationship between national and foreign goods. It
was hoped that this policy would mitigate the fluctuations in foreign trade,
avoid speculative gains which had traditionally been favoured by policy
instability, and stimulate production of exports and import substitutes more
effectively and efficiently. From April 1965, the real exchange rate rose con-
tinually,andtowardstheendof1968itreachedalevelwhichwas19percent
higher than the April 1965 value without prompting speculation, changes in
economicactivityorpublicunrest.Oneofthemainvirtueswasprobablythat
theexchangeratepolicyadoptedallowedtheeconomicauthoritiestorespond
opportunely.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
In the years 196668 there were elements of a policy of export incentives.
The government had expressed in its programme its intention to diversify
foreign trade markets as a way to achieve greater economic stability. During
1966anexportincentivelawwaspassedwhichextendedthebenefitsconceded
undertheprevioustwogovernments'legislation.Thislawgrantedataxreturn
ordrawbackofupto30percentofthevalueofexports.Theuseofthistoolasa
subsidy for exporters was widespread throughout the years 196770, covering
almostallexportsexceptironandcoppermining.
By 1969, three-quarters of the huge tax payouts in drawback were closely
linked to the exporter's inefficiency, in the sense that the demonstration that
his costs had risen constituted an acceptable reason for increasing the
15
percentage. In this sense, the drawback policy was similar to the traditional
policyofimportsubstitution,onlynowwithanoticeablybetterexchangerate
policy. A serious effort was made to widen export markets, to disseminate
information amongst exporters and to design standardization, quality and
other similar processes. Furthermore, with the help of the new exchange rate
policythereslowly began toemerge an èxport mentality'in acountrywhere
during previous decades all theefforts hadbeenput into importsubstitution.
The volume of non-traditional exports grew 16 per cent between 1965 and
1970.
As abasic component of foreignpolicy,Chile triedto reinforceits relations
with Latin American countries. As is widely known, during the whole period
19641970theLatinAmericanFreeTradeAssociation(ALALC)waspractically
atastandstill.However,withregardtofinancialintegrationtofacilitateintra-
regional trade, this was a highly active period for ALALC member countries.
The Andean Pact, signed in 1969, grouped together those countries who
wanted to advance at a faster rate in the process of trade integration and
harmonizationofeconomicpolicies.
The general tendency of the new trade policy was to reduce the spread of
effective protection in an attempt to reach a level just over the minimum
necessarytoprotectactivitiessubstitutingforimports.Fromthenon,apolicy
of gradual reduction was followed, accompanied by an industrial strategy of
selective development. This contrasted with the policy of producing a bit of
everything.Thepolicywasgearedtowardsbothhomeandforeignmarkets,and
hadtobecoordinatedinthefaceofacalendarofagreementsscheduledinthe
PactwhichtheAndeanMarketcreated(Ossa,1971).
Clearly,theshort-termeffectsonthelevelofimportsarenottheonlyfactors
importantinmeasuringtheefficacyofthenewpolicy.Theeffectonindustrial
productivityandontheefficiencyoftheallocationofresourcesisakeyissue.
Inthisrespect,theobjectivewastofixtariffrateswhichwere`sufficientlylow'
so as to convince home producers that they should now face up to foreign
competition and improve productivity, and `sufficiently high' to as to avoid
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
bankruptcies and unemployment. Over as long a time period as possible,
the rates were to be reduced gradually in a process which was to last roughly
five years, while resources were moved from industries in decline to growth
areas.
In the presidential elections of 1970 the official candidate lost. The new
authoritiessuspendedthereformsandreintroducedmoreextremeversionsof
thepoliciesofinterventionandchangewhichwerecharacteristicofthe1952
55 period. Inparticular, asharp rise in inflation caused a noticeable apprecia-
tion in the real exchange rate. This was followed by the creation of multiple
rates, which widened in their spread more than ever before. High inflation
madethedifference.
Theliberalizationprocess,197382
16
The central element of trade policy was the sharp reduction in protection on
whichtheproducersofimportsubstitutesreliedatthestartofthenewregime.
The rapid liberalization process, which began in 1973 and officially ended in
June1979,entailedanotablechangeincomparativeadvantageasitmodified
the composition of effective protection as much as its average level. Overall,
the changes implied a drastic reduction in the effective protection of both
intermediateandconsumermanufactures.
Thegoalsoftheliberalizationprocessunderwentsignificantchangesduring
itsapplication.Whatinitiallyseemedtobeamoderatereform,withmaximum
tariffs of 60 per cent and a depreciating real rate of exchange, ended up as a
uniformtariffof10percentwithanappreciatingexchangerate.Overtheyears
from 1973 onwards, official spokesmen repeatedly underlined that the real
exchangerate,giveninpesosperdollar,woulddevaluealongwithreductions
in effective tariff protection. However, in practice and especially given the
presence of large movements in capital, the goalposts moved significantly in
relationtothestatedaim.
Duringtheperiodsofgreatesttariffreduction, therealexchangeratefellin
parallel. Table 5.8 displays information about the evolution of the total ex-
customscostperUSdollarofCIF-importedgoodsformerchandiseaffectedby
themaximumandaveragetariff,onselecteddatesaccordingtothechangein
the cost of imports and according to the relation between the following two
elements:theexchangerateandnominaltariffprotection.Accordingtothese
criteria,fourphasescanbedefined.
Thetariffreductionsmadeinphase1,whichcoverstheperiodfromtheend
of1973untilApril1975,werecarriedoutunderrelativelyhighexchangerates;
furthermore, a large part of the tariff reduction corresponded to levies which
were redundant, since they were mostly levels of protection which were not
used or excessive. On the other hand, as shown by massive liberalization
experiments in other countries, non-traditional imports grow, but with lags,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table5.8 Chile:costperdollarofimports,inrealterms(1977pesosperUSdollarofthe
sameyear)
Nominaltariffs Importcost
(percentages) perdollar
Dates Real Percentage Max. Ave. Max. Ave.
(month/year) exchange variation
rate perphase
(1) (2) (3) (4) (5) (6)
10/1973 20.3 .. 220 94 65.1 39.5
04/1975 34.0 67.2 120 52 74.8 51.7
07/1977 20.6 39.4 45 22 29.9 25.2
06/1979 23.5 14.1 10 10 25.9 25.9
06/1982 17.8 24.3 10 10 19.6 19.6
Notes: Incolumn(1)thenominalpriceofthedollarwasdeflatedbytheconsumerpriceindex(CPI)
andinflatedbyanexternalpriceindexrelevantforChile.ThisisaweightedindexexpressedinUS
dollarsfromtheindicesoftheunitaryvalueofexportsoftheUSA,France,Japan,theUKand
Germany.TheweightscorrespondtotheshareofeachcountryinChile'sforeigntrade(exportsplus
imports).Columns(5)and(6)wereobtainedbymultiplyingcolumn(1)by[1(3)/100]and[1(4)/
100],respectively.
Sources: CentralBankofChile,MonthlyBulletin,variouseditions;CortazarandMarshall(1980);
Ffrench-Davis,LeivaandMadrid(1991),table4.
since they require the creation of marketing channels basic information on
alternative supplies, selling, technical services. During this phase the real
exchange rate accelerated, along with a crisis in the balance of payments; it
easily offset the liberalizing effects of the lowest nominal tariff. The average
cost per US dollar of imports by the end of the phase was 30 per cent
higher than at the beginning. For this reason, the process was mainly con-
cerned with rationalization during this phase, generally decreasing the
high degree of spread in effective protection without registering a noticeable
impact on the production of import substitutes. The liberalization of
imports and the increase in exchange rates did have a positive impact on
exports. Given that before liberalization had begun, exporters had at their
disposition free-trade zones for their imported components, it was the
exchange rate policy which had the greatest effect in this phase (Ffrench-
Davis,1979b).
The situation faced by imported goods changed considerably in phase 2;
namely, between April 1975 and mid-1977. The reductions in nominal
protection were more substantial: the average decreased from 52 to 22 per
cent,andthemaximumratefrom120to45percent,whiletherealexchange
rate decreased by 39 per cent, strongly reinforcing the first effect. In parallel,
and despite a recuperation, aggregate demand was still very depressed due to
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
therecessionof1975.Bothofthesestrongimpactsgavelittleopportunityfor
a gradual adjustment in economic activity, because they were unexpected,
especially in the sense that they contradicted the reiterated official
declarations thatthe exchange rate management would above all target com-
pensation. The net result was a rapid rise in `non-traditional' imports, espe-
cially of manufactured consumer goods, and a sharp fall in industrial
production. The result was a deficit in the current account of the balance of
payments in 1977 which was covered by inflows of financial capital from
abroad.
Inphase3,whichcoverstheperioduptomid-1979,whentheexchangerate
was frozen, the policy returned to a situation of compensatory exchange rate
adjustments.Naturally,theproductswhichhadpreviouslybeenprotectedlost
their position when customs duties converged to the uniform rate of 10 per
cent. Inparticular,therewasasignificantreductionintheprotectionofnon-
basic consumer goods and intermediate products produced locally. These
changes operated within an economy which was more sensitive than before
totheinternationaleconomy.Theaveragetariff,throughoutthethreephases,
hadfallenfrom94to10percent,witharealexchangeratewhichwas,bythe
end of the process, only 16 per cent higher than the rate established at the
beginningofthepolicywhichculminatedinunrestrictedopening-upinJune
1979. In this way, the exchange rate readjustment only managed to offset a
fifthoftheoverallimpactoftariffliberalizationonrelativecosts(seeFfrench-
Davis,1986).
Intheperiodwhichfollowed,asignificantrevaluationoftheexchangerate
occurred. It is commonly argued that the serious problems which emerged in
the Chilean economy in 1982 were due to the revaluation registered between
1979 and 1982. For this reason, it is interesting to examine what had been
happening in the manufacturing sector up to 1978. Separate information
shows that out of 29 industrial sectors, exports played a dominant role in
only two paper and wood and a significant role in a further two food
(fishmeal) and chemicals (especially molybdenum oxides). In contrast, the
impact of growing imports was felt in ten sectors. In another four the
main factor was the relative fall in demand (Vergara, 1980), although there
was also an impact from desubstitution of imports. In part this fall was asso-
ciated with the noticeable deterioration in income distribution registered in
thedecade.
Lastly, in phase 4 the real exchange rate was persistently revalued. This
was a result of a fixed nominal rate and a rate of internal inflation which
was higher than abroad during the three years after the freezing of the
exchange rate. The growing capital inflows registered until the beginning of
1982financedacurrentaccountdeficitwhichgrewcontinually.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Socialandagriculturaldevelopment
Socialpolicies
Upto1973,thestatehadsteadilyintensifieditsfunctionasaproviderofsocial
welfare,throughthedevelopmentofaframeworkdesignedtoprovidebenefit
covertothemiddleandworking-classurbangroups.Intheperiodbeginningin
1965, when an active social agrarian policy was defined, farm workers were
incorporatedintothisframework.
Socialdevelopmentuptothe1970scouldbecharacterizedbythefollowing
three central traits: firstly, there was a large allocation of resources for social
ends. Between 1920 and 1972 public social expenditure grew ten times faster
thantheGDP(Arellano,1985,p.50).Secondly,thedirectionofsocialpolicies
changed course from an early period (the 1920s and 1930s), when policies
sought to provide workers with a minimum level of protection in the face of
employers' excesses and abuses, and moved to a later period (the 1940s to
1960s) when the tendency was to provide more social cover against the risks
of urban life, as well as greater provision in the form of benefits such as
education and housing. Thirdly, an additional factor, which was the main
stimulus for these policies, came from social mobilization and the demands
madebytheworkers.Allofthiscontributedtotheproliferationofverydiverse
systemsofwelfare,whichgeneratednumerousdiscriminations,distortionsand
inequalities.
AclearresultofsocialpoliciesinChilewasthenoticeableimprovementina
collectionofindicatorsofsocialwelfare,inacontextofgrowingurbanization
whichincreasedfrom50percentin1930to76 percentin1970,accordingto
the population census (Arellano, 1985, table 1), and moderate economic
growth,oftheorderof4percentovertheperiod194070.
17
The improvement in working conditions was also substantial. In 1924 a
series of laws was passed on a wide variety of subjects: contracts of employ-
ment, union organization, the right to strike, conciliation and arbitration
tribunals,indemnityforaccidentsatworkandobligatorysocialsecurityinsur-
ance. The following were also created: a Ministry of Hygiene, Assistance and
SocialWelfare,andSocialSecurityFundsforprivateandpublic-sectorworkers.
In 1952 the National Health Service was organized, which achieved notable
progresswhichhasalreadybeendescribed.Tothisframeworkshouldbeadded
the governments' decisions to assign a growing volume of resources to social
welfare.Between1935and1955realexpenditureonsocialprogrammesinthe
publicsectormultipliedby4.5times,risingasaproportionofGDPfrom5per
centin1935to15percentin1955and26 percentin1972(Arellano,1985,
chart3,chapter1).
From the point of view of distributive equality, this social development
undoubtedly represented very important progress. The working population
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
wasabletoincreaseitsshareinrealincomeduetotheredistributionofferedby
thesesocialprogrammes.However,withinthissectorofsocietyvarioussources
of inequality were generated. On the one hand, the distinction between
employees and workers allowed a discrimination in benefits which favoured
theformer,especiallyduringthe1940sand1950s.Farmworkersconstituteda
severelyignoredsector,whichonlybegantobeincorporatedintothesebene-
fitsfrom1965.Ontheotherhand,insideeachofthesesegmentsworkersand
employeesotherkindsofdiscriminationemergedduetotheproliferationof
differentrules,andaccordingtothecapacityforpoliticalpressure.Anexample
ofthisislegislationgoverningtheminimumwage:itwasinplacein1934for
nitrate workers a period of crisis within the sector and in 1937 for private
sector employees, then for workers in leather, mills, bakers and print workers
(until 1951), in 1953 for farm workers, and in 1956 for all industrial workers
(Arellano, 1985, p. 36). Another notorious case was that of Social Security
Funds,whichproliferateduntilthereweremorethan35bythemid-1950s.
Theimpactofthesepoliciesontherelativedistributionofactualincomewas
important,asshownbyastudycarriedoutin1969:30percentofthepoorest
familiesreceived18percentofthebenefitsofferedbythesesocialprogrammes,
compared with a share in income of below 8 per cent (Foxley, Aninat and
Arellano, 1980). However, the fiscal impact was regressive as long as the state
was unable to increase tax revenue in proportion to the increase in social
expenditure, therefore contributing to budget deficits and inflationary pres-
suresduringtheperiodwhichaffectedpreciselythepoorestgroups.Ifincome
tax revenue and insurance payments are added together, as a proportion of
GDP,theyonlyincreasedfrom15.6 percentin1935to18.7percentin1955,
compared with a threefold increase in the social expenditure coefficient (Are-
llano, 1985, tables 3 and 5, ch. 1). At the beginning of the 1970s, Chile was
placed amongst the most socially developed Latin American countries. The
standard of education, the National Health System and the system for state
housing construction were the most advanced in the region. Social security
cover was also very high, reaching four-fifths of the workforce, and a massive
programmeoffeedingforpreschoolandschool-agechildrenwasinplace.
Agriculturaldevelopment
The inability of the agricultural sector to keep pace with industrial develop-
ment was a permanent theme in the literature on postwar Chilean economic
development.
18
Thefigures speakfor themselves: between 1940and 1953 the
industrialsectorgrewatanannualrateof7.5percent,whilsttheagricultural
sectorgrewbyonly1.8percent.Throughoutthisperiodtheinternalsupplyof
agricultural products did not manage to expand at the 2.5 per cent rate of
growthoftheurbanpopulation,whichledtoanincreaseinimports.However,
in spite of the increase in agricultural imports which were meant to fill the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
internaldeficitandpricecontrolinordertopreventpressuresoncosts,whole-
salepricesoftheseproductsgrewbynearly60percentmorethanthepricesof
industrialproductsbetween1928and1953.
There are various interpretations of the causes of the agricultural standstill.
Oneclassicexplanationmaintainedthatagriculturewasdiscriminatedagainst
by industrializationpolicies whichextractedresourcesfromitand transferred
them to the industrial sector (Mamalakis, 1965). In particular, price controls
on, and subsidized imports of, agricultural productsacted asa disincentive to
agricultural producers. Inaddition, agriculturewasburdened by importlevies
on capital and intermediate goods. Estimates of real protection for various
sectorsoftheChileaneconomyshowthatin1961agriculturestoodatalevel
equivalent to20 percentof theaverage arithmetic levelforall thesectors
(Behrman,1976).
This conclusionis notclear-cut,however,becauseatthesametime agricul-
turereceivedaseriesofotherbenefitsnottakenintoaccountbythesecalcula-
tions.Ontheonehand,althoughtherewerenominalimportleviesoncapital
and intermediate products for agriculture, in practice there were numerous
exemptionsdesignedpreciselytoloweragriculturalproductioncosts(Hurtado,
1984).Tothisshouldbeaddedtheloansatsubsidizedratesofinterest,verylow
taxation,agriculturalrealwagesthatwerewellbelowthelevelofurbanwages
and capital earnings derived from the revaluation of land in periods of high
inflation. The final balance of the package of incentives and disincentives in
agricultureisnotclear.
For their part, structuralist interpretations, without denying many of the
distortions in the incentive systems, emphasized the institutional aspects
such as the tenancy structure of land and the general absence of modern
businessmen in agriculture. Jorge Ahumada, one of the most perceptive ana-
lystsofdevelopmentproblemsandofChileanagriculture,maintainedthat
iftheconditionsforagriculturalproductionhadbeenreformed,ourhistory
wouldhavebeenverydifferent.Bysimplyeliminatingthelatifundio,orlarge
estates, there would have been a greater elasticity in the supply curve of
businessefforts.Therecanbenodoubtaboutthis...
and he continues `technical improvements, better farmers and better prices
togethermakeupaninseparabletrilogy fortheagricultural transformationof
thecountry'(Ahumada,1958).Ahumadawasoneofthemostinfluentialforces
ofagrarianreform,whichbegantobeimplementedduringthegovernmentof
President Frei Montalva (196470). Kaldor, in his study of the Chilean eco-
nomyin1956 (Kaldor,1964)haddevelopedsimilararguments.
Amoreintegralagriculturalpolicybegantobeappliedinthesecondhalfof
the 1960s. The main objective was to incorporate the sector into the global
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
developmentofthecountry.Thatnotonlyinvolvedapolicyofanincreasein
productionandareallocationofresourcestowardssectorswithgreaterpoten-
tialforgrowth,likethefruitproducingsectorCORFOputforwardaplanfor
the development of fruit production it also involved a social policy of
improvingconditionsforfarmworkersincludingtherighttounionorganiza-
tion, the equivalence of the minimum agricultural wage to the minimum
industrial wage and an increase in family allowances for farm workers, and
thebeginningsoflarge-scaleagriculturalreform.
19
Theresultofthispolicywasanincreaseintherateofgrowthofagricultural
production.Between1964and1970therewasanestimatedannualgrowthin
agricultureof4.9percent.Thetwolargestsub-sectorswerearableproduction,
which expanded to 3.7 per cent per year, and livestock production, with an
increaseof5.7percentperyear(PPEA,1976).Arableproductionunderwenta
somewhat uneven development since some crops suffered an absolute fall in
productionvaluewhilstothersgreatlyincreased.Beetcultivationstandsout;it
wasstimulatedbyanexplicitagro-industrialpolicywhichhadbegunattheend
of the 1940s. The introduction of sugar beet in Chile is an interesting case of
import substitution in agriculture, in a framework of a long-term strategy
creatingexternalitiesandopportunitiesfortechnologicallearning.
20
However,lookingatthelongterm,theobviousresultofagrarianbackward-
nesswasthatitaffectedtheprocessofindustrializationnotonlybecauseofthe
impactonrelativepricesandthestandstillindemandforindustrialgoods,but
also because of the pressures on the balance of payments. The allocation of
currency resources to import essential agricultural goods which could be pro-
ducedinthecountrycompetedwithimportsofcapitalgoodsandintermediate
products for industry. Also, since the deficit of agricultural supplies was
unstable and unpredictable in the medium term, the control of the balance
of payments also suffered from this instability, and this lack of predictability
affectedgeneraleconomicpolicy.
Finalconsiderations
Hopesandfrustrationsofpostwarindustrialization
Economicpoliciesfromthe1930sonwardswereveryheterodoxincomparison
towhathadbeenthetraditionalfocusofeconomicpolicy,asmuchbecauseof
thedifficultanduncertaintimesinternationallyasbecauseofoneundeniable
andoftenoverlookedfact:inarelativelybriefperiod,thestatehadtoassume
centralizedcontrolofaneconomywhichwasseriouslyrestrictedbyitsexternal
sector,butthatatthesametimehadserioussocialdemandsonit.Itscapacity
to manage policy was rapidly exceeded. For example, the adjustment of
expenditure on imports required measures with very negative implications
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
for internal stability and the real incomes of the middle and lower classes.
Furthermore, the control of the many imbalances had to be carried out in
difficultcircumstances:aseriousshortageofqualifiedpersonnel,anunaware-
nessofthesecondaryeffectsofthemeasuresapplied,andalackofinformation.
Consequently, it is not surprising that the process should have been riddled
with inefficiency and incoherence, which in the end led to inflationary pres-
suresandinstability.
Theaverageyearlyrateofinflationinthedecadeofthe1940sstoodat18per
cent,butin195458itacceleratedto51percent(Zahler,1978).Theeconomic
andsocialinequalitiesbecamemoreapparentinthefaceofintensemigration
fromthecountrysidetothecities,andanatmosphereofcrisisandpessimism
spreadwhichinhibitedthedevelopmentprocess.Thelossofpoliticalsupport
n for a government such as Ibaez's, which had prepared far-reaching
institutional and social changes, was added to the acceleration of inflation,
economicinstabilityandrisingunemployment.However,thevirtualpolitical
limbointowhichitfellshouldhavebeenovercomebytheinterventionofthe
controversial KleinSaks Mission (195558) appointed by the government in
thefaceoftheimpossibletaskofnegotiatinganeconomicagreementbetween
the different political and social lobbies
21
(see Ffrench-Davis, 1973, section
III.1.c).
Inthiscontexttherewasafullcriticaldebateonthedirectionbeingtakenin
Chileaneconomicdevelopment.InfluentialChileanandforeigneconomists
such as Jorge Ahumada, Anibal Pinto and Osvaldo Sunkel were amongst
othersinthemaingroupwhowerealsolinkedtoECLAC;andThomasBalogh,
David Felix, and Nicholas Kaldor, in the second, foreign, group formulated
novel analyses of the structural deficiencies which supposedly affected the
Chilean economy and were preventing a more accelerated and balanced
economy. All of them agreed that Chile was a country very well-supplied
with natural and human resources, whose modest rate of economic growth
4percentannualgrowthinthe1950s,seeTable5.1didnotcorrespondtoits
potential in resources. The main obstacles were to be found in the character-
istics of the system of economic organization and in the inadequacies of the
rulingelitewithregardtosavingsbehaviour.
Thequestionaroseofwhatwasgoingwrongintheindustrializationstrategy.
If wefocus on theaverage rate ofindustrial growth thefigures showthat this
was quite dynamic and sustained at over 5 per cent throughout the 20 years
between 1934 and 1953 (Munoz, 1968). But this was also a period of severe
allocation distortions, since there was a strict policy of exchange and import
restrictions,withheavyexemptions,pricecontrol,andsoon.Althoughitcan
bepresumedthatthecostsofinefficiencyintheallocationofresourcesplayed
arole,theydidnotimpedeindustrialexpansion.Ontheotherhand,therestof
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
theeconomy,especiallytheminingandagriculturesectors,performedpoorly,
whichispartlyexplainedbytherecessioninworldmarkets.
Industrialdeclinewasevidentinthemid-1950swheninflationarypressures
were so great that orthodox stabilization methods were applied by means of
restricting aggregatedemand. TheKleinSaksMission'sstrategyconsisted ofa
combination of orthodox stabilization methods and fiscal reforms to regain
financialequilibrium.Thepolicyofrestrictionofaggregatedemandcontribu-
ted to lower inflation, although at much higher levels than in the previous
decade. There wasareformof theforeign trade regime,which replaced direct
andquantitativecontrolswithadvaloremtaxes,andreductionsinthenumber
ofdifferentialratesofexchangetonomorethantwo,allofwhichcontributed
to improving the efficiency of the system. This package of policies was com-
plemented by an external loan negotiated with the International Monetary
Fund.However,theeffectofthesepolicieswasrecession.Industrialproduction
decreased and unemployment increased to the highest postwar levels. In the
middle of intense criticism the Mission ended its work without even having
established a new political consensus or a strategy for development able to
rescuethecountryfromdeadlock.
Thiswouldsuggestatfirstsightthatshort-termpolicies weremoreinfluen-
tialintheindustrialcrisisthanlong-termones.However,theproblemwasnot
that simple. The short-term imbalances reflect, in some measure, long-term
imbalances.Thestructuralistapproachofthe1950sputforwardthehypothesis
that the inflationary acceleration was due to rigidity of supply, especially in
farming produce and foreign exchange. All of this, in the face of dynamic
economic growth, translated into pressures on costs through some crucial
elementsliketheexchangerateandthepriceoffarmproduce.
Theproblemrevolvedaroundthequestionoftowhatextenttheserigidities
were due to the policies themselves, or to external factors such as a fall in
export prices. It is often argued that inelastic supply of foreign currency was
due to a stagnation in exports which was caused by distortions in trade and
exchangeratepolicies.Thestandstillinexports, inrealterms, duringmostof
the1940sand1950s,isanacceptedfact(ECLAC,1977).Butattheendofthe
war export prices recovered fairly steadily until the second half of the 1950s;
consequently external market factors cannot be seen as having a negative
effect. Furthermore, international trade returned to normal, especially after
1950.Ineffect,thegreatestshareofresponsibilitylayintheinternalworkings
of the economy. Copper was the main export and was operated by foreign
companies; it can be presumed that they were not interested in expanding
productionduetotheriseintaxation.In1925taxincome,asaproportionof
thevalueofcopperproduction,wasat5percent;thisfigurehadrisento17per
cent in 1945 and to 33.6 per cent for the average over the 195054 period
(World Bank, 1980). If we add to this the burden of an overvalued exchange
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ratefortheconversionofreturnswhichisestimatedtohaveamountedtoatax
of 16.7 per cent of the value of production in 195254 (Ffrench-Davis, 1974,
table 4), we can perceive that there was a conflict of interests between those
companiesandtheChileanstatewhichclaimedanincreasinglyhigherpropor-
tionoftheearningsofminingresources.
Therewerealsootherdisincentivestothegrowthofnewexports.Inaddition
tothetaxburdenofhighimportleviesonintermediateproductsnecessaryfor
final export, there was an exchange rate policy in place which kept the
exchange rate well below its equilibrium level. This policy was motivated as
much by the above reasons, which led to the use of the exchange rate as a
taxationtoolonforeigncompanies,asbyredistributionofincomeobjectives.
Acommonbeliefwasthatdevaluationsreducedworkers'incomes,whichwas
correctintheshortterm,butnotinthelongterm.Mamalakis(1976)estimated
thattheaveragerateofexchangefortheperiod194670was36percentofits
realvaluein193233and62percentofitsrealvaluein193445takinginto
account Chilean and US inflation. Price distortions were additional factors
which worsened the situation, given the structural factors which had already
prevented adynamicexportgrowth suchas theconflict ofinterest between
thecopperproducersandtheChileanstate.Thedeclineoftheexternalsector
with respect to GDP was extreme. From around 30 per cent which this sector
had reached on the eve of the Great Depression, by 195059 it had come to
represent 10 per cent (Corbo and Meller, 1981). This was a very restrictive
structureforarapidlyexpandingindustrialsector,whichrequiredaveryelastic
andflexiblesupplyofbothcapitalandintermediateproducts.
Insearch ofsocialtransformation:towardsradicalpolitics
Duringthe1960ssuccessivegovernmentsattemptedtoreformeconomicpoli-
cies, in the sense of a modernization favouring a more widespread use of the
market,agreater transferofresponsibilitytotheprivatesector,anincreasein
theinternationalinsertionoftheeconomyand,from1965,theexplicitobject-
ive of increasing equity. However, simultaneously the country was experien-
cinganincreaseinthedegreeofradicalpoliticsandideologicalstruggleswhich
createdagrowingfeelingofuncertainty,especiallyamongprivateinvestors.
Towards the end of 1958 the government of President Jorge Alessandri
(195864) was established. A prominent business leader, he promoted in the
nameofpragmatismastrategyofgreatermarketfreedom,preeminenceofthe
private sector and renewed attempts at stabilization. Paradoxically for a busi-
nessman, the need to reactivate the role of the state through stimulating
private investment and aggregate demand was put forward. Attempts to curb
inflationweremadebyreducingimportleviesandfreezingtheexchangerate,
and in order to finance the probable deficit in the balance of payments new
foreignloanswerenegotiated.Butbeyondshort-termpolicies,thegovernment
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
sought to define a long-term goal of gaining private sector confidence. The
diagnosis blamed the main causes of economic standstill as being overregula-
tionoftheeconomy,pricecontrols,bureaucracy,andthedistrustwhichprev-
ious governments had shown towards private businesses. It was necessary to
establish frameworks for mutual cooperation between the state and the capi-
talistprivatesector,butwithoutstateinterferenceinbusinessactivity.CORFO
hadtofulfiltheroleofdevelopmentbank,notofbusinessagentforthestate.
Thisstrategywaspartiallysuccessfulinitsattemptstoreactivategrowthrates
and the level of investment, but it failed to control inflation and to liberalize
foreigntrade.Thebalanceofpaymentdeficitsreacheduncontrollablelevelsin
196162andthegovernmentwasforcedtodevaluedrasticallyandreestablish
highimportlevies.Thedistributionofincomealsosuffered,negativelyaffect-
ingworkerswhoseshareintotalincomediminishedfromalmost52percentin
1960to45percentin1964,accordingtoofficialestimates.
Atthistime thediagnosis ofthe problemsof development acquired amore
radicalcontent.TheinfluenceoftheCubanRevolutionandtheUSprogramme
of the Alliance for Progress were persuasive in as much as the problems were
much deeper than the effect of changes in economic policies. Structural
reforms were required in the socioeconomic system, and particularly in prop-
erty rights. There was genuine ideological rivalry between the programmes of
reform put forward by the Christian Democrats and the Socialists. The latter
group had a long tradition in Chile, antiimperialist in tendency and actively
fightingforworkers'rights,especiallyinthefaceofthepresenceofUScompan-
iesinmining.Inthe1960sitdevelopedadistinctlyrevolutionaryprogramme
insearchofsocialism.
The Christian Democrat programme from 1965 tried to reconcile capitalist
modernizationwithsocialreformstendingtodemocratizethepoliticalsystem
and redistribute wealth in workers' and peasants' favour. Agrarian reform, a
stimulusforunionorganizationandland-sharingreinforcedaprocessofsocial
mobilization which rapidly gathered its own momentum. Conservative polit-
icalsectors,especiallylandowners,perceivedthattheirinterestswereseriously
threatened, but Christian Democracy had achieved solid political support
which allowed it to push forward its programme. However, political mom-
entum and the rivalry of the socialistcommunist block, especially in the
unionmovement,generatedaconflictbetweentheaccelerationofthechanges
and their consolidation. The government was weakened in the face of a
strengthening of conservative as well as socialist and communist sectors.
Whilst the former demanded an end to the reforms, the latter pushed for an
intensificationofthechanges.
Economic growth slowed down in the last years of the decade and
inflationacceleratedagain,afterashortperiodofbeingundercontrol.Never-
theless, income distribution improved, especially in favour of the agricultural
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
workforce.Employees'shareofincomeregained52percentin1970,thefarm
workers'realminimumwageincreasedby24percentbetween1964and1970
and the distributive effects of the state's social programmes significantly
benefited the lowest income groups. Progress was also made on the partial
nationalizationofthecoppercompanies,onthemodernizationandrationali-
zationofthestateapparatus,andonthepartialeliminationofthelatifundio, or
large estates, and the anachronistic social relations in the country areas. The
policy of industrialization stimulated the development of new high-tech sec-
tors like telecommunications, the petrochemical industry and the infrastruc-
tureforagro-industry.Italsocorrectedsomedistortionsinthekeyinstruments
for the price system such as the exchange rate, real interest rates, rationaliza-
tion of the tariff structure and stimuli for exports, especially through new
agreementsforregionalLatinAmericanintegration.
The triumph of the Unidad Popular in the 1970 elections was unexpected
evenbyitsownsupporters.Thatyearsawtheformalendofthepartialpolitical-
economic consensus which had prevailed for almost 40 years, during which
time the capitalist and mixed economy had not been seriously challenged.
From1970theUnidadPopularpolicyproposedatransitiontosocialismthrough
totalexpropriationofbothnationalandforeigncapital,radicalizationofagrar-
ianreformand,eventually,theestablishmentofanewsocialistinstitutionality.
The main objective was the formation of an area of social or state property,
which would be composed of 91 of the largest industrial companies in the
country, all large-scale mining and the banking system. This area of social
propertywouldbecometheleverfortheaccumulationofcapital.
The history of the Unidad Popular years (197073) is well-known, and it is
therefore not necessary to repeat it here (Bitar, 1979; Meller, 1996). A lack of
connection between political and economic strategy led to unmanageable
imbalancesintheeconomy.Theareaofsocialpropertywascarriedawaybya
tideofpoliticalmomentum,asmuchinitscompositioninSeptember1973it
had nearly 500 small and medium-sized companies as in its operation its
losses generated a public deficit which was 22 per cent of GDP in 1973,
compared to 3 per cent in 1970. On the other hand, the gap between the
magnitudeofthechangessoughtandtheminoritynatureofthegovernment's
politicalsupportgaverisetoadilemma. Thegovernmenthadeithertoseeka
political alliance with moderate reformist sectors in order to consolidate the
changes made, or to proceed without negotiation even at the price of over-
ridingtheinstitutionalsystem,usingonlyitsabilitytorallypopularsupport.In
addition to these problems threats came from sectors which favoured open
insurrection and destabilization of the government. The chosen option to
proceed without negotiation corresponded to the ideological radicalism
which had been underway since the previous decade, and had been practised
byChristianDemocracyatthetime.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Paradoxically, it was the military regime which was in power in September
1973 which brought about the greatest transformation in the last fifty
years.Withthesupportofthemilitaryandtheuseofunprecedentedrepression
the institutional and socioeconomic systems were changed. There was now
uninhibited capitalism. These changes came about progressively (Vergara,
1985).Initiallythechangeswereconceivedto`re-establishthefracturedinsti-
tutional system' and to rectify economic imbalances inflation in 1973 was
over 800 per cent. Shortly afterwards several areas were examined: the indus-
trialization strategy, which had been in place since the 1920s, the state's role
andthesocioeconomicreformscarriedoutsince1965.Fromthisthefollowing
objectivesemerged:tocompletelyliberalizetheeconomy,exceptforthelabour
andcurrencymarkets;toreducetheroleofthestatetothatofasubsidiaryarm
oftheprivatesector;topropeltheinsertionoftheeconomyintotheinterna-
tionalcapitalistsystem;andtostimulateforeigninvestment.Inthelongterm
thisstrategywascomplementedbyvariousshort-termpoliciesgearedtowards
thecontrolofaggregatedemand,theeliminationofthefiscaldeficit,financial
relaxationofcontrolsandautomaticadjustmentinthefaceofexternalshocks
(Foxley, 1982). These reforms saw the end of a model of development geared
towardstheinternalmarketinwhichthestatehadtakentheleadingroleand
income redistribution had been central to economic policies; these aspects
would nowbe left tothe market. These neo-liberal reforms wentfurther than
placingthemarketandtheprivatesectorattheeconomicforefront,andwere
foundedonexportgrowthandliberalizationofthefinancialsector.Inthisway,
theexportsectorcametoplayafundamentalroleinthestrategyfordevelop-
mentinitiatedafter1973.
The growth in exports in the 1970s and 1980s was sustained in part by the
newlycreatedinstitutionalsystem,butalsobydevelopmentswhichhadbeen
in gestation since the 1960s. At the beginning of the 1990s, the Chilean
economy had already undergone a great transformation and was in a mature
phaseofreform.Anewbusinessclass,whichwasmoredynamicandfocusedon
the long term in comparison to the old business class, had now become the
leading investors, and even ventured to invest in other emerging markets
abroad. This new business class imposed a more pragmatic administration on
its companies. The market economy has consolidated during the 1990s
and has been able to adjust to new democratic conditions. This transition
not only meant the restoration of a system of free elections, freedom of the
press and of opposition, but also the modification of aspects central to
politicalinstitutionalitywhichaffectthepowerbalance.Thetwomajortopics
of public debate in the 1990s are consolidation of economic growth, with
greater stability and equity, and political transformation towards a more
genuinedemocracy.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Notes
1 Thebeginning of`developmentfromwithin'shouldnotbe identifiedasindustrial-
izationassuch.Thisearlierprocessoriginatedattheendofthenineteenthcentury
underthestimulusofthenitrateworksboomandasaresultofpublicpoliciesmore
activelyfavourabletoindustrialization,particularlyviaprotectionism(Munoz,1977)
andisexplainedinPalma'schapterintheparallelvolumeonexporteconomies.
2 MarfanandArtiagoit a(1989)giveratesofproductioncapacityfor196088similarto
those obtained in this work. The main methodological differences with our work
are:(i)theoriginofcapitalfiguresintheyearsbefore1982(officialaccounts),and(ii)
the definition of capital formation (excluding residential). The similarity of the
results,namelythetimelapsecommontobothstudies,isdeterminedbythechoice
ofthe`peaks'methodandtheconcordanceintheidentificationofthese.
3 Here we are associating the increases in production with the formation of fixed
capital. Apart from problems in measuring investment in human capital, there
have been changes in the use of other factors. The active population grew 1.9 per
centperyearinthefirsttwodecades,2percentin197381and3.6 percentin1981
89.Ontheotherhand,theintensityofuseofnaturalresourcesgrew.Consideration
of both factors leads to an estimation of the pure productivity of capital which is
significantlylessthanthatcalculatedinthistext.Thegapbetweenthetwoisclearly
greaterinthe1980sthaninthe1950sand1960s.
4 In the estimation of net production and investment a fixed coefficient of deprecia-
tion of 8 per cent is used. The increase in this, for example, in response to the
indiscriminate liberalizing of imports in 197481, must be considered as one of
the costs in an analysis of the benefits and costs of this policy. On the other hand,
therapidity(andnature)oftechnologicalchangetendedtoshortenthelifeoffixed
capital,probablyraisingthemeanfactorofdepreciationofthecapitalstock.
5 Labourproductivitycorrespondsheretovalueaddedperworker.Thus,itspatternis
affected,amongstotherthings,bychangesinthecapitallabourratio,bytherateof
useofcapitalandbythequalityoflabour.
6 Theseinitialscorrespondto:CentreforMiningandMetallurgicalResearch(CIMM),
the Technological Institute of Chile (INTEC), Forestry Institute (INFOR), and Insti-
tuteofFisheriesDevelopment(IFOP).
7 The historian, Hobsbawm (1996, pp. 449) points out that the First and Second
WorldWarswerebothdecisiveinbringingaboutstateinvolvementintheeconomy.
Thewareffortsprovidedanimpetusforplanningsystemsandcentralizedallocation
ofresources.
8 They included Desiderio Garc a, who years later designed the legislation which
would create the Corporation of Development and Production (CORFO), and Guil-
lermoDelPedregal,itsfirstvice-president.Thisgroupoftechnicianswithabusiness
vision in the service of the state were of crucial importance in the capacity for
institutional change in the 1930s and 1940s. The group and their work were rein-
forced by the inclusion of a new generation of engineers who would have a prom-
inentroleinthecreationofpublic-sectorcompaniesduringthe1940s.Thoseworthy
of mention are Rau l Saez, architect of the National Electricity Board, ENDESA, and
Eduardo Simian, who discovered oil in Chile and founded the National Oil Com-
pany,ENAP(Munoz,1993).
9 Althoughtheprivatesectorwasveryinterestedanddemandedthatthestatehavean
activedevelopmentpolicy,theconcernsoverfurtherinterventionwereexpressedin
a very critical parliamentary debate which threatened to paralyse the bill. Political
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
negotiation with right-wing sectors and agriculture representatives meant that the
billwasfinallypassedinexchangeforanagreementthatthegovernmentwouldstop
itstradeunionandreformistinitiativesinagriculture(Mun
ozandArriagada,1977,
pp.359).
10 These state companies had as their main objective an increase in the supply of
strategicindustrialinputs, suchas the NationalOil Company(ENAP), the National
EnergyCompany(ENDESA)andthePacificSteelCompany(CAP).Theywerealsoto
developagriculturalareaswithlittlealternativeuse,liketheNationalSugarIndustry
(IANSA).Inthefieldoftechnologicalresearch,CORFOpromotedthestudyofnatural
resourcesandtheirpotentialuse.
11 Although this proposal did not abandon the idea that the state should maintain a
predominantroleinindustrialpolicies,andevenintheownershipofmanycompan-
ies, it did aim for efficient management of public-sector companies, rejecting the
notionofthepopulistandbureaucraticstate,andhavingalong-termvision.
12 Tobringaboutthischangeseveralsuggestionsweremade:stablerealexchangerate
policies,rationalizationofthetariffstructure,agreatefforttoincreaseproductivity
and quality, and the organization of an export promotion system which would
representChileinthemainexternalmarkets.
13 TheseinitialsstandforCopperCorporation(CODELCO),NationalMiningCompany
(ENAMI)andNationalOilCompany(ENAP).
14 Adetailedexamination of tradeand exchange ratepoliciesis presented in Ffrench-
Davis(1973).
15 The lack of a coherent policy of industrial development contributed to the
limitations of the export subsidy. This was reinforced by an absence of standard
methods of social evaluation of projects, which could apply to different state
institutions. Drawback was meant to be used to close the gap between social and
private prices, within the development strategy. In reality, neither was taken into
account.
16 ThesetopicsareexaminedindetailinFfrench-Davis(1986)andFfrench-Davis,Leiva
andMadrid(1991).
17 Considerthe changesinthe followingindicators: (i)mortalityrates:downfrom27
perthousandin1940to9perthousandin1970;(ii)childmortalityrates:downfrom
118perthousandlivebirthsin1952to32perthousandin1970and17perthousand
in1980;(iii)illiteracy,downfrom20percentofthepopulationovertheageof15in
1952 to 11 per cent in 1970 and 7 per cent in 1980; (iv) enrolment in primary
education of the populationaged 614,up from64 percentin 1950 to94 percent
in 1970; (v) enrolment in secondary education of the population aged 1518, up
from9percentin1950to38percentin1970and53percentin1980.
18 Forexample,ArthurLewisemphasizedtheeffectsoftheagriculturalstragnationon
industrial costs, real wages and income taxes. Agricultural productivity determined
the real wages of the farm workers and, as such, their capacity to buy industrial
products.Arapidgrowthinthisproductivitycontributedtotheincreaseinsustain-
abledemandforthesemanufacturedgoods(Lewis,1980).
19 DuringthegovernmentofJorgeAlessandrialawofagrarianreformwasdecreed,but
itwasnotwidelyapplied.
20 At the end of the 1940s CORFO formulated a plan which included incentives to
farmers to begin to cultivate sugar beet, with an investment plan which would be
carriedoutbyCORFOitself.Thereforein1949theNationalSugarIndustry, IANSA,
was set up as a branch of CORFO. Amongst the incentives offered to farmers were
technicalandfinancialassistance,theestablishmentofcontractsforproductionand
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
buying, as well as the extra benefit that beet, in rotation with other annual crops,
increased the productivity of the land (Munoz, 1993). Although there was a great
deal of debate on the social cost of this policy, the sector developed significantly.
Between 1964 and 1970, the beet sector expanded 170 per cent whilst the total
expansion in annual crops was 25 per cent throughout the period (PPEA, 1976,
table3.1).
21 TheproposalsputforwardbytheKleinSaksMission(195657)wereoneofthefirst
attemptsatpartialliberalizationoftheeconomy,butinfacttheydidnothavestable
politicalsupport,notevenfromtheright-wing.Adetailedanalysisoftheseproposals
isinFfrench-Davis(1973).
References
Ahumada,J.(1958)Ùnatesissobreelestancamientodelaeconom achilena',Econom a,
Nos.601,Santiago:UniversityofChile.
Arellano, J.P. (1985) Pol ticas sociales y desarrollo: Chile 192484, Santiago: Ediciones
CIEPLAN.
Behrman, J. (1976) Foreign Trade Regimes and Economic Development: Chile, New York;
London:ColumbiaUniversityPress.
Benavente, J.M. Crespi, G. and Katz, J. (1996) El sistema innovativo nacional chileno,
Santiago:DDPECEPAL/UniversidaddeChile.
Bitar,S.(1979)Transicion,socialismoydemocracia:laexperienciachilena,MexicoCity:Siglo
XXI.
Cabezo n,P.(1971)Àntecedenteshisto ricosdelasimportacionesydelapol ticacomer-
cialenChile',CuadernosdeEconom a,vol.25,December,Santiago.
Carrasco,C.andFfrench-Davis,R.(1969)`Cuatroan ndelanuevapol tica osdeaplicacio
decambios',BancoCentraldeChile,Bolet nMensual,no.498,August.
CONICYT(1993)IndicadoresCient ficosyTecnologicosparaChile,Santiago:CentralBankof
Chile,1952,`MemoriadelBancoCentral,1952'.
Corbo, V. and Meller, P. (1981) Àlternative Trade Strategies and Employment Implica-
tions:Chile',inA.O.Kruegeretal.(eds),TradeandEmploymentinDevelopingCountries,
1,IndividualStudies,Chicago;London:UniversityofChicagoPress.
CORFO(1969)Basesdediscusiondeunaestrategiadedesarrolloindustrialparaladecadadel
70,Santiago:GerenciadeIndustrias,Divisio ndePlanificacio nIndustrial.
(1970)Basesdeunaestrategiadedesarrolloindustrialparaladecadadel70,Gerenciade
Industrias,Divisio ndePlanificacio nIndustrial,4vols,Santiago.
ndePromocio Corporacio nUniversitaria(CPU)(1987)Eldesarrollocient ficoytecnologico
enChile.Unanalisiscualitativo,19651985,Santiago.
Cortazar and Marshall, (1980) Ìndices de Precios al consumidor en Chile, 197078',
ColecciondeEstudiosCIENPLAN,no.4,Santiago.
ECLAC (1977) America Latina:relacion de preciosdel intercambio,19281976, Santiago:E/
ECLAC/1040,August.
Ffrench-Davis,R.(1986)ÌmportLiberalization:TheChileanExperience,197382',inJ.S.
ValenzuelaandA.Valenzuela(eds),MilitaryruleinChile,Baltimore,Md.;London:The
JohnsHopkinsUniversityPress.
(1999a) Reforming the Reforms in Latin America, London: Macmillan; New York: St
Martin'sPress.
(1999b)Entreelneoliberalismoyelcrecimientoconequidad:,Santiago:DolvenEditors.
(1982)ÈlexperimentomonetaristaenChile:unas ntesiscr tica',ColeccionEstudios
CIEPLAN,vol.2,September,Santiago.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
(1980) `Liberalizacion de las importaciones: la experiencia chilena en 197379',
ColeccionEstudiosCIEPLAN,no.4,November,Santiago.
(1979a)Èxportacioneseindustrializacio nenunmodeloortodoxo:Chile,197378',
RevistadelaCEPAL,no.9,December,Santiago.
(1979b)`LasexperienciascambiariasenChile:196579',ColeccionEstudiosCIEPLAN,
no.2December,Santiago,andinWilliamson(1981).
(1974)Ìntegracio ndelagranminer aalaeconom anacional:elroldelaspol ticas
econo micas', in E. Tironi (ed.), El cobre en el desarrollo nacional, Santiago: Ediciones
NuevaUniversidad.
(1973)Pol ticaseconomicasenChile:195270,Santiago:EdicionesNuevaUniversidad.
(1964)Èlementosparaunapol ticadecomercioexterior',inDialogossobreeldesar-
rollo economico de Chile, July: Instituto de Econom a, University of Chile; reprinted,
1965,inCuadernosdeEconom a,vol.5,April,Santiago.
Ffrench-Davis,R.andMunoz,O.(1990)`Desarrolloecono mico,inestabilidadydesequili-
briospol ticosenChile:195089',ColeccionEstudiosCIEPLAN,vol.28,June,Santiago,
and in S. Teitel, 1995, Hacia una nueva estrategia de desarrollo para America Latina,
WashingtonDC:Inter-AmericanDevelopmentBank.
Ffrench-Davis,R.,Leiva,P.andMadrid,R.(1991)`LaaperturacomercialenChile,Estudios
dePol ticaComercial,vol.1,UNCTAD,NewYork:UnitedNationsPublications.
Foxley, A. (1982) Èxperimentos neoliberales en America Latina', Coleccion Estudios CIE-
PLAN,vol.7,March,specialedition,Santiago.
Foxley,A.,Aninat,E.andArellano,J.P.(1980)Lasdesigualdadeseconomicasylaacciondel
Estado,Mexico,D.F.:FondodeCulturaEcono mica.
Hachette,D.andLuders,R.(1992).LaprivatizacionenChile,Santiago:CentroInternacio-
nalparaelDesarrolloEcono mico(CINDE).
Hobsbawn,E.J.(1996)TheAgeofExtremes,NewYork:VintageBooks.
Hofman,A.(1992),`CapitalAccumulationinLatinAmerica:ASixCountryComparison
for19501989',ReviewofIncomeandWealth,series38,no.4,December,NewYork.
Hurtado, H. (1984), El sector agropecuario chileno: 196474, Programa de Posgrado en
Econom aAgraria,Santiago:PontificiaCatholicUniversityofChile.
Ibaez, A. (1983) `Los ingenieros, el Estado y la pol tica en Chile: del Ministerio de n
Fomento a la Corporacio n de Fomento, 19271939', Historia, vol. 18, Santiago: Cath-
olicUniversityofChile.
Kaldor, N. (1964). Èconomic Problems of Chile', in Essays on Economic Policy vol. II,
London:Duckworth.
Leniz,S.andRozas,M.(1974)`Compatibilizacio nCuentasNacionalesODEPLAN-CORFO
19401962,19601967',Documentodetrabajo,vol.21,Santiago:InstitutodeEconom a,
CatholicUniversityofChile.
Lewis, A. (1980), Evolucion del orden economico internacional, Mexico City: El Colegio de
Mexico.
Mamalakis, M. (1976), The Growth and Structure of the Chilean Economy, New Haven,
Conn.:YaleUniversityPress.
(1965)`PublicPolicyandSectoralDevelopment:ACaseStudyofChile,19401958',
inM.MamalakisandC.Reynolds(eds),EssaysontheChileanEconomy,Homewood,Ill.:
RichardD.Irwin.Inc.
Marcel,M.andMeller,P.(1986)ÈmpalmedelascuentasnacionalesdeChile19601985:
Metodos alternativos y resultados', Coleccion Estudios CIEPLAN, vol. 20, December,
Santiago.
Marcel,M.andPalma,G.(1990)`Kaldoryel``discretoencanto''delaburgues achilena',
ColeccionEstudiosCIEPLAN,vol.28,June,Santiago.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Marfan,M.(1984)`Pol ticasreactivadorasyrecesio nexterna:Chile19291938',Coleccion
EstudiosCIEPLAN,vol.12,March,Santiago.
Marfan, M. and Artiagoit a, P. (1989) Èstimacio n del PGB potencial: Chile 19601988',
ColeccionEstudiosCIEPLAN,vol.27,December,Santiago.
Meller, P. (1996) Un siglo de econom a pol tica chilena (18901990), Santiago: Editorial
AndresBello.
Munoz, O. (1995) Los inesperados caminos de la modernizacion economica, Coleccion IDEA,
Santiago:EditorialUniversidaddeSantiago.
(ed.)(1993)Pol ticaspublicas,historiaspersonales,Santiago:EditorialLosAndes.
(1977) Èstado e industrializacio n en el ciclo de expansio n del salitre', Colleccion
EstudiosCIEPLAN,vol.6,January,Santiago.
(1968) Crecimiento industrial en Chile, 19141965, Instituto de Econom a, Santiago:
UniversityofChile.
oz, O. and Arriagada, A.M. (1977), Òr genes pol ticos y econo Mun micos del Estado
empresarialenChile',ColeccionEstudiosCIEPLAN,vol.16,September,Santiago.
Ossa,C.(1971),`Consideracionessobreunaestrategiadedesarrolloindustrial',Trimestre
Economico,vol.149,JanuaryMarch,Mexico:FondodeCulturaEcono mica.
Pinto, A. (1973), Chile, un caso de desarrollo frustrado, 2nd edn, Santiago: Editorial Uni-
versitaria.
PPEA, (1976), Èl sector agro pecuario chileno: 196474' Programa de Post Grado de Econ-
om aAgraria,CatholicUniversityofChile,Santiago.
Vergara,P.(1985)Augeyca dadelneoliberalismoenChile,Santiago:FLACSO.
(1980) Àpertura externa y desarrollo industrial en Chile: 19741978', Coleccion
EstudiosCIEPLAN,vol.4,November,Santiago.
Williamson, J. (1981) The Crawling Peg. Past Performance and Future Prospects, London:
Macmillan.
WorldBank(1980)Chile:AnEconomyinTransition,WashingtonDC.
Zahler,R.(1978)`Lainflacio nchilena',inR.Zahleretal.,Chile19401975,treintaycinco
anosdediscontinuidadeconomica,Santiago:InstitutoChilenodeEstudiosHuman sticos.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
6
ImportSubstitutionandGrowthin
Brazil,1890s1970s
MarcelodeP.Abreu,AfonsoS.BevilaquaandDemosthenesM.Pinho*
Introduction
It is extremely difficult to disentangle for analytical purposes the origins and
deepening of import substitution in Brazil from the consolidation and con-
tinuousgrowthofthecoffeeexporteconomy.Inanalmosttrivialsense,thisis
becausethedemandgeneratedbytherapidly growingcoffeeeconomyserved
as an engine of growth. First coffee stimulated products whose domestic pro-
duction was protected by relatively high transportation costs; then its influ-
ence spread more generally, as animal spirits, manpower availability, credit
facilities, tariff protection and foreign supply disturbances made possible the
substitutionofmoresophisticatedimports.
Import-substituting industrialization (ISI) was important well before 1930,
under the òld Republic'. It became even more important under Vargas in the
1930sandearly1940s,andflourishedfromtheendofSecondWorldWaralmost
untilthemilitarycoupin1964.Whileafterthemid-1960sitneverrecoveredits
former strength as a factor explaining industrial growth, it was qualitatively
important as from the mid-1960s to the 1980s it affected industrial sectors
producingtechnologicallymoresophisticatedgoods.
Import substitution had already had a significant impact in the domestic
supplyofindustrialgoodsbytheturnofthetwentiethcentury.Afterthebegin-
ning of the Depression in the late 1920s there was, with a lag, a new wave of
import-substitution investments. So, although one can think of the 193060s
periodascorrespondingtothecoreofimport-substitutingindustrializationin
Brazil, it is important to emphasize that this process was already at work
from the 1890s, well before reversal from outward-looking to inward-looking
*TheauthorsthanktheparticipantsintheOxfordandPaipaprojectworkshopsfortheircommentsas
well as Regis Bonelli for help with references and the competent research assistance of Fernando H.
Alvares,LuizGustavoChermanandAlvaroB.A.Motta.Theauthorsarealsogratefulforthefinancial
supportofFINEPandCNPq.
154
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
MarcelodeP.Abreu,AfonsoS.BevilaquaandDemosthenesM.Pinho 155
development. It is true that earlier ISI occurred without a turmoil in world
economic and financial conditions such as the Depression in the early 1930s;
nordiditdependongovernmentdirectinterventionintheallocationofscarce
foreign exchange; but the decisive development was there a sustained sig-
nificantreductionintheshareofimportsinthedomesticsupplyofindustrial
goods.
High tariffs, or non-tariff barriers after 1930, have been a crucial feature of
importsubstitutioninBrazil.Thestandardliterature(Dean,1969),stressesthe
convergenceofinterestsbetweencoffeegrowersandindustrialistsratherthan
the opposition of such interests as suggested by Furtado (1959). Furtado's
interpretationstressessuchoppositionofinterestsmainlyinconnectionwith
theexchangeratepolicyandtheso-calledsocializationoflosses.
1
Dean(1969),
in his well-established revision, underlined the complementarity of interests
between coffee growing and industrialization: often coffee growers were also
entrepreneurs both in industry and infrastructure facilities. Others have
stressedthelackofcommitmenttolaissez-faireinBrazil,asshown,forinstance,
by the intervention in the supply of labour to an expanding coffee-growing
economy.Interventionisttendenciesculminatedintheendorsementofcoffee
`valorization'bythefederalgovernmentafter1907.ThecaseofBrazilhasbeen
contrasted to that of Argentina, where before 1930 tariffs were much lower
than in Brazil. One important reason for the lack of a laissez-faire tradition in
BrazilwasthatBrazilwasaratherpeculiarsmalleconomyasitwasthedomin-
ant supplier of the coffee market for more than a century. As it was a price-
making export economy, domestic policies had an influence on world coffee
pricesandrent-seekingbehaviourwaslessconstrainedthaninothercommod-
ityexportingeconomies(Abreu,1994).
In another important aspect the Brazilian economy can be contrasted with
most other Latin American economies: it had a significant domestic market
withapopulationexceeding17millionin1900,evenifwithanextremelylow
GDPpercapita.
2
Sotheconventionalargumentsaboutsizeofthemarketand
lackofscaleforimportsubstitutionmustbeusedwithmuchcircumspection,if
atall.
Import substitution deepened in the 1890s, as measured by the share of
domestic production in total supply, then again in the 1930s and from the
late1940s.Itsimportanceforindustrialgrowthvariedwiththefluctuationsof
theworldeconomyandtherecurrentbalance-of-paymentsdifficulties.Inlater
periodsimportlevelsweresolowasaproportionoftotalsupplythatitbecame
impossible for import substitution to continue to play an important role as a
sourceofindustrialgrowth.
Standard hypotheses on the impact of foreign sector variables on growth
include in a central role the idea that òutward-looking' policies enhance
growth. There is much ambiguity in the definition of an outward-looking
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
156 ImportSubstitutionandGrowth inBrazil
stance by a country: it may mean opening up domestic markets, fostering
exportsorkeepingtheexchangerateatthe`rightlevel'.Alinkbetweengrowth
andopennesshasbeenshowntobelikelybycross-sectionwork,evenifsome
of the literature is based on rather ad hoc definitions of outwardness (Green-
away and Nam, 1988 and World Bank, 1987).
3
But high protection of the
domesticindustryagainstimportcompetitiondidnotpreventanalmostspec-
taculargrowthperformancefromthebeginningofthecentury:Brazil's1900
73growthperformancewasonlybetteredbyJapan'sbyanamplemarginand
Finland'sbyaverymodestmargin(Maddison,1989).
Thischapterisdividedintofoursections,besidesthisintroduction.Thenext
sectionconsiderstheoriginsofimportsubstitutionfromtheearlydaysofthe
Republic in the 1890s to 1930 and proposes an original explanation to con-
ciliate high protection and a high growth path. In the following section the
high-noonperiodofimportsubstitutioninBrazilfromthe1930suntiltheearly
1960s is analysed, together with a discussion of possible alternative policies
giventheinternationalcontext.Wethenfocusontheperiodfromtheshiftin
economicpoliciesfollowingthemilitarycoupof1964untilthelate1970s,and
the possible links between import substitution and the declining growth
performancearediscussed.Thefinalsectionpresentstheconclusions.
Importsubstitutionbefore1930
4
BytheturnofthiscenturyBrazilianindustrywasalreadywell-establishedasan
industrialboomfollowedthehighinflationduringtheEncilhamentoperiodin
theearly1890s.Theboominimportsofcapitalgoods,firststressedbyFishlow
(1972) based on five-year average British export data, is confirmed for all
relevantBraziliansuppliers(Suzigan,1986,table18).In1900importswere40
per cent of the total domestic supply of textiles, but by 1919 the share of
imports in total supply of industrial products was reduced to only 25 per
cent.Itis,ofcourse,well-knownthatthecomparisonofimportsanddomestic
industrial production leads to the underestimation of the importance of
imports in total supply, as imports should be added to their industrial inputs
to make the figure comparable to domestic industrial production which
includes final goods and their inputs. Techniques of correction depend on
inputoutput data which are not available for such an early period. It is thus
reasonable to adopt cruder methods which result in an adjustment for the 25
percentratioin1919toaround37percentona`valueadded'basis(Fishlow,
1972, pp. 3234). Industrial output increased almost five-fold between 1900
and1930,whiletotaloutputincreasedonly3.5times(Haddad,1978).
TariffsinBrazilfluctuatedverysubstantiallyduringthenineteenthcentury,
andtheratiobetweentariffrevenuesandtotalimportspeakedatalmost50per
cent in 1888. In the first half of the 1890s it fell to almost 20 per cent as the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
exchangeratedepreciatedtolessthan30percentofits1889level,thenitrose
to35percentin1896 andreachedapeakof50percentin190607.Itstarted
tofallintheearly1910sandfellbelow20percentduringthewar.Itincreased
subsequently, and in the second half of the 1920s it was over 25 per cent.
Disaggregated data on ad valorem equivalent tariffs which would have been
applied to selected goods without duty exemptions or reductions indicate
particularly high tariffs for low-quality cotton textiles, declining from almost
400 per cent in 1901 to more than 100 per cent in 1928. Protection on inter-
mediateandcapitalgoods,althoughlower,wasalsoveryhigh(Villela,1993,p.
181).
Tariffs correspond to a lower bound of the price divergence between do-
mestically-produced and imported goods in Brazil, as there were additional
charges which may have had a significant additional impact on the cost of
imports. These included, at different times, statistical taxes, administrative
taxes, taxes to improve harbour facilities and roads, and discretionary social
contributions(NunesandSilva,1929,pp.123).Illegalstatetaxesonimports
mayalsohavebeenimportant,especiallyinthepoorerstates.Tariffexemptions
werealsofrequentinBrazil,butinmanycasesrequiredregistrationofimport-
ing firms generally, but not exclusively, railway companies and public
utilities. However, after 1911 duty exemption could not benefit goods which
could be produced domestically and an official register of domestic firms was
kept (Nunes and Silva, 1929, pp. 25270). The concept of `similar domestic
production' would be transformed into the most effective non-tariff barrier
providingabsoluteprotectionformanyindustrialsectors.
The contrast with other major commodity exporters such as Argentina is
very marked. By theearly 1920s the Argentine average tariff wasconsiderably
belowCanada's:15percentcomparedto24.9percent(Solberg,1987,p.105).
Commitment to economic liberalism, and particularly to a low tariff, was
virtuallynon-existent inBrazil.Groupswhichwouldbefavouredbyitparts
oftheemerging,butpoliticallyunorganized,urbanmiddleclass,andthework-
ing class lacked the clout to influence the economic policy formulation
process. Bad memories counted also, as the low tariff ceiling imposed by the
treaties with Britain until 1845 in a curiously overlooked episode in the
business imperialism debate had resulted in severe fiscal constraints during
theperiodofpoliticalturmoilwhichfollowedindependence.Coffee-planters,
on the other hand, could barely stress with any credibility a serious commit-
ment to economic liberalism. From the early nineteenth century their stand
had emphasized rent extraction through illegal extension of the slave trade,
followed by subsidized immigration paid by the government and restrictive
land policies. This trend had culminated in the coffee `valorization' schemes
after 1906, based on a freeze of productive capacity and aimed at reaping
monopolyprofitsbasedonthecountry'smarketpowerinthecoffeemarket.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
The segregation between agricultural and import-substituting industrial
interestswasnotwell-defined.Thereisinfactevidenceofconsiderableinvolve-
ment of coffee interests in the fast growing Paulista industry already at the
turn of the century. About 45 per cent of industrial workers in Sao Paulo in
1901 were employed in firms controlled by coffee interests.
5
This confirms a
trend ofportfolio diversification by coffeegrowers which startedinthe1870s
and extended to investment in the export infrastructure, including railways,
and in the processing of agricultural products. Contemporary evidence indic-
ates that the industrial lobbies had considerable weight. From quite early on,
industrialinterestswereabletocorrectlygaugetheirabilitytoextractconces-
sionsfromthegovernment,includingaveryprotectivetariff.
6
A feature of paramount importance in this context which has been over-
lookedintheliteratureisthelinkbetweentarifflevelsandtheextenttowhich
each particular commodity-producing country was able to influence relevant
commodityprices.Thelowerthemarketshareofagiveneconomyinaspecific
market,theweakerwillbetheinfluenceofdevelopmentsinthiseconomyon
worldpricesofthiscommodity.Insucheconomies,thescopeforhighprotec-
tion of domestic production, and in particular of goods demanded by the
export sector, is constrained by the need to maintain costs of production in
linewiththoseofcompetitors.
Brazil, however, is a special case among commodity exporters. Since quite
early in the nineteenth century its share of the world markets of coffee was
such that cost conditions in Brazil tended to determine world prices. The
Braziliancoffeesectormarginalcostcurvewastoaverylargeextentequivalent
totheworldcoffeesupplycurve.Inthelongrun,ofcourse,theBrazilianprice
ùmbrella',bymakingpossiblethesurvivaloflessefficientcompetitors,ended
upbyunderminingBraziliandominance.ButBrazil'sdominanceofthemarket
continueduptothe1960s.Brazilalsohadamajorshareoftheworldmarketof
rubber from the 1870s to the early 1910s. Thus, the country had degrees of
freedominitscommercialpolicywhichdidnotexistincommodityexporters
whichwereprice-takersinworldmarkets.Therewasscopefortheadoptionofa
highimporttariffas`theforeignerwouldpay'.Thisisanessentialelementfor
understandingthecoexistenceoveralongperiodofaveryprotectionisttrade
policyandagoodgrowthperformance.TheevidenceispresentedinTable6.1.
Inordertotesttheempiricalrelevanceofthishypothesis,astandardreduced
form equation for the determination of world coffee prices was estimated in
logarithmic form, using annual data for 1880 to 1960.
7
In the basic specifica-
tion,realcoffeeprices(PRICE)areafunctionoftheirownlaggedvalues,ofthe
laggedvaluesofavariablethattriestocapturethesupplydemandbalancein
the coffee market (MARKET) and a variable that captures the increase in pro-
ductioncostsassociatedwithhigherimporttariffs(COST).Thesupplydemand
balancevariableisconstructedastheratioofworldcoffeeconsumptiontothe
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table6.1 Brazil:coffeepriceregressions,18801960
a
Variable Equation1 Equation2 Equation3 Equation4
18801930 18801930 18801960 18801960
CONSTANT 2.59 3.43 1.24 1.00
3.58 3.54 (3.29) (2.38)
PRICE(1) 0.76 0.77 0.87 0.83
(9.93) (9.98) (16.95) (12.98)
MARKET(1) 0.50 0.48 0.33 0.37
(2.94) (2.77) (3.54) (3.66)
COST(-5) 0.52 0.26
(3.41) (3.11)
TARIFF(-5) 0.23 0:02
(2.26) (0:40)
RER(5) 0.57 0.22
(3.50) (2.79)
AdjustedR
2
0.79
Numberofobservations 51
Standarderror 0.17
0.78 0.84 0.84
51 81 81
0.17 0.16 0.16
a
Statisticsinparentheses.
Sources: Dataadjustedtocalendaryears.Coffeeprices:importsintoUSA,United
States(1975);worldcoffeeproduction,consumptionandstocks:BachaandGreenhill
(1992),statisticalappendix;domesticpricesUSA:UnitedStates(1975);exchangerates:
Brasil(1941);averagetariffs:computedfromBrasil(1941),Brasil(1990)andFritsch
(1988).
sum of world coffee supply and world coffee stocks. Therefore, an increase in
thisvariableshouldhaveapositiveimpactonrealcoffeeprices.Theinclusion
of this explanatory variable is justified by the special characteristics of the
coffeemarket.Sincetheproductcanbeeasilystoredandproductionresponds
topriceswithalagofseveralyears,astandardmodelwheresupplyanddemand
are functions of current prices and determine the market price through a
clearance condition is not appropriate for the case of coffee (De Vries, 1975).
The COST variable is constructed as the product of the real exchange rate
defined so that an increase in the index corresponds to a depreciation of the
domestic currencyand theimplicit tariffrate. Thisvariable enters theequa-
tionwithalagoffiveyears.Thereasonforsuchalagisthatinthebeginningof
the century production started, in general, four years after coffee trees were
planted, and some three-fourths of total costs in the coffee sector were asso-
ciated with fixed costs (Rowe, 1936, p. 37). Therefore, cost increases should
havemostofitsimpactoncoffeepricesaboutfiveyearslater.
Table6.1presentsthemainestimationresults.Equation1displaysthebase-
linecoffeepriceequationforthe18801930period.Allestimatedcoefficients
arestatisticallydifferentfromzeroatstandardconfidencelevels.Asexpected,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
realcoffeepricesincreasewhenconsumptionraisesinproportiontothesumof
coffee production and coffee stocks. The results show, that the COST variable
hastheexpectedpositiveimpactonrealcoffeepricesanditsestimatedcoeffi-
cientissignificantlydifferentfromzeroatveryhighsignificancelevels
ThereisachancethatthemostoftheeffectoftheCOSTvariableoncoffee
prices occurs through real exchange rate devaluations and not through tariff
increases.Inordertoexaminethispossibility,equation2inTable6.1separates
theCOSTvariableintoitstwocomponents:tariffs(TARIFF)andrealexchange
rate(RER).
8
Theresultsindicatethatthecostvariableisnotsimplyaproxyfor
realexchangeratedevaluations.Whentariffsandtherealexchangerateenter
the equation separately, both have the expected signs and their estimated
coefficientsarehighlysignificant.
Equations 3 and 4 replicate the econometric exercise for the 18801960
period. Most estimated coefficients remain highly significant, but when the
two components of the COST variable are separated in equation 4, the tariff
coefficientappearswiththewrongsignandisnotsignificantlydifferentfrom
zero.Thisshouldbeexpectedsinceafterthebalance-of-paymentscrisisfollow-
ing1930,exchange controls becametherule forthe wholeperiodupto1960,
andtariffsareaverypoormeasureofprotection.
The argument linking market power in the world coffee market and the
adoption of a high tariff is akin to that familiar in the literature, which links
world coffee prices and the Brazilian real exchange rate, or for that matter
worldpricesandtheexchangerateinanycommodityexporterwhichholdsa
substantialshareoftherelevantmarket.
9
Itisremarkablehowlittletherecognitionofsuchlinkshasbeenreflectedasa
qualificationofthestandardargumenton`socializationoflosses'.Inthecaseof
aprice-makersuchasBrazil,thedevaluationoftheexchangerateincreasedthe
amount of domestic currency generated by each unit of foreign exchange
received by exporters in the short run; but it also weakened the world prices
denominated in foreign currency, partly eroding the initial redistributive
impact mainly through the inducement to a release of stocks. In the long
term,exchangeratedevaluationhadmorecomplexeffects.Itimpliedinduce-
ments to increase production due to higher export revenues denominated in
domesticcurrency,aswellasinducementstoreduceproductionduetohigher
costsofimportedinputsdenominatedindomesticcurrency.Theeconometric
results show that after a lag of five years, the net effect of a devaluation is to
increaseprices,compoundingtheeffectofanincreaseintariffs.
Thegoldenageofimportsubstitution:1930toearly1960s
As elsewhere, the great Depression and the consequent balance-of-payments
shock provoked a significant shift in relative prices by making imports more
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
expensiveastheforeignexchangeratewasmassivelydevalued.Importcontrols
werealsoimposedasinmostdevelopingeconomiesandcanbeconsideredthe
mostimportantinstrumentofindustrialpolicy.Itisindeeddifficulttoexagger-
atetheimportanceofthetransitionin1930toaregimeofalmostpermanent
governmentinterventioninthedistributionofforeignexchangecover which
was to survive in other disguises until well into the 1990s. Domestic policy
changesinrelationtotheroleofthestatewerethuspromptedtoalargeextent
bychangedconditionsintheworldeconomy.
There were unexpected advantages in having adopted inefficient trade and
industrialpoliciesbefore1930.WiththeDepression,thescopeforexpenditure
switchingwasmuchamplerinrelativelymoreprotectionisteconomiesthanin
economies which had adopted more liberal commercial policy regimes and
lacked industrial capacity. These are important qualifications to the standard
evaluationofcommercialpoliciesstrictlybasedonstaticefficiencygrounds.
It waspart of the standard answerto theDepression in many economies to
maintain the exchange rate overvalued and meet excess demand for imports
with rationing. This was due to fiscal reasons, as governments feared the
impact of expenditures in foreign currency on their budgets, and, as already
mentioned, in countries with an important share ofcommodity markets, due
to the awareness that exchange devaluation in the short run did have an
unfavourableimpactonworldpricesdenominatedinforeigncurrency.
The impressive rates of GDP growth in Brazil from 1932 to 1937 relied to a
great extent on the increased importance of domestic industrial production
(see Table 6.2) favoured by devaluation and import controls imposed for the
government by the Bank of Brazil. The fall in import quantum following the
recessionwassubstantial:itreachedaminimumof40percentofthe1928level
in 1932, and towards the end of the decade was back around 8090 per cent.
From 1919 to 1939 the importdomestic supply ratio, using current prices,
decreased only from 25 per cent to 20 per cent. But the change in relative
prices, with imports becoming much more expensive in relation to domestic
prices,hidesthesharpadvanceofimportsubstitution.Using1939pricesthisis
nothingshortofspectacular astheimportstotalsupplyratiofellfrom45per
cent to 20 per cent. Even at current prices some subsectoral decreases of the
importsupplyratiosweresubstantial:from1214percentto2to5percentfor
textilesandfood,from60percentto40percentformajorintermediategoods
suchasmetallurgicalgoodsandchemicals,fromnearly100percentto6580
percentformechanicalandelectricalequipment(Fishlow,1972,pp.3234and
tablesIII,VIIandIX).
Fromtheendofthenineteenthcenturyindustrygainedgroundasapropor-
tion of GDP basically at the expense of agriculture with the share of services
maintainedroughlyaround50percent.Thistrendwasmonotonicafter1930
andmoremarkedinthe1950sthaninanyotherdecade(seeTable6.3).Dataon
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table6.2 Brazil:GDPandindustrialoutputgrowthrates,193080
Year GDP Industrial Year GDP Industrial Year GDP Industrial
output output output
1930 4.2 6.7 1947 3.9 3.3 1964 3.2 5.2
5 1931 3:8 1.2 1948
1932 3.7 1.4 1949
1933 11.1 11.7 1950
9.6 12.3 196 2.9 4.0
7.9 10.7 1966 6.7 11.7
6.6 12.3 1967 4.2 2.2
8 1934 9.1 11.1 1951 4.6 5.7 196 9.8 14.2
9 1935 3.8 11.9 1952 7.6 5.5 196 9.5 11.2
1936 12.5 17.2 1953 4.8 9.2 1970 10.3 12.0
1937 4.5 5.4 1954 7.9 9.1 1971 11.4 11.8
1938 3.6 3.7 1955 8.5 10.9 1972 11.9 15.1
1939 3.0 9.3 1956 3.0 5.7 1973 13.9 16
1940 2.0 2.7 1957
1941 5.5 6.4 1958
1942 3.7 1.4 1959
1943 10.1 13.5 1960
1944 8.0 10.7 1961
1945 2.9 5.5 1962
1946 11.6 3
.6
7.7 5.4 1974 8.2 7.8
10.7 16.8 1975 5.1 3.8
9.8 13.2 1976 10.2 11.9
9.4 10.7 1977 4.9 2.1
8.5 11.0 1978 4.9 6.2
6.7 8.0 1979 6.8 7.0
18.5 196 0.6 0.2 1980 10.2 9.2
Sources: 193047:Haddad(1978),sectoraldataaggregatedusing1947weights;194780:Brasil
(1990).Industrialoutput197175manufacturingindustryonly.
Table6.3 Brazil:GDPshares(%),
a
191080
Year Agriculture Industry Services
1910 35.8 14.0 50.2
1920 31.9 17.1 50.9
.5 1930 30.6 16 52.9
1940 25.0 20.8 54.2
1950 24.3 24.1 51.6
1960 17.8 32.2 50.0
1970 11.5 35.8 52.6
1980 10.1 40.9 48.9
a
GDPatfactorprices,includingfinancialintermediation.
Servicesincludegovernment.
Sources: seeTable6.2.
food production show that the supply per capita was roughly maintained
between1900and1930.Forthe193047periodfoodproductionincreasedat
a yearly rate of 2.7 per cent and outpaced population growth by 0.4 per cent
yearly (data from Haddad, 1978, pp. 601, and Brasil, 1990). From 1947 to
1980,withtheexceptionofbeans,andtoalesserextentbeef,theexpansionof
output of all items of popular consumption amply exceeded population
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
growth.WiththeexceptionofwheatBrazildidnotrelyonfoodimportstoan
important extent.
10
In contrast with food exporters Brazil was not an import-
ant consumer of its major exports, so that the traditional trade-off between
domesticconsumptionandexportswasnotfaced,atleastdirectly.
To a large extent the advance of domestic industrial production was due to
expandedproductionoftraditionalwagegoodsandtoalesserextentofinter-
mediate goods. The contribution of consumer durables and capital goods was
limitedandwouldcontinuetobesountilthe1950s.ThisisshowninTable6.4
bydataonindustrialproductionfromthe1919,1939,1949and1959censuses.
The share of consumer goods in total industrial value added fell, but not
spectacularly, between 1919 and 1939. Textiles still corresponded to 22 per
cent of industrial value added in 1939 compared to 24.4 per cent in 1919. It
was in food processing that the contraction was more substantial from 32.9
percentto23.6 percentofindustrialvalueaddedopeningspaceforincreased
production of intermediate goods and, to a much lesser extent, consumer
durablesandcapitalgoods.
Ascouldhavebeenexpected,importsubstitutionproceededinwaves,affect-
ing first the more basic industrial branches from the point of view of capital
divisibility,availabilityofmanpowerwiththerequiredskillsandaccesstotech-
nology:textilesandclothing,andfood-processing.Thenitreachedothercon-
sumergoods,intermediategoodsand,onlyafterlongmaturationandoftenwith
theinvolvementofforeigncapital,consumerdurablegoodsandcapitalgoods.
Table6.4 Brazil:distributionofindustrialvalueadded,191959
1919 1939 1949 1959
Consumergoods 80.2 69.7 61.9 46.6
Textiles 24.4 22.0 19.7 12.0
Clothing 7.3 4.8 4.3 3.6
Food 32.9 23.6 20.6 16.4
Other 15.6 19.3 17.3 14.6
Consumerdurables 1.8 2.5 2.5 5.0
Intermediategoods 16.5 22.9 30.4 37.3
Metallurgy 3.8 7.6 9.4 11.8
Non-metallicminerals 2.8 4.3 6.5 6.1
Chemical 0.8 4.2 4.7 8.3
Wood 5.7 3.2 4.2 3.2
Other 3.4 3.6 5.6 7.9
Capitalgoods 1.5 4.9 5.2 11.1
Mechanical 0.1 1.3 2.1 3.4
Electrical 0 0.3 0.8 1.0
Transportequipment 1.4 3.3 2.2 6.7
Source: Fishlow(1972),tablesIII,VIIandIX.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
From1937to1942theBrazilianeconomystagnated,followingtheUS1937
recessionandtheshocksrelatedtothewar(seeTable6.2).Butstartingin1942
the Brazilian economy entered a period of 20 years of almost continuous
growth in which industrial output growth and import substitution would
play a key role. However, due to the massive recovery of imports which took
placeafter1945,thecontributionofimportsubstitutiontoincreasedindustrial
production between 1939 and the early 1950s was insignificant. This is
reflected in the relatively slow change of industrial structure between
1939 and 1949 (see Table 6.4).
11
Industrial output almost always expanded at
highergrowthratesthanGDP;onlyinrecessionyearswasthisnotthecase(see
Table6.2).
During the Second World War the expansion of industrial capacity was not
veryimportantduetosupplyconstraints.ButthebuildingofVoltaRedonda, a
government-owned modern steel mill, which started to operate in 1945, con-
stitutedalandmarkinthehistoryofBrazilianindustrialization.Supportbythe
military was an essential element to explain the political mobilization which
made possible the construction of the mill during wartime. Not only was the
plant considered vital from a national security point of view, but military
personnel were directly involved in plant design, building and operation.
After1930andwellintothe1980spoliticalsupportby themilitarywasoften
essential to explain the high priority accorded to import-substitution indus-
trialization. Direct government involvement proved essential as foreign in-
vestors did not show interest, and was the first significant tangible
demonstration that direct government intervention could be effective in
industry.Whileduringthe1930ssuchinterventionwassignificantlyincreased,
partly reflecting ideological contrasts, it is interesting to note that Volta
Redonda was a result of state intervention by default as private investors were
unwilling or unable to be involved in the project. US support of the project
reflectedboththehighpoliticalpricewhichBrazil'sgoodwillcarriedinvirtue
of its strategic importance and the victory of political groups in Washington
whichsupportedimportsubstitutionindevelopingeconomies(Abreu,1984).
Policy based on an overvalued exchange rate, which had been adopted for
most of the 1930s and the early 1940s, became a pillar of Brazilian economic
policy after 1946. Ironically, it was the proponents of orthodox economic
policies who initially placed emphasis on maintaining a grossly overvalued
exchange rate with import controls due to the importance attached to price
stabilization. After a brief experiment without exchange controls, the imbal-
anceoftradewiththedollar areaandthelackofavailablereservesinconver-
tiblecurrencyledtotheirreimpositionin1947,againwith theBankofBrazil
playingthekeyroleinthedistributionofavailableexchangecover.Theimplica-
tions of such policies for the industrial sector have been recognized in the
literature. Industry would benefit from absolute protection due to import
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
controls,whichprohibitedimportsofcompetitivegoodsandalsofromaccess
to inputs and capital goods purchased at the overvalued exchange rate or
cheaperratesincaseofmultipleexchangerateregimes.Suchregimestendedto
benefit relatively more ISI in industrial sectors engaged in the production of
luxurygoods,ascompetitiveimportswerecertaintofailthecriteriaofessenti-
alitybasedonwhichimportcontrolswereimposed.
IntheyearsfollowingtheendofthewartheotherimportantissueinBrazil,
besidesstabilization,washowtofinanceinvestmentinaninfrastructuremuch
dilapidatedsince1930.Thedirectroleofthestatewasnowcrucialduetothe
negotiatedwithdrawalofmostBritishcapitalinvestedininfrastructure.There
wasmuchdisappointmentwiththeUSAwhosefinancialsupportwasdeemed
modest. In the early 1950s new government-controlled enterprises were cre-
atedmainlyintheenergysectoraswellasfinancialintermediariessuchasthe
BNDE(NationalEconomicDevelopmentBank),whichwastoplayakeyrolein
financing,first,infrastructuredevelopmentinthe1950s,andthentheexpan-
sionofselectedindustrialsectors.
Specific foreign exchange regimes varied between 1947 and 1964, but the
favourable distributive features from the point of view of domestic industry
remained in place. In the second half of the 1950s rapid industrialization
became the core of the government's economic strategy with emphasis on
the completion of import substitution in the consumer durable sector, espe-
cially in the auto industry, and making a big effort to reduce dependence on
imports of intermediate products. Government policies attracted foreign cap-
italinthesecondhalfofthe1950s,whenexportearningswerefallingrapidly
withthecollapseofcoffeeprices,byamassiveuseofsubsidieswhichincluded
thepossibilityofimportingcapitalgoodsatfavourableexchangeratesandthe
useoftheirmarketpowerbehindanextremelyhightariffwall.Fiscalsubsidies
andeasycreditwerealsomobilizedbyadhocsectoralgroupscreatedtoimple-
mentthePlanodeMetas.Fromthelate1940simportduties,whichwerespecific
andthuserodedbyinflation,hadbecomeirrelevantandprotectiondepended
on import controls or multiple exchange rate regimes. From 1957, however,
veryhighadvaloremimportdutieswereestablished,reachinginsomecases150
per cent. On top of this, imports were restricted by the `similar domestic
production'non-tariffbarrierembodiedinexchangecontrolrules.
Thetransferstotheindustrialsectorentailedbyexchangerateovervaluation
in194752havebeenestimatedasbeingbetween15percentand20percentof
income originating in the manufacturing sector. Implicit subsidies created by
incentives to attract foreign capital were also significant later in the 1950s,
rising from 2.1 per cent in 1955 to a peak of nearly 20 per cent in 195859
and still at 11.6 per cent in 1960 (Fishlow, 1975, tables IV and VIII). Income
transferstotheindustrialsectorrelatedtosuchforeignexchangepoliciesbefore
1964 were sizeable, but it is not clear to what extent they were transferred
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
toconsumers.Itisreasonabletoarguethatthesetransfersareinverselycorre-
lated to market power. The more competitive an industrial subsector is, the
more likely it is that consumers ended up appropriating the benefits of such
lowimportedinputandcapitalgoodscoststhroughlowerprices.
Wasthereanalternativetothecontinuationofanimport-substitutionstrat-
egy in postwar Brazil? Under the postwar system built in Bretton Woods and
Geneva industrialized countries performed quite well after initial difficulties.
The US answer to these difficulties was the Marshall Plan, launched in 1947,
whose transfers to Europe up to 1952 amounted to nearly US$14 billion,
approximately 5 per cent of the US GDP in 1948. Nevertheless, developing
countries continued tobe marginal in thisinternational reconstruction effort
andhaddifficultiesadaptingtothefixedexchangerateregime,beingcontinu-
ally forced to devalue their currencies or impose import controls due to
recurring foreign exchange crises. The situation was worse in Latin America
where prices of the major agricultural exports registered a steep downward
trend after having reached their peak during the Korean War in the early
1950s. This generated serious balance-of-payments problems for countries
thatreliedalmostentirelyuponthoseproductsforforeignexchangeearnings.
Thesituationofthesecountrieswasfurtheraggravatedbythelackofcapital
inflowssincetheUSA,theonlypotentialsupplier,hadassignedhighpriorityto
otherregionsoftheworld,andproposalsfortherevisionofthepolicytowards
Latin America were repeatedly denied by the State Department.
12
To obtain
foreignfinancing,countriesthatwerenotheldtobestrategicprioritiesforthe
USAdependedonloansfrominternationalinstitutions,butthevolumeofsuch
loans was almost negligible.
13
The perspectives for primary exporting nations
substantially worsened throughout the 1950s and 1960s. Agricultural com-
modities faced a rise of protection in the developed economies and, in the
case of Brazil, coffee prices collapsed after 1953, following their long-term
cyclicalpattern.
Anotherroadwasallegedlyopentodevelopingcountrieswhichbythe1950s
had already established an industrial base in processed food and light manu-
factures, accomplished basically through the import substitution of the inter-
warperiod.ItisarguedthataftertheSecondWorldWarthesecountriesshould
have taken advantage of their competitive edge in the production of those
goods due to low labour costs and inputs availability and promoted these
exports to developed countries. This strategy would have had the virtue of
providingamuchmoreefficientpatternofdevelopment comparedwithcon-
tinuingimport-substitutingindustrialization.
Theargument,however,ignoresthestateoftheworldeconomythroughout
the postwar period. The possibility of following a manufactured-based, out-
ward-oriented strategy was not available at least until the mid-1960s; the
expansionoftradeinthe1940sand1950swaslargelylimitedtotradebetween
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
industrializedcountries.Infact,therewasaclearlyidentifiablepatternamong
the industrialized nations by means of which not only nominal tariffs, but
effective protection grew in inverse proportion to the sophistication of the
productionprocess:themorecapitalandtechnology-intensivetheproduction
process, the lower the required level of protection.
14
In labour-intensive pro-
ductionprocesses,wheresomedevelopingcountrieshadanobviouscomparat-
ive advantage, the effective level of protection in developed economies was
inordinatelyhigh.Inturn,productsthatrequiredmoresophisticatedtechnol-
ogy such as electrical and non-electrical machinery and transport equipment
facedlowernominaldutiesandeffectiveprotection.Thereisalsoevidencethat
non-tariffbarriers were disproportionately leviedon exports ofless-developed
countries(Balassa,1965,andWalter,1971).
Thecritiqueassumesfurtherthatsuchanexport-promotionstrategywasnot
pursued due only to policy-makers' miscalculations regarding the capacity of
international markets to absorb developing nations' exports. Accordingly, the
superiorityoftheexport-promotionstrategyisallegedlyillustratedbytheEast
Asian experience with an outward-oriented, distortion-free strategy in the
1960s. The major problem with this critique is that it essentially ignores his-
tory. The shift in Asia towards an outward-oriented strategy occurred only in
the 1960s; the world economy in the 1950s presented real obstacles for an
outward-oriented strategy. Brazil and other more advanced Latin American
economiesventuredintoanewphaseofimport-substitutingindustrialization,
yet import substitution did not haveto be incompatible with amore rational
exportpolicy.IntheBraziliancase,thegovernmentcouldhavecreatedincent-
ivestopromotetheexportsofmanufacturedgoods,suchascottontextiles,in
whichthecountryalreadyhadsomeexportingexperience.
The golden age of import substitution in Brazil was roughly between the
early1950sandtheearly1960s.ISI,whichinthepasthadmainlyaffectedwage
goods and intermediate products, also became relevant for durable consumer
goods such as household appliances and, from the mid-1950s, motor cars.
Domestic motor car production stimulated the establishment of a complex
systemofsuppliersofpartsthusinducingtheadoptionofhigherstandardsof
production and management. Other state enterprises were established in the
early1950s,especially suppliersofindustrialinputsandservices,
15
affecting a
widerangeofsectorssuchassteel,petrochemicals,fuelsandelectricity.About
31 per cent of industrial growth between 1949 and 1962 was due to import
substitution.
16
Theimportsupplyratiowhichhadstoodat14percentin1949
fellto6percentin1964.Formachineryonlyitfellfrom65.7to30.9percent
(WorldBank,1983,p.35).Industrialgrowthbetween1958and1961wasatthe
average yearly rate of 11.9 per cent (see Table 6.2), comparable to earlier
industrial booms in 193336 and 194850 when average yearly rates were 13
and11.6 percentrespectively.Buttheboomofthelate1950slastedlongerand,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
moreimportantly,itwasbasedneitherontheutilizationofidlecapacitynoron
theexpansionoftraditionalsubsectors.AscanbeseenfromTable6.4,thiswas
the period when it is possible to speak of a decisive change in industrial
structure.
Suchaprocesswasinterruptedbyarecessionin1962,andthereisanample
literature on its causes. The limits to import substitution as an engine for
growthhavebeenunderlined;insomecasesusehasbeenmadeofarguments
aboutincreasingmarginalcapitaloutputratiosduetosubstitutionofimports
inmorecapital-intensivesectors,orthelargeoptimalsizeofplantsinrelation
to the Brazilian market. Others have emphasised the lack of compatibility
betweensupplyanddemandstructuresduetotheemphasisontheproduction
ofconsumerdurableswhichwasespeciallymarkedaftertheimplementationof
Kubitschek'sPlanodeMetas.
17
Importsubstitutionpetersout:early1960stolate1970s
With the recovery of the balance-of-payments position after 1964 there was
once again negative import substitution. The importtotal supply ratio for
manufacturing industry increased from 6.1 per cent in 1964 to 11.9 per cent
in1974,butwasstillbelowits1949level.Formachinerytheratiofrom1964to
1974 was roughly constant at around 3032 per cent. This was a surprisingly
slowtrendinthedirectionofamoreopeneconomygiventherecoveryofthe
balance-of-paymentspositionandtheavowedemphasisofpoliciesonopening
uptheeconomy(WorldBank,1983,p.35).
In any case, 1964 marked a shift towards a more open economy as many
distortionswereremovedorreducedbyextensivepolicyreform,includingthe
foreign exchange regime. The new approach to trade policy relied on the
implementationofseveralincentivestomanufacturedexports(see,forexample,
Neves and Moreira, 1987). Subsidies in some years were equivalent to a deva-
luation of more than 70 per cent of the ruling foreign exchange rate. Rent-
seeking,whilecontinuingtopursueasanobjectivelowcostnowintheformof
discretionary exemption of import duties, rather than the more overvalued
categories of exchange rates, started to include export incentives as a target.
Theresultsofthesepolicymeasureswerequiteimpressive.Therateofgrowth
of manufactured exports was twice as great asthat of overall exports between
1965 and 1975, and well above the rate of expansion of international trade.
Table6.5showsthedramaticimpactofexportpromotionofmanufactureson
exportstructurefromthe1960s(WorldBank,1983,pp.1848,andNeves,1985,
p.245).Exportswhichexpandedthemostintheperiodwerepreciselythoseof
sectors which were either established or consolidated during the postwar ISI:
motor vehicles, communications and transport equipment, ship-building,
iron,steel,basicchemicals,andaircraft(WorldBank,1983,pp.191).Brazilian
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
exportperformanceafterthe1960swouldnothavebeenpossiblewithoutthe
industrializationeffortwhichprecededitasexportgrowthwaslargelybasedon
sectorsestablishedthroughISIinthe1950s.
But there was only a half-hearted commitment to increased openness. The
more liberal import tariff regime of 1967 was partly reversed in the following
yearsyieldingtopressurebyestablishedrent-seekinginterests,includingmulti-
nationals in a prominent role (Coes, 1991, pp. 479 and ch. 5). A high tariff
remained in place as a crucial factor in the attraction of direct investment as
therewasamplescopefortheuseofmarketpowertoextracthighprofits.These
`miracle years' from 1968 to 1973 seem to have corresponded to a missed
opportunity in terms of an overhauling of traditional ISI. After the first oil
shock the government followed a policy of explicitly deepening IS to reduce
theeconomy'svulnerabilitytoexternalshocks.
From 1974 to 1979, the contribution of import substitution to industrial
growth was again positive but extremely limited: no more than 10.1 per cent
for industry as a whole. It was more important for capital goods explaining
16.1 per cent of growth less so for intermediate goods (14.6 per cent) and
negligible (2.5 per cent) forconsumer goods. In 197585, import substitution
was negligible as an explanation for industrial output fluctuations (Bonelli,
1986).Importsubstitutionasasourceofgrowthwasexhaustedbytheprogress
ofimportsubstitutionitself.
The decision-making processes involving industrial and trade policies
remained very closed for the whole period under consideration. While much
lip-service was paid to the virtues of a liberal system, the state played a
dominant role in shaping industrial policies. Decision-making was heavily
influenced by the military and in many instances national security reasons
Table6.5 Brazil:percentageshareofcoffeeand
manufacturedexportsintotalexports,193080
Coffee Manufactured
products
1930 62.9 0
1940 32.0 13.5
a
1950 63.9 1.1
b
1960 56.1 2.6
b
1970 34.2 11.2
1980 12.3 44.8
a
Estimated:classesIII(animalproducts)andIV,AEB
(194045),pp.2624.
b
Estimated:classesVtoVIII,Brasil(1961),p.83.
Sources: Brazil(1961),andBancoCentral,Relatorio,
severalissues.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
wereinvokedtodefinepoliciesandtheroletobeplayedbydifferentinterests.
A common format was the partnership in joint ventures of the state, foreign
capital and the domestic private sector. By the late 1970s, ISI had resulted in
the creation of an extremely diversified industrial structure producing a wide
range of goods which included motor cars, aircraft, armoured vehicles and
most types of capital goods. But many projects promoted during these years
proved to be unable to survive when in the 1980s government support was
reducedandsomedegreeofcompetitionwasfosteredbytheopening-upofthe
economy.
The government has proved to be a poor representative of interests which
were underrepresented or simply not represented in the decision-making
process. These excluded interests are generally those who paid the bill. Even
today,severalyearsafterthefullrestorationofdemocraticrule,thediscussion
of sectoral policy issues in the so-called `camaras setoriais', which involve
government,domesticproducersandtradeunions,areinsomesensedeemed
tobe`democratic',evenbyoppositionparties.Nowonderthatinmostsectoral
issuesthereisaclosecoalitionofaffectedproducersandworkerstotheexclu-
sion of consumers and/or taxpayers. The difference in relation to the past, of
course, is that trade unions were not engaged in such negotiations before the
mid-1980s.
It is reasonable to suppose that there are significant links between the ratio
of capital goods imports in total investment and GDP growth rates (Carneiro
and Werneck, 1993). The underlying idea is that the lower this ratio is, the
higher the capitaloutput ratio will be, as domestic substitutes become more
expensive and/or are less efficient than imported capital goods. Trends of
the relative cost of investment and specifically on domestically produced
capital goods seems to lend support to this hypothesis. The ratio between
the investment deflator and GDP deflator increased by about 40 per cent
fromthemid-1970stotheearly1980s.Bytheendofthe1980sithadreached
100 per cent. The ratio of imported capital goods in total investment,
which was typically 1314 per cent in 197175, fell below 6 per cent for
most of the 1980s. Preliminary econometric results indicate that, in
addition to capital stock and labour, the share of imported capital goods
intotalinvestmentisasignificantvariableexplaininggrowthinthelongrun
inBrazil.
Ifthe197480and198191periodsarecompared,itisapparentthatonlya
very low proportion of the sharp fall in GDP growth rate can be linked to a
reduction in savings, as the savings ratio fell only very modestly. About two-
thirdsofthefallinGDPaveragegrowthfrom7.2percentto1.5percentis
linkedtotheriseinthecapital-outputratioandone-thirdtotheincreasedcost
ofinvestment.Between196773and197480,astherewasaslightincreasein
the savings ratio with a constant cost of investment, the fall in the average
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
growth rate is wholly related to a deterioration of the capitaloutput ratio.
Reasons for increased capitaloutput ratios in the 1970s, and even more in
the 1980s, are partly rooted in the imperfect substitution of imported capital
goods by domestically produced substitutes. Also of paramount significance
were the microeconomic consequences of the major macroeconomic imbal-
ances faced in the period which resulted in curtailment of excessively ambi-
tiousinvestmentplans,extensionofperiodsofmaturationofinvestmentand
sharpreductionsinplannedratesofreturn.
Itisimpossibletodealadequatelywiththelinksbetweensocialdevelopment
andimportsubstitutioninthecontextofthischapter.Totheextentthathigh
growthandimportsubstitutionwerelinkedatleastuntiltheearly1960s,one
cansaythattheevolutionofsocialindicatorswasratherreasonable,especially
in the 1950s, if compared with stagnation in the 1960s and a rather slow
improvementthereafter(seeTable6.6).Itiswell-knownthatBrazilisanoutlier
if comparison is made between income per capita and the level of the main
socialindicators.Givenitspercapitaincome,itshouldhavemuchbettersocial
indicators(Abreu,1987).Butthisprocessofdeteriorationaswellasthegrowing
inequalityinthedistributionofincomewouldseemtohavespeededupafter
thegoldenageofimportsubstitution.Similarly,Brazilhasalsobeenshownto
beverybackwardinrelationtoeducationindicators. In1940theliteracyrate
was only 43 per cent and in 1980 still only 74.5 per cent. It may perhaps
be argued that at least to some extent, ISI-related public expenditure
crowded out expenditure in social programmes, but there is no indication
that if ISI-induced expenditure had been lower this would have resulted in
increased social expenditure. An important link between ISI and social
development was through the possibility of significant expansion of the
population employed in the industrial sector where productivity was about
four times higher than in agriculture in the census years of 1940, 1960
and 1980. Between 1940 and 1970 the increase in the population employed
intheindustrialsectorwasalmost350percentwhileinagricultureitwasonly
35percent.
Table6.6 Brazil:lifeexpectancyatbirth,
inyears,194080
Year Lifeexpectancy
1940 42.7
1950 45.9
1960 52.4
1970 52.7
1980 62.0
Sources: Abreu(1987).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Conclusions
Some basic ideas have been advanced in this chapter as a basis for an under-
standingofimportsubstitutioninBrazilinthelongrun.Somearenew,some
well-established in the literature. It is suggested that high protection was
crucial for the genesis of Brazilian industrialization and deepening of ISI. The
explanationadvancedhere,thathightariffswerepossiblebecauseofBrazilian
market power in the coffee market, is of crucial importance to explain why
substantialindustrializationoccurredinBrazilwithouthurtingexportproceeds
inthemid-term.ItisalsousefulasanexplanationforthepersistenceinBrazil
of the idea that high protection of the domestic market and good growth
performancearelinkedwithouttakingintoaccountthedecreasingimportance
ofcoffeeasashareofGDP.
Ithasalsobeenarguedthat,giventheinternationalcontext,therewaslittle
roomforanalternativeindustrialstrategyforBrazilintheearly1950s,evenif
exportpessimismresultedinanalmosttotaldisregardofexportincentives.The
importance of import substitution as an engine of growth between the late
1940s andearly 1960s hasbeenstressed, aswasthe idea thatimportsubstitu-
tionafterthemid-1960skilleditself,sinceimportpenetrationwassolowthat
thedomesticmarketandexportsweretoplayafortiorithemostimportantrole
inexplainingindustrialgrowth.
While changes in the international environment and the emergence of
populism implied an increased importance of state intervention, there were
important elements of continuity in such intervention as ISI in Brazil was
importantlongbefore1930.Thecrucialroleofthemilitaryintheconstitution
ofasemi-competitiveenvironmentafterthe1960shasalsobeenemphasized.
Finally, it has been suggested that it was likely that as at late stages import
substitution affected capital goods, there was an adverse impact on growth.
This is because domestically produced capital goods are more expensive or
technologicallylessadvancedthancompetitiveimports.Capitaloutputratios
areincreasedandsuchconsequencescombinetoproduceadeteriorationofthe
macroeconomic situation and result in stagnation in spite of relatively high
savingsandinvestmentratios.
Notes
1 In spite of this emphasis, Furtado (1959) also mentions that import-substituting
industryafteritsinstallationhadavestedinterestinforeignexchangedevaluation.
2 SeeAbreu(1994),table1foralternativeestimates.BrazilianGDPpercapitain1913
wasaboutafifthofthatofArgentina.
3 SeeTaylor(1996)andFrankelandRomer(1996).
4 ThissectiondrawsfromAbreu(1994).
5 Dean(1969,pp.378),basedonBandeiraJr(1901).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
6 SeeStein(1957,pp.967)fortheirabilitytodosointhe1890sand1900s.
7 Between1880 and 1930 the average Brazilian share of world coffee productionwas
about67percent.Thisfellto56percentbetween1930and1960.
8 Granger causality tests show that one cannot reject the hypothesis that there is no
causal link between changes in real coffee prices and changes in the Brazilian
exchangerate.
9 For early perceptions of such a link see Gudin (1933) and Williams (1934). See
Cardoso(1983)foranempiricalinvestigationofthemoretraditionalcausality,that
isoftheBrazilianexchangeratebeingexplainedbycoffeeprices.
10 AttemptstoextendIStowheatproductionfailedspectacularlyinthe1950sandonly
succeededinthe1970sand1980sunderextremesubsidization.
11 Malan, Bonelli, Abreu and Pereira (1977, p. 349). The import quantum doubled
between 1944 and 1947 and remained at least 70 per cent above the 1944 level in
194850.
12 The`Sovietthreat'wasthesinglemostimportantelementintheallocationofaid,as
thefollowingofficialmemoillustrates:`SecretaryDullessaidthatofcourseallofus
would like to see our economic objectives in the under-developed countries
achieved through the use of private capital investment. But some of the most
critical of these under-developed countries existed under conditions where they
will have to be able to see genuine hope of a transformation provided by the
West, or else they will turn to the USSR. So large were these under-developed
areasthatiftheyturntotheSovietUniontheareaoftheFreeWorldwillshrinkby
another two-thirds. Accordingly we have got to provide economic development
assistance,and furthermore, we must as a nation realise more fully the importance
of this assistance for our national security'; in `Review of Basic National Security
Policy: Foreign Economic Issues Relating to National Security', in United States
(1987),p.182.
13 Brazil,moreover,hadverydifficultrelationswiththeIMFandtheWorldBankfrom
theearly1950sto1964.
14 SeetheUnitedNations1962specialreportondevelopingnations'trade:`thehighest
tariffratesarethoseappliedtoconsumersgoods(cottontextileandfootwear)which
compete with domestic production in developed countries', and further it was
observed that ìndustrial countries maintain a clear progression in their tariff rates
accordingtothedegreeofprocessing',UnitedNations(1962,PartI,pp.667).
15 SeeShapiro(1994)foracomprehensiveanalysisoftheautomotiveindustryinBrazil
sincethe1950s.
16 Fishlow(1972).WorldBank(1983)estimatestheratioat24percentfor194964.
17 UnitedNations(1962)andFurtado(1968).Forashortreviewofexplanationsofthe
19627downturnseeAbreu(1990),pp.2089.
References
Abreu,M.deP.(1994)`ThePoliticalEconomyofProtectionisminArgentinaandBrazil,
18801930',inP.H.Lindert,J.V.NyeandJ.Chevet(eds),PoliticalEconomyofProtection-
ismandCommerce,Eighteenth-TwentiethCenturies,ProceedingsoftheEleventhInterna-
tionalEconomicHistoryCongress,SectionB7,Milan:Bocconi.
(1990)Ìnflacao,EstagnacoeRuptura:19611964',inM.deP.Abreu(ed.),AOrdem a
do Progresso: Cem Anos de Pol tica Economica Republicana, 18891989, Rio de Janeiro:
EditoraCampus.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
(1987) `Stability andSocialPolicyin Brazil:TheWay Ahead',Bulletinof LatinAmer-
icanResearch,vol.6 (2).
(1984)ÒBrasileaEconomiaMundial(19291945)',inB.Fausto(ed.),HistoriaGeral
daCivilizacoBrasileira,tomoIII,vol.4,SaoPaulo:Difel. a
Bacha,E.andGreenhill, R.(1992) 150AnosdeCafe noBrasil, RiodeJaneiro:Marcellino
MartinsandE.Johnston.
Balassa, B. (1965), `Tariff Protection in Industrial Countries: An Evaluation', Journal of
PoliticalEconomy,vol.lxxiii,no.6,December.
BancoCentraldoBrasil,Relatorio,variousissues.
BandeiraJr,A.F.(1901)AIndustriadeSaoPauloem1901, SaoPaulo.
Bonelli, R. (1986) Ensaios sobre Pol tica Econo a mica e Industrializaco no Brasil, Rio de
Janeiro:ServicoNaiocnaldeAprendizagemIndustrial(SENAIDN/DITEC/DPEA).
Brasil(1990)FundacoInstitutoBrasileirodeGeografiaeEstat stica,Estat sticasHistoricas a
do Brasil. Series Economicas, Demograficas e Sociais de 1550 a 1988, 2nd edn, Rio de
Janeiro.
Brasil(1961)InstitutoBrasileirodeGeografiaeEstat stica,BrasilemNumeros1960,Riode
Janeiro.
Brasil(1941)InstitutoBrasileirodeGeografiaeEstat stica,AnuarioEstat sticodoBrasilAno
V1939/1940,RiodeJaneiro.
Brasil, Fundaco Instituto Brasileiro de Geografia e Estat stica, Contas Nacionais, various a
issues.
Cardoso, E. (1983) Èxchange Rates in Nineteenth-Century Brazil: An Econometric
Model',JournalofDevelopmentStudies,vol.19(2),January.
Carneiro,D.D.andWerneck,R.F.(1993)ÒbstaclestoInvestmentResumptioninBrazil',
in E. Bacha (ed.), Savings and Investment Requirements for the Resumption of Growth in
LatinAmerica,WashingtonDC:Inter-AmericanDevelopmentBank.
Coes,D.(1991)`Brazil',inD.Papageorgiou,M.MichaelyandA.M.Choksi(eds),Liberal-
izingForeignTrade,vol.4,TheExperienceofBrazil,ColombiaandPeru,Cambridge,Mass.;
Oxford:BasilBlackwell.
Dean,W.(1969)TheIndustrialisationofSaoPaulo18801945,Austin:UniversityofTexas
Press.
De Vries, J. (1975) `Structure and Prospects of the World Coffee Economy', World Bank
Staffworkingpaperno.208,June.
ECLAC (1964) `The Growth and Decline of Import Substitution in Brazil', Economic
BulletinforLatinAmerica,vol.9,March.
Fishlow, A. (1972) Òrigins and Consequences of Import Substitution in Brazil', in L.E.
DiMarco(ed.),InternationalEconomicsandDevelopment.EssaysinHonorofRaulPrebisch,
NewYork:NationalBureauofeconomicresearch
(1975) `Foreign Trade Regimes and Economic Development: Brazil', mimeo, Berke-
ley:UniversityofCalifornia.
Frankel, J.A. and Romer, D. (1996) `Trade and Growth: An Empirical Investigation',
NationalBureauofEconomicResearch(NBER)workingpaperno.5476,March.
Fritsch, W. (1988) External Constraints on Economic Policy in Brazil, 18891930, London:
Macmillan.
Furtado,C.(1968)UmProjetoparaoBrasil,RiodeJaneiro:Saga.
(1959)FormacoEcono a micadoBrasil,RiodeJaneiro:FundodeCultura.
Greenaway,D.andNam,C.H.(1988)ÌndustrialisationandMacroeconomicPerformance
inDevelopingCountriesunderAlternativeTradeStrategies',Kyklos,vol.41(3),August.
Gudin,E.(1933)CambioeCafe,RiodeJaneiro:Laemmert.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Haddad,C.(1978)CrescimentodoProdutoRealBrasileiro,19001947,RiodeJaneiro:Foudo
deGetulioVaugas(FGV).
Maddison,A.(1989)TheWorldEconomyinthe20thCentury,Paris:OECD.
Malan,P.S.,Bonelli,R.,Abreu,M.deP.andPereira,J.E.C.(1977)Pol ticaEconomicaExterna
eIndustrializacodoBrasil:19391952,RiodeJaneiro:InstitutodePesquisaEcono a mico-
SocialAplicada/InstitutodePesquisas(IPEA/INPES).
Neves,R.B.(1985)ExportaceseCrescimentoIndustrialnoBrasil,RiodeJaneiro:Institutode o
PesquisaEconomico-SocialAplicada/InstitutodePesquisas(IPEA/INPES).
o Neves,R.B.andMoreira,H.(1987)ÒsIncentivosasExportacesBrasileirasdeProdutos
Manufaturados19691985',PesquisaePlanejamentoEconomico,vol.17(2),August.
Nunes,R.C.B.andSilva,J.R.(1929)TarifadasAlfandegas.Annotada,CommentadaeExpli-
cada pelos Conferentes da Alfandega do Rio de Janeiro, vol. 1, Rio de Janeiro: Oficinas
GraphicasdoJornaldoBrasil.
Rowe,J.W.F.(1936)MarketsandMen.AStudyofArtificialControlSchemesinsomePrimary
Industries,Cambridge:CambridgeUniversityPress.
Shapiro,H.(1994)EnginesofGrowth.TheStateandTransnationalAutoCompaniesinBrazil,
Solberg,C.E.(1987)ThePrairiesandthePampas.AgrarianPolicyinCanadaandArgentina,
18801930,Stanford,California:StanfordUniversityPress.
Stein,S.J.(1957)TheBrazilianCottonManufacture.TextileEnterpriseinanUnderdevelopment
Area,18501950,Cambridge,Mass.:HarvardUniversityPress.
Suzigan,W.(1986)IndustriaBrasileira.OrigemeDesenvolvimento, SaoPaulo:Brasiliense.
Taylor, A. (1996) Òn the Costs of Inward-looking Development: Historical Perspectives
onPriceDistortions,Growth,andDivergenceinLatinAmericafromthe1930stothe
1980s',NationalBureauofeconomicresearch(NBER)workingpaper.
UnitedNations(1962)WorldEconomicSurvey,NewYork.
UnitedStates,DepartmentofState(1991)ForeignRelationsoftheUnitedStates,19551957,
Washington,DC:GovernmentPrintingOffice.
United States, Department of Commerce (1975) Historical Statistics of the United States.
ColonialTimesto1970,Washington,DC:GovernmentPrintingOffice.
Villela, A. (1993) `Pol tica Comercial e Importaces na Primeira Repu o blica: 18891930',
MAdissertation,DepartmentofEconomics,CatholicUniversity,RiodeJaneiro.
Walter,I.(1971)`NontariffBarriersandtheExportPerformanceofDevelopingEconom-
ies',AmericanEconomicReview,May.
Williams, J.H. (1934) Àmerican Foreign Exchange Problems in Brazil, Argentine, Chile
andUruguay,1934',reprintedinForeignRelationsoftheUnitedStates1934,Washington,
DC:GovernmentPrintingOffice.
WorldBank(1987)WorldDevelopmentReport1987,NewYork:OxfordUniversityPress.
(1983)Brazil,IndustrialPoliciesandManufacturedExports,Washington,DC.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
7
TheProcessofAccelerated
IndustrializationinMexico,192982
EnriqueCardenas*
Introduction
TheeconomichistoryofMexico'sindustrializationdatesbacktotheendofthe
nineteenth century when the first large industrial firms were established.
Although these businesses were so huge that they would satisfy at least 30 or
40 per cent of the domestic market, they only touched a relatively small
segmentofsociety.Butduringthetwentiethcentury,aftertheMexicanrevolu-
tion(191017),industrializationreflectstoalargeextentthetransformationof
thewholeeconomy,withitscorrespondingimpactonthepoliticalsystemand
society as a whole. It encompasses the way by which industrial activity
becomes important for a larger share of the population, its relevance to urb-
anizationandthetransformationofagricultureandotherprimarysectors,the
relationship of the private sector and labour with the government, and the
greaterroleoftheinternationaleconomy.
There are different ways to approach the subject, all of which are to some
extentarbitrary.Oneofthemisbylookingattheeconomic,politicalandsocial
forcesthatstimulatedtheprocessofindustrialization.InMexico,aswellasin
othercountriesofLatinAmerica,boththemarketandthestateplayedamajor
role.Anexpandingdomesticmarket,stimulatedbypopulationgrowthandthe
more extensive use of other resources created opportunities for private in-
vestors. Public sector activism included the implementation of policies that
aimed to protect domestic producers and stimulate the transfer of resources
* A large part of the content of this paper is the result of several years of research on the economic
historyofindustrialization,economicpolicyandgrowth,whichhasbeen,orwillsoonbepublished.I
thereforedrawextensivelyfromthosesources.IwishtoexpressmygratitudetoRosemaryThorpand
V ctor Urquidi for insightful comments throughout the paper, as well as to Sergio Negrete and the
participants of the workshop that originated this volume. Naturally all remaining responsibility is
mine. I also want to express my appreciation to Rodolfo Martell and Elia Sol s for their help in
producingthetablesandgraphsincludedinthistextandforreviewingthemanuscript.
176
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 177
fromprimaryproductiontourbandevelopment.Additionally,thegovernment
made alliances with the business class and with organized labour to enhance
industrial growth, among other means. During the twentieth century, the
processofindustrializationinMexicohasbeenstimulatedbydomesticfactors,
suchasthegrowthofpopulationandthecorrespondingurbanizationprocess,
andbytheexistenceofincentivestoinvestsuchastradeprotectionandfiscal
subsidieswhichimpliedhigherrelativeprofitratesinindustry,amongothers.
Industrialization has also been stimulated, or retarded, by external shocks, by
the availability of foreign capital and demand, and by the development of
financial and other institutions created for that purpose; that process has
beengovernment-sponsoredattimes,andmarket-ledinotherinstances.
The purpose of this chapter is to analyse the process of accelerated indus-
trializationinMexicofromtheGreatDepressiontotheendoftheoilboomand
thecollapseoftheeconomyasaconsequenceofthedebtcrisis.The1930sare
only briefly reviewed here since their full treatment is well-developed in the
previous volumes of the series. The approach is chronological and therefore
considers thedifferent stagesofindustrialization, dependingon thenatureof
themaineconomicforcesthatstimulatedtheprocess.Therefore,thechapteris
dividedintofoursections:thefirstbrieflycoversthe1930swhenindustrializa-
tionwasdrivenbymarketforcescausedbytheimpactoftheGreatDepression
in general, and by the real exchange depreciations in particular. The second
sectioncoverstheSecondWorldWarperiodwhenindustrializationwasmainly
pushed by external demand, contrary to conventional wisdom, and by the
strong building andcapitalization of economic infrastructure financedby the
state. The third section covers from the end of the war until the early 1960s,
when the model of import substitution was established, forced by circum-
stancesatfirst,andthengraduallyevolvingintoawholedevelopmentstrategy
based on protection from foreign competition, alliances between the private
sectorandthegovernment,resourcetransfersfromagriculturetoindustryand
theexpandinginternationalmarketsforMexico'sexports.Theperiod196271
ischaracterizedbyanumberofadditionalprotectionistmeasuresthatnotonly
expanded protection, but also implicitly meant a development path that was
bound to collapse eventually. Although some efforts were made to increase
exports, the path finally created a number of macroeconomic imbalances as
wellasmicroeconomicdifficulties,exacerbatedbythedemographicboomand
the increase in the labour force as domestic savings were increasingly insuffi-
cient to finance the required investment to generate the necessary jobs. The
only way to sustain it for some years more was by the exploitation of major
national resources and by borrowing abroad, which by definition also had its
limits, which were finally reached in the early 1980s with the debt crisis. The
last section considers the 1970s and early 1980s, a period that witnessed a
process referred to as `Mexico's economic populism', when rapid economic
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
178 IndustrializationinMexico,192982
growth was only made possible by the huge additional dollar revenues that
came from the renewed exploitation of oil resources and from the lending
boom that occurred in the 1970s and early 1980s. When those sources began
todiminish,theeconomyenteredintoadepressionpaththatithasnotbeen
abletoovercome.Letusnowturntothosestages.
Market-ledimportsubstitutioninthe1930s,192939
TheGreatDepressionof1929thatstartedintheUSAhadanenormousimpact
on the world economy, and particularly in Latin America, Mexico being no
exception. Mexico's exports declined 64.9 per cent between 1929 and 1932,
whilethetermsoftradefell20.8percentduringthesameperiod.Inturn,the
fallinMexicanexportsreduceditscapacitytoimport,andconsequentlythere
was a growing trade deficit that had various consequences. On the one hand,
the export sector, which was formed essentiallyby oil andmining companies
in foreign hands, nevertheless had a considerable domestic component in its
gross value and, consequently, the decline in exports did have a negative
impactondomesticnationalincome.
1
In addition, the trade deficit generated multiple depressive monetary and
fiscal effects. First, because the monetary system was based on gold andsilver
withoutbankbills,theyawningtradeaccountdeficitimpliedanoutwardflow
of silver and mainly gold as foreign exchange because most commercial part-
ners of Mexico were under the gold-exchange standard.
2
As international
reserves began to fall, that created a reduction in the money supply with the
correspondingdepressiveimpact.Between1929and1931,nominalmonetary
balancesdeclined60.2percentandrealinterestratesincreased.
Thesecondimpactwasinpublicfinances.SinceMexicohadfallenintodebt
moratorium since the revolution, there had been no access to foreign credit.
NeitherdidithavedomesticcreditinspiteoftheestablishmentoftheBancode
Mexico in 1925 and its corresponding power to issue money, for the public
distrusted paper money (Cardenas and Manns, 1987, pp. 37594). Con-
sequently, a decline in fiscal revenues had to be matched by a reduction in
publicspending.Asapproximately50percentoffiscalrevenuesoriginatedin
the foreign sector, areduction in thevalue of international trade had adirect
impactonfiscalrevenuesandthereforeonpublicexpenditure.Forthatreason,
theGreatDepressionbroughtwithitareductioningovernmentspendingwith
adeflationarymultiplier effect. Governmentspending,especially investment,
declined23percentfrom276 millionpesosto212millionpesosbetween1929
and1932,andthisactuallyproducedsmallbudgetsurpluses(Cardenas,1987,
pp.8891).
TherealimpactoftheGreatDepressionwasademandshockthatproduceda
declineinGDPof18percentbetween1929and1932.Agriculturedidnotsuffer
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 179
asmuchbutthereductioninindustrialoutputreached31percent.Unemploy-
ment increased and real wages declined, aggravated by the deportation of
approximately 300 000 Mexican migrant workers from the USA (Carreras de
Velasco,1974,pp.1734),whichweakenedevenfurtherthelabourmovement.
Fearingpublicdisapprovaloroutrightrejection,inearlyDecember1931the
BancodeMexicoacceptedratherreluctantlytheissueofbankbillswithsilver
orsilver-denominateddocumentsinsteadofgold.Giventhescarcityofmeans
ofpayment,thepublicreactedimmediately andbeganabsorbingnewbillsto
the extent that the supply of banknotes increased from 0.1 million pesos in
November 1931 to 4 million three months later.
3
To reactivate aggregate
demand, the government also started paying bureaucrats and the army with
Banco de Mexico bills, so that by the end of 1932 the issue of notes reached
42.1millionpesosandthecoinageofsilvercoinshadbeenrestarted.
Butthemostimportantfactor,onethatnotonlycontributedtotherecovery
butalsohadaprofoundlong-termstructuraleffect,wastherealdepreciationof
the exchange rate. The fact that Mexico was only formally under the gold
standard, but de facto had a bimetallic exchange rate system since most do-
mestic transactions were carried out in silver currency, implied that fluctua-
tionsintherelativepricesofgoldandsilvermodifiedMexico'sexchangerate.
AsmanymonetarysystemsintheworldsufferedasaconsequenceoftheGreat
Depression, the price of gold relative to silver increased. Therefore, the silver
peso depreciated gradually versus the gold peso and the US dollar from
1930.WhengoldreservesintheBancodeMexicobecameexhaustedandanew
monetary law was passed by which gold was demonetized in July 1931, the
exchange rate collapsed and depreciated from 2.67 pesos per US dollar before
the change, to 3.71 by June 1932. It then gradually recovered and was finally
fixedat3.60pesosperUSdollarbytheendof1933.Theimmediateeffectwasa
drastic change in relative prices of importable and domestic goods. Imports
became more expensive and demand was redirected towards domestically
produced goods, in spite of a certain decline in tariff rates.
4
The long border
withtheUSAandthelargeamountofeconomicbordertransactionsprevented
the establishment of exchange controls which had become commonplace in
LatinAmerica.Thechangeinrelativepricesbroughtaboutbythedevaluation
oftheexchangeratewasthemajorforcebehindindustrialrecoveryandgrowth
duringthefirsthalfofthe1930s,whichwaslaterreinforcedbyanewdevalua-
tionoftheexchangerateinMarch1938,
5
andbyexpansivefiscalandmonet-
arypoliciesthattookplaceinthemidstoftheDepression.
6
Lazaro Cardenas' presidency (193440) marked a restructuring of the state
anditsrelationtosocietyandtheeconomy.Thepoliticalbaseshiftedfromthe
army and wealthy groups to the peasants and workers, but left the private
industrial sector untouched, with the exception of oil. A sweeping agrarian
reform, supported by the recently created Banco Nacional de Credito Ejidal a
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
ruraldevelopmentbankthatbenefitedmorethanonemillionfamilies,about
one-third of the population at that time, stimulated the shifting of resources
from commercial agriculture to industry and created a simple but potential
consumermarketthatjoinedvarioussegmentsofsociety(HernandezChavez,
1993,pp.20810).
Therefore, the price effect of the two devaluations acted rapidly and with
strengthtoreorientdemand.Realindustrialproductionincreased118percent
from1932to1940,animpressiveaverageannualrateof10percent.Duringthe
1930s, theindustrial sectorclearlybecametheengine ofeconomicgrowthby
contributing38percentofthetotal,andincreaseditsshareofGDPfrom15per
centin1929to19percentin1940.Whichwerethesourcesofdemandgrowth
during the decade? How were they met by increasing supply? First of all, the
relativepriceofimportsofconsumergoodsrelativetodomesticpricesofsimilar
goods increased, with certain fluctuations, by over 91 per cent from 1929 to
1939.Iftheimpactofadecliningimplicittariffrateandindirectimportprices
aretakenintoaccount,therelativeincreasewassomewhatsmaller,around56
percent,butstillconsiderable(Cardenas,1987,p.122;andtable5.6).
Import-substitutionindustrializationduringthe1930swasimportant,essen-
tiallythroughthepriceeffectcausedbythedevaluations.Ithasbeenestimated
that the relative price elasticity of substitution between imports and domest-
icallyproducedgoodswasquitehigh,rangingfrom1.9to6.4dependingonthe
specificindustry.Inaddition,therewasalsoaneffectonthesupplyside,both
because prices also increased for producers, and because the supply curve in
certainindustriessuchastextiles,cement,ironandrubberproductsshiftedon
averageeachyearbetween8and30percent(ibid.,table5.7,p.125).Finally,the
expansive fiscal and monetary policies that were applied throughout most of
the decade had a continuing impact on aggregate demand, and consequently
onindustrialoutput.Therelativeimpactofimport-substitutiononthegrowth
of aggregate demand in industry was almost 37 per cent during the 1930s,
while domestic demand contributed with 56 per cent and external demand
with 4 per cent (Table 7.1).
7
Naturally, consumer and intermediate goods
industries were the ones where most substitution occurred, while capital
goods had a negative effect. It is clear, then, that import substitution indus-
trialisationwasstrongandstimulatedessentiallybyachangeinrelativeprices
thatreorienteddomesticdemand.
PartlyasaresultofCardenas'policiesagainstcommercialagricultureandthe
oilindustry,resourcesshiftedtotheindustrialsector,henceincreasinginvest-
ment and productivity. First, profit rates rose substantially across industries
during the second half of the 1930s, after some decline due to the Great
Depression. Partly because of the high wage increases, but especially
becauseofthecapitalintensivenessofimportedtechnologyandthestrengthen-
ing of labour unions under the Cardenismo, the process of investment was
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 181
Table7.1 Mexico:sourcesofindustrialgrowth,192939(%)
Domestic External Import Relative
demand demand substitution share
a
Total 56.4 4.3 36.9 100.0
Consumergoods 51.5 5.7 44.7 60.5
non-durable 54.9 5.6 40.5 57.3
durable 88:4 9.6 216.3 3.2
Intermediategoods 55.5 0.2 48.5 24.1
Capitalgoods 36.1 0.0 256:1 15.1
Note: Thesumofdomesticdemand,externaldemandandimportsubstitutionisnot
exactlyequalto100percent.Differencesareduetochangesindemandstructure.
a
Relativeshareofeachgroupofgoodsintotalsupply;averageofyears1929,1934and
1939.
Source: Cardenas(1987),table5.1,p.113.
capital-intensive which increased profit rates. Second, the orientation and
growth of real public investment, 7 per cent a year during the 1930s, created
positive externalities and economies of scale. In particular, the road system
increased from 695 kilometres of roads in 1928 to 9929 kilometres ten years
later.Itconnectedmajorcitiesandcomplementedtherailroadnetworkalready
inexistence.Also,dams,powerplantsandothermajorpartsoftheinfrastruc-
ture were significantly expanded. Nacional Financiera a development bank,
mainlyforindustryandtheBancoNacionaldeComercioExteriortohelpthe
export sector were established in 1934 and 1937 respectively, in order to
supporttheindustrializationeffort.Asawhole,realgrossinvestmentincreased
5percentduringthe1930s,whileprivateinvestmentin1940wasover47per
centaboveitspeakbeforethebrakeoftheDepression.AsapercentageofGDP,
totalinvestmentgrewfromonly5percentin192526,toalmost8percentin
193940(ibid.,pp.1425,159).
With regard to foreign investment, it is also important to mention the
increase during the 1930s in manufacturing, present in certain industries
since the end of the nineteenth century. First, small businessmen continued
comingintoMexico,butinsmallernumbersthanwasthecaseinotherLatin
Americancountries.Second,largefirms,especiallyintheauto-industrysuchas
Ford and General Motors, established plants in Mexico during the 1920s and
1930s. This process would gain much more importance in the 1950s. And,
third, there were joint ventures between domestic and foreign capital that
permitted technology andskill transfers. Naturally, nationalindustrial groups
alsodeveloped,butinanoligopolisticfashionwhichhadbeeninheritedfrom
the Porfiriato regime (Cardenas, 1997, pp. 6592). As a consequence, labour
productivity increased during the 1930s in most industries. Depending on
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
specificindustriesandstatisticalsources,labourproductivitygrewbetween37
and54percentduringthedecade(Cardenas,1987,pp.16777).
Consequently,theprocessofimportsubstitutionduringthe1930swasvery
strongandconstitutedtheleadingengineofgrowth.Itchangedtheroleofthe
industrialsectorandincreaseditsshareofGDPmorethaninanyotherdecade
since then. The process was stimulated by market forces, especially by the
devaluationoftheexchangeratethatmodifiedrelativepricessothatdemand
shifted from the foreign to the domestic sector. Tariff protection had, if any-
thing, a counter effect because the implicit tariff rates and ad valorem rates
applicableatthetimeactuallydeclinedduringthe1930s;however,thevalueof
exportsor,moreimportantly,thecapacitytoimport,wasabletorecoverafter
the crisis partly as a consequence of silver appreciation and partly because
Mexican exports were quite diverse which enabled the country to face the
crisis well. Finally, an expansive economic policy also helped maintain a
high level of demand, coupled with high profit rates and increases in
productivity through positive linkages from public investment projects,
especially of basic infrastructure. By the end of the 1930s, the Cardenas
populist reforms had reached the limit admissible by certain segments of
the population and acceptable to the main political forces. Therefore, the
official candidate for the upcoming presidential election was Manuel Avila
Camacho, a centre moderate candidate, rather than Francisco J. Mu jica, the
most likely candidate to follow the path of Lazaro Cardenas' reforms. In a
sense, the Mexican Revolution was over (Aguilar Cam n and Meyer, 1989,
pp.1825).
TheSecondWorldWarandexport-ledexpansion,194045
Two years after the Second World War broke out a new period of economic
expansion started for Mexico. Economic growth was occurring in a political
system that was becoming increasingly centralized, authoritarian and presi-
dentialist. With the idea that economic prosperity was synonymous with
industrialization, urbanization became widespread among the postcardenista
regimes,andmosteffortswerecentredonexpandingeconomicinfrastructure
forthebenefitofindustry,commerceandthecities.Thegoalofsocialjustice,
whichwasatthecentreduringtheCardenasperiod,wasreplacedbyefficiency
andeconomicgrowth(ibid.,pp.1914).
The war brought with it an expansion of imports from the USA as well as
capitalinflowsthatalmostimmediatelyincreaseddomesticaggregatedemand.
Theflowsoffinancialcapital,fleeingfromriskyareasduetothewar,contrib-
uted to generate a monetary expansion that the Banco de Mexico tried to
sterilize in order to avoid excessive inflation. The central bank reserves
increased by US$22 million while the real money supply grew 20 per cent in
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 183
1940 alone, pushing prices up. Similarly, exports increased more rapidly than
imports,essentially becausetherealexchange ratehadbeenfloating andhad
depreciated15percentsincethedevaluationof1938.Therefore,thebalanceof
payments observed an accumulated surplus of US$62 million in 193940. In
spiteofastrongdeclineinthetermsoftradebetween1939and1941(25per
cent),theexportexpansioncreatedapositiveimpactonnationalincomebut
withhighinflation.
During the following years the terms of trade fully recovered as a con-
sequence of the improvement of export prices, and the volume of exports
increased; by 1945, the level of exports was 48 per cent above that of 1940.
However, in spite of some trade restrictions imposed by the USA military
authorities,MexicanimportsfromtheUSAincreasedevenfasterthanexports.
From1940to1945,importsinUSdollartermsincreasedatanannualaverage
rate of 22 per cent. Of the total, between 70 and 75 percent corresponded to
intermediate products and capital goods (Ortiz Mena, R. et al., 1953, p. 413),
which reflected a larger level of investment and consequently of economic
growth. Indeed, real GDP increased at an average rate of 5.2 per cent during
thewar(193945),butthisgrowthwasgreaterintheindustrialsector:7.6 per
cent a year. Where did that growth come from? How was it financed? What
role,ifany,didbusinessandthegovernmentplay?Whatimpactdidithaveon
incomedistribution?
Untilquiterecently,conventionalwisdomstatedthattheprocessofimport
substitutionstartedduringthewarduetothedeclineininternationaltradeasa
consequenceoftherelativeisolationthattheLatinAmericaneconomieshadto
face. Although that assertion is in principle correct for most countries, the
Mexican case is different, andthe reason is geographic. TheUSA andMexico,
beingneighbours,didnothavetraderestrictionsduringthewar,exceptthose
imposedbythemilitaryauthorities.Inanycase,Mexicodidhelpthewareffort
by providing labour and by keeping its borders open to trade according to a
trade agreement signed in 1943. Consequently, the expansion of US demand
forMexicancommoditiestendedtoincreaseduringthewar,whichstandsout
as a major difference with regard to other Latin American countries which
facedalmostcompleteisolationduringthewar.
8
Exports changed their composition dramatically. Mineral exports, which
historically had been responsible for more than 70 per cent of total exports,
declinedsteadilyinthefirsthalfofthe1940storeachonlyalittleover20per
cent.Agriculturalproductsdidincreasetheirsharesomewhat,butthegreatest
impacttookplaceinthemanufacturingsector.Fromaround7percentin1939,
manufacturing exports increased without interruption until 1945, reaching
almost 38 per cent of the total. That implied an 850 per cent increase in the
volumeofmanufacturingexportsduringthewar(ibid.,tables130,131,andpp.
309,401).Infact,suchahighshareofmanufacturingexportswasnotreached
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
againinMexicountilthe1980s.Sothewarwasindeedaverypeculiarperiodin
thatregard.
Therefore, the growth in the manufacturing sector was based on external
demand and the full use of capacity, especially cotton textiles which counted
up to 60 per cent of total manufacturing exports, in spite of the increasingly
overvaluedexchangerate.Indeed,thefixedexchangerateregime,andthefact
thatdomesticinflationwashigherthanthatintheUSA,graduallyovervalued
the peso which appreciated 32 percent between 1940 and 1945. That in turn
inhibited exports by making them less competitive. However, the war itself
eliminated other competitors and the demand was very large, so industrial
outputexperiencedaboomanyway.
AsTable 7.2shows, external demandcomprised 78.9 percent oftotal man-
ufacturing growth during the war, while internal demand comprised of 29.6
per cent. Import substitution was reversed, essentially because of the over-
valuation of the exchange rate. The price effect was again important in the
substitution process, but in the reverse direction to that experienced in the
1930s.Thetextileindustrywasmostimpactedbytheexportboom,alongwith
food, beverages and tobacco, which in turn affected other sectors which were
reallynon-tradable,suchasconstruction.Actually,constructionwaspositively
affectedbytheheavyprogrammeofpublicinvestmentininfrastructurewhich
took place in those years, which had a far-reaching impact on the economy.
Just as had happened in the previous 15 years, the programme of public
investmenthadbeentargetedtoinfrastructure,especiallyroadsandirrigation
projects. The increasing volume of foreign trade during the war increased the
level of real fiscal revenues by 7 per cent yearly on average, which in turn
allowedforanexpansionofpublicexpenditureofasimilarmagnitude.Infact,
publicspendingincreasedatapproximatelythesamerateasGDP,whichmeant
much more public investment supported by Nacional Financiera and other
agencies. In fact, public investment increased 15 per cent a year in real terms
duringthewar.Thisprogramme,whichalsoincludedsomestatemanufactur-
ing firms considered strategic,
9
along with the rapid growth of private invest-
ment, established a production base that permitted output expansion. Total
realinvestmentduringthewargrewatanannualaveragerateof13percent.As
the capitallabour ratio increased, output per worker in the industrial sector
also increased, especially because of the productivity gains obtained by the
expansion of economic infrastructure. Also, a law to stimulate industry
through fiscal exemptions was passed in 1941, and reinforcedin 1946, which
providedfiscalincentivesandothergovernmentalpublicsupport.
10
However, income distribution worsened during the 1940s. The full use of
productive capacity as well as the increase in prices tended to redistribute
income from labour to capital, although there was a slight overall
improvementinthestandard ofliving.Theshiftingofoccupationalactivities
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 185
Table7.2 Mexico:sourcesofeconomicgrowthofindustry(%)
Domestic External Changesin Import
demand demand demand substitution
structure
194045
Allindustries 29.6 78.9 0:1 8:5
Textiles 78.6 173.4 2.0 3.1
Food,beveragesandtobacco 86.3 19.0 1.0 4.3
Construction 102.7 0.1 0.4 2.2
Otherindustries 15.1 118.0 0.4 3.4
194550
Allindustries 130.2 54.0 1.8 25.5
Textiles 345.1 246.8 0.3 1.4
Food,beverages,andtobacco 90.4 0.2 0.4 10.2
Construction 75.1 5.7 0.6 18.6
Otherindustries 192.0 94.3 2.2 4.4
Source: Cardenas(1994,tableIV.8,p.126).
fromlow-wageagriculturetogrowingindustryandservicesectorswithhigher
productivity,whereunionnegotiationsbecamemorecommon,alsoimplieda
redistributionoflabourincomebetweenthesetwosectors(Hansen,1971,pp.
723). Over the years, this pattern would become a trend with long-term
consequencesforMexico'sdevelopment.
Therefore, industrial expansion during the war was caused by the pull of
foreign demand, especially from the USA, and on the supply side by the
availability of capital and intermediate goods imports from that country
which permittedmore investmentandfull utilization of capital,as wellasan
increase in investment in infrastructure. This was in turn facilitated by the
increase in public expenditure financed by the additional fiscal revenues
obtained through the expansion of international trade. This contrasts with
the conventional view which states that the war stimulated the process of
importsubstitution,aseachcountryhadtorelyonitsownproductivecapacity.
Mexico's long borderline with the USA ruled out such a situation; Mexico's
industrialexpansionduringthewarwasexport-driven.
Protection-ledindustrialization,194662
TheendofthewarunderlinedthefactthatMexicosquarelyfellwithintheUS
sphereofinfluence,farfromaweakenedEuropethatcouldhavesomecounter-
vailing weight. The confrontations of the days of the oil expropriation just a
few years earlier were left aside, and both countries became allies during the
war. An agreement of monetary stabilization for US$50 million was accorded
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
betweentheUSDepartmentoftheTreasuryandtheBancodeMexicoin1941
tostrengthenthepesodollarparity(NacionalFinanciera,ElMercadodeValores,
1955, p. 587), while a trade agreement kept Mexico's borders open, and Mex-
icanmigrantlabourersworkedintheUSA.Duringthewar,theinflowofcapital
as well as foreign remittances originating from the so-called `Bracero Pro-
gramme'
11
generated a large inflow of dollars that impacted the monetary
base and consequently the supply of money, thus contributing to the rise in
prices.Inturn,andgivenafixedexchangerate,thepesobecameincreasingly
overvalued. When exports reduced their rate of growth due to the end of the
war,importsincreasedevenmorerapidlyandby1947were93percenthigher
than in 1945, and 444 per cent more than in 1940. Imports kept increasing
becausetherewerenolongertraderestrictionsofanysortandbecausethelevel
ofnationalincome,giventhefactthattheexchangeratebecameovervaluedby
approximately 40 per cent, continued to increase in US dollar terms quite
rapidly (Cardenas, 1994, p. 100). In 1946 and 1947, average GDP valued in
US dollars increased 23percent each year, whileGDP inreal pesosonly grew
3.4 per cent. Such a trend was possible only as long as there were sufficient
internationalreservesinthecentralbanktopayforimports.
The large trade deficits that emerged after the war, which reached US$578
million in just two years, were the major cause of the decline of reserves in
194647.
12
On the political front, a new government had taken office in
December1946, headedby MiguelAleman,whowouldbecome thePresident
mostcloselyidentifiedwiththeindustrializationeffort.
Facing enormous pressures on the trade account, the Mexican government
decreed a set of protective measures to contain the increase in imports.
Although the government had tried to establish higher nominal tariff rates
during the war, Mexico's trade agreement with the USA from the war years
prevented it. After some efforts to impose some restrictions,
13
the Mexican
governmentfinallyapprovedalonglistofarticlessubjecttoimportrestrictions
thatcameintoeffectuntil1947.
14
Thiswasthebeginningofatrendtoprotect
domesticindustry,notsomuchasadevelopmentstrategybasedonprotection-
ismwhichwastobethecaselateron,butasashort-termmeasuretocontain
the avalanche of imports. However, by 1948 the situation became unbearable
asthelevelofreserveshadfallen67percentwithrespectto1945.Thefinancial
authoritiesallowedthepesotofloatfreelyinJulyof1948,anditimmediately
depreciated.WhentheexchangeratewasfixedagainalmostayearlaterinJune
1949, the peso had depreciated almost 80 per cent. As a consequence, the
relative price of imports substantially increased compared with domestic
goods, which redirected demand. Therefore, the impact on the sources of
industrial demand were immediately felt, as import substitution became
important again both for the devaluation as well as for the beginnings of
protection(Table7.2).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 187
Theefforttoprotectdomesticindustrywasatfirstmodest.Onlyagroupof
`luxury'goodswasincludedatfirst,whichamountedto18percentofimports
in 1946. A few months later, in November 1947, the system of specific tariff
rates was changed for ad valorem rates to avoid the impact of inflation on
protectionlevels.Ayearlater,theprotectionschemewasreinforcedwhenthe
Ministry of National Economy was allowed to change the list of goods under
restriction, with the only requirement being to consult with the Ministry of
Hacienda to check its impact on fiscal revenues (Torres, 1984, pp. 2489).
Throughthe1950s,thepolicytoprotectdomesticindustrycontinued,gradu-
allyenlargingitsscopeandeventuallyapplyingtheconceptthatanewindus-
trywhichsubstitutedimportsautomaticallyobtainedfullprotection.Thatisto
say, as soon as a domestic company, of either national or foreign capital,
producedcertaingoodswhichuntilthenwerebeingimported,thegovernment
wouldestablishimportquotastoprotectit.Asthe1950sadvanced,theideaof
protectingdomesticindustrybecameawidespreadstrategyofeconomicdevel-
opment. The main idea was to stimulate domestic producers by isolating the
national market, and at the same time to substitute imports. It was generally
thought that the economy's dependence on the availability of foreign
exchange to import capital and intermediate goods made the national eco-
nomy more vulnerable to external shocks, a situation worsened by the belief
that the terms of trade for the developing countries would decline over time.
Also, by stimulating domestic firms a stronger domestic market would be
created, backward and forward linkages with other firms would occur, wages
wouldincreaseandthedependenceonforeignexchangewouldbediminished.
The level of import substitution from year to year also had to do with the
level of the real exchange rate. In particular, when the exchange rate was
overvalued the import coefficient would increase. Before the devaluations of
1938 and194849, the import-substitution coefficient tended toincrease as a
result of the overvaluation of the peso and the actual cheapening of imports.
Afterthedevaluations,theimportcoefficientdeclinedasforeigngoodsbecame
moreexpensive(Figure7.1)(Cardenas,1994,tablesIV.9andV.4).Nevertheless,
alreadybytheearly1950s,therewereseveralindustrialsectorswhosesupplies
came mostly from domestic producers, but with imported inputs. Indeed, for
the textile, food, beverage, tobacco, shoes, soap, rubber, alcohol and glass
sectors among others, almost 95 per cent of the domestic market was already
supplied by domestic firms. Consequently, the growth of industrial demand
duringthe1950scameessentiallyfromthedomesticmarket,highlyprotected
fromforeigncompetition.Importsubstitutiondidnotplayanysignificantrole
(Table 7.3). In other sectors where domestic producers entered the market
undernewtradeprotection,someimportsubstitutionoccurred.
15
Once this process was well-underway, another important phenomenon
appeared:anextensiveflow offoreigntransnational corporations. Itis logical
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
i
n
d
e
x

120.0
100.0
80.0
60.0
40.0
20.0
0.0
Overvaluation
Import coefficient
1925 1930 1935 1940 1945 1950 1955
Figure7.1 Mexico:exchangerateovervaluationandtheimportcoefficient
Source: Cardenas(1994).
Table7.3 Mexico:sourcesofindustrialdemandgrowth(%)
Domesticdemand Externaldemand
195054 195458 195058 195054 195458 195058
Allindustries 99.5
Textiles 156.0
Food,beverages,and 93.2
tobacco
Construction 102.4
Otherindustries 94.8
84.4 92.5 0.6 5.5 2.9
99.2 129.6 57.1 9.8 34.7
73.6 86.6 8.5 27.3 16.6
97.6 99.9 0.4 1.8 0.8
93.6 95.0 0.1 0.6 0.4
Importsubstitution Changesinstructure
195054 195458 195058 195054 195458 195058
Allindustries 0.4
Textiles 1.7
Food,beveragesand 1.5
tobacco
Construction 1.3
Otherindustries 3.3
9.0 1.7 1.5 1.1 2.9
7.0 2.4 2.8 3.6 2.7
1.3 0.1 0.2 0.4 1.1
1.2 0.4 0.7 0.6 0.3
5.3 1.5 1.8 0.5 3.0
Source: Cardenas(1994,tableV.3,p.142).
that corporations thatused toexport to Mexico under free trade in the1930s
and1940swouldmoveinandestablishaplantinthecountrywhentheyfound
that by transferring to Mexico they would receive full trade protection and a
captivemarket.Thattrendbecameincreasinglyimportantfromthelate1940s
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 189
until the Mexicanization laws in the 1960s, when another step in the protec-
tion process took place: that of shielding domestic businessmen, not only
domesticmarkets.
The entry of foreign companies naturally increased private as well as do-
mesticinvestment.Indeed,the1950switnessedamajorboomininvestment,
mostof whichwasdone bythe privatesectorinmanufacturing, bankingand
other services. During the whole decade, real total investment grew at an
averagerateof17percent,whileGDPdidsoatarateof6.6percent.Therefore,
investmentasapercentageofGDPincreasedfrom14percentin1950to17per
cent in 1960. Private investment reacted quite well to the flow of public
investment that had been occurring for many years, crowding in private
resources. Investment from the private sector increased at a real annual rate
of 15 per cent during the decade, while public investment, essentially in
energy, iron, roads, education, health and urban public works, grew at a real
rateofover5percent(ibid.,tablesV.2,A.18andA.29).
Theimpressivegrowthinproductivecapacityprovidedtheroomforgrowth.
Overall productivity during the 1950s increased at a rate of approximately
4percentper yearinspiteoftherapidincreaseinpopulationandthelabour
force,
16
andthatinturnwasreflectedinthelevelofrealwageswhichincreased
2.2percentonaverageperyear.
17
Inessence,theextremelyrapidgrowthofthe
economy, 6.2 per cent on average (195062), and of the industrial sector in
particular, 7.3 per cent, was the result of the rapid increase in private
investment stimulated by the strategy to protect domestic producers and
provide a captive market. In order to do so it was necessary to keep the cost
of living at a low level, supply financial capital at reasonable rates at least for
working capital, and generate sufficient foreign exchange to finance the
necessaryimportsofcapitalandintermediategoods.Thetransferofresources
fromagriculturetoindustryandservicestookplacethroughthepricesystem.
The state-owned company Compana Exportadora e Importadora de Ma z
CEIMSA, later CONASUPO, or Compana Nacional de Subsistencias Populares
established to purchase and distribute staple commodities became a monop-
sonist. Thus minimum `guaranteed' prices actually became maximum prices
forthebenefitofurbanconsumersandattheexpenseofagriculturalproducers
(Izquierdo, 1994, table IV.7). Urban labour unions loyal to the government
wereabletoobtainfurtherconcessionsandprotecttheirowninterests,along
with business, at the expense of the peasants and food producers. The CTM
Confederacion de Trabajadores Mexicanos had become by far the most
important intermediary between government and workers, with Lombardo's
successor, Fidel Velazquez, as its perennial leader.
18
On the other hand,
independent labour unions were repressed when they became a threat to
the status quo, as was the case of the railroad labour movement in the late
1950s.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Itis estimated thatduring the1950s andearly 1960s,approximately 40per
cent of private investment was composed of imported equipment and ma-
chinery (Nacional Financiera, 1978, p. 47). Thus, it was important to have
sufficient foreign exchange which, during this period, was supplied by the
export of cotton as well as other agricultural commodities, and by services
such as tourism and border transactions, which actually helped finance the
trade deficits (ibid., pp. 37780). Similarly, the resources channelled through
the financial system trebled in nominal terms, and most of the increase went
to the private sector. In fact, while during the 1940s 43 per cent of the
additional resources of the financial system were devoted to the private
business sector and other private individuals, by the end of the 1950s that
percentagehadincreasedto70percent(BrothersandSol s,1967,p.145).
Ontopofthefinancingonpreferentialtermsthatdevelopmentinstitutions
such as Nacional Financiera granted to the most important players from the
privatesector,anadditionalperkwasthatthegovernmentactedvirtuallyasa
last-resortbuyerforanimportantenterpriseiftheventurewentsour.Typically,
Nacional Financiera invested resources in a company as a minority share-
holder; if the business was a success, then it reduced its participation to a
symbolic amount. However, if bankruptcy loomed, the other shareholders
`requested' that it the government buy from them and take over. The
reasons why the government accepted were basically twofold: to save jobs,
and because the company was considered important for economic develop-
19
ment. Thispolicymarksthestartingpointofwhatwouldbecomeagigantic
andheterogeneousgroupofstate-ownedenterprises.
Finally,theeconomywasalsopositivelyimpactedduringthe1950sandearly
1960sbytheinternalizationofpositiveexternalitiescreatedbytheinvestment
in infrastructure referred to above. It is calculated that, during that period,
almost 20 per cent of GDP growth was caused by externalities and other
organizational changes (Cardenas, 1978, tables 3, 13 and 14). This figure is
relativelyhighandshowsthattherewerestillampleopportunitiestoincrease
overallproductivitybyinvestinginstrategicprojectswithhighsocialproduct-
ivity,likeroadsandenergyprojects.
Therefore, from the late 1940s to the early 1960s, the industrialization
process was stimulated by the strategy of protecting the domestic market
which spurred private investment, by the crowding-in effects of public
investment in projects of high social return, by a low cost of living sustained
by subsidies and transfers from rural to urban producers through the price
of foodstuffs, and by the availability of financial resources at reasonable
rates which were essentially used for working capital. The development
model was a success, but it seems that by the early 1960s the model began to
show signs of deterioration which today, with the benefit of hindsight, are
muchclearer.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 191
Furtherprotection-ledindustrialization,196381
20
The1960switnessedoneofthemostaffluentperiodsofcontemporaryMexico.
Real GDP increased at an average rate of7.1 percent, while output percapita
grew by 3.6 per cent. There was a deepening in the urbanization process,
especially in the larger cities, so that by the end of the decade almost half of
thepopulationlivedintownsandcities.
21
Thisprocessunfortunatelyreflected
thedeclineinbasicsectors,suchasagricultureandmining,thathadtradition-
allybeenimportantinMexico'sdevelopment.Therewasarelativestagnation
ofagriculture,whichdespitegrowingatarateof3.2percentayearstayedthe
same in per capita and terms and lost almost 25 per cent of its share in GDP.
On the other hand, the modern sector showed impressive rates of growth,
including manufacturing (9.3 per cent), power infrastructure (14.2 per cent),
construction (8.3 per cent) and government services (8.5 per cent).
22
As a
whole,theindustrialsectorcontributedalmost40percentoftotalgrowth.
At the same time, inflation was comparable to that observed in developed
countries less than 3 percent on annual average so theexchange ratewas
keptfixedduringthoseyears.Ontheotherhand,inflationwaskeptatsuchlow
levelsthankstothefactthatseveralkeypricesofpublicsectorgoodswerekept
fixed for long periods of time. While that implied an increasing subsidy for
productionandconsumption,thecostfortheexchequerincreasedalongwith
it(Figure7.2).
30
50
70
90
110
130
150
170
190
210
230
250
270
290
Gas for Industry
Electricity for Industry
Gasoline and diesel
Diesel
Gasoline
1
9
6
0
1
9
6
1
1
9
6
2
1
9
6
3
1
9
6
4
1
9
6
5
1
9
6
6
1
9
6
7
1
9
6
8
1
9
6
9
1
9
7
0
1
9
7
1
1
9
7
2
1
9
7
3
1
9
7
4
1
9
7
5
1
9
7
6
1
9
7
7
1
9
7
8
1
9
7
9
1
9
8
0
1
9
8
1
1
9
8
2

Figure7.2 Mexico:evolutionofkeyrealpublicpricesforindustry(1960100)
Source: BasedonOrtizDietz(1985),anddeflatedwiththeGDPdeflator.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
The supply of capital services, that is the flow of new net investment,
increased at a rate of 9 per cent in 196371 so that the capitallabour ratio
grew 7 per cent annually and therefore productivity by 5 per cent, most of
which was transferred to wages which increased 4 per cent in real terms
(Cardenas, 1978, tables II, XI and XII; Bortz, 1988, p. 266). Similarly to the
experience of the 1950s, GDP growth had relied on enormous capital forma-
tion,wherepublicinvestmenthadattractedmoreprivateinvestmentbypush-
ingprofitratesupandbyextendingtheprotectionfromexternalcompetitors.
Gross capital accumulation increased 10.3 per cent on average from 1963 to
1970, with a share of GDP that reached 18.5 per cent, and then declined in
1971.
23
That process implied that a wide variety of firms, with differing com-
petitiveedges,wereabletofunctionandgrow.Naturally,thosemorecompet-
itive firms, usually with ahighercapitallabour ratio and technology of more
recentvintageenjoyedalargerprofitmarginthantheirfellowcompetitorswho
couldonlysurvivethankstothehighlevelofprotection.
24
The business community was obviously in favour of the idea of further
protecting the domestic market from foreign competition. From the mid-
1950s, but mainly from the beginning of the 1960s, the private sector repre-
sentedbytheindustrialchambersCanacintraandConcaminCamaraNacional
de la Industria de la Transformacion and Confederacion de Camaras Industriales
exertedmorepressureonthegovernmenttoreceiveadditionalprotection.The
governmenttookanumberofmeasures,mainlyto`Mexicanize'theeconomy
by limiting new investment, such as entry barriers to foreign firms in certain
industries already producing in Mexico, a maximum of 49 per cent foreign
control of shares and thus a need to invest along with Mexican businessmen,
preference for foreign financing rather than foreign investment, and so on
(Izquierdo,1994,pp.10919).
25
Consequently,theprivatesectorwasreceiving
an additional level of protection: quantitative import restrictions and strong
limitations for foreign investors. No wonder the flows of foreign investment
declinedsteadilyduringthe1960s,whenitrepresentedonly5percentoftotal
capital accumulation and 7.5 per cent of private investment.
26
By the same
token,thevolumeofinternationaltradealsodeclinedduringthe1960s,from
25percentofGDPto20percent,butreflectingalargertradedeficitasimports
increasedmorethanexports(NacionalFinanciera,1978,p.46).
Thepolicytodeepenthestrategyofimportsubstitutionprovedfar-reaching,
withlong-termconsequencesfortheeconomyandforsocietyasawhole.Itis
naturalthattheauthorities,withtheenthusiasticsupportoftheprivatesector,
decidedtocontinuethesamepathasbefore,asithadproventobesuccessful
whenmeasuredbyalmostanyyardstick:therehadbeennomajorproblemswith
theexchangerateandinflation;output,wagesandemploymenthadincreased
at substantial rates; and domestic investment continued to grow. However,
this successful story was hiding certain weaknesses in the development
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 193
strategythatdidnotsurfaceclearly.Infact,incomedistributionhadcontinued
to be extremely unjust, or even worsened, and the economy began to use its
capacity to borrowin theinternational capital markets in1963, atreasonable
interestratesandwithvirtuallyacleanslateasaforeigndebtor.
The first sign of weakness was the loss of competitiveness of the industrial
sector. The strategy implied the strengthening of an oligopolistic industrial
structure, as internal competition was curtailed, and by 1960 a large share of
consumergoodsindustrieswaspracticallyself-sufficientinthesensethatthere
were virtually no imports in those industries except for inputs of certain raw
materialsandcapitalgoods.Forinstance,textiles,food,beverageandtobacco
hadanimportcoefficientof1.8percent,shoes4percent,soap0.01percent
and rubber 1 per cent. Intermediate goods had also diminished its coefficient
significantly,suchasironandsteel18percent,cement0.1percent,paper40
percent,fertilizers69percentandtheauto-industry41percent.Forindustry
asawhole,theimportcoefficientwasonly9.5percent(Cardenas,1994,table
V.4). Therefore, a number of consumer-goods industries, at the time called
ìnfant' industries, had already had by 1960 a decade of protection and fiscal
stimulus that had allowed them to grow and develop. In the case of some
intermediate and capital goods, the protection had not really favoured them
because ithad beenconsidered necessary to implement the process ofimport
substitutioninconsumergoods.
Continuingtoprotectconsumergoodsindustrieshadanumberofimplica-
tions. First, the protection from foreign competition through quantitative
restrictions and non-tariff barriers, with subsidized agricultural products,
energy,transportationandotherbasicinputs,andwithfavourablefiscaltreat-
ment and public investment with positive externalities, led to a situation in
which industry in general was quite protected and therefore was not interna-
tionally competitive. Their prices and quality were not always sufficient to
compete abroad, as evidenced by the failure of various efforts to increase
manufactured exports; although they did increase as a percentage of total
exports, they were still low. In principle, the level of protection gave way to
uncompetitiveness, but in the early 1970s there were some signs that some
internal manufactured goods prices were declining compared to those in the
USA, thus implying that they were not as high as one would have expected.
One possible reason for this was that both internal competition and the long
and permeable border with the USA permitted certain smuggling activity
which somewhat counteracted the protection policy with regard to certain
goods(ShveidFinkelstein,1972,tables7.7to7.9andtheconcludingchapter).
Nevertheless,manybusinesseswereinefficient,withexcessiveprofitratesanda
high cost to public finances in the form of subsidies at the expense of con-
sumers and tax-payers. As Leopoldo Sol s (1981b, p. 204) put it `protection to
the infant industry...had no sense for the export market that would make it
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
internationally competitive and often it only protected inefficiencies'; or as
GerardoBuenowrotereferringtotheprotectionpolicy,that
on the one hand, industries that have been created under this instrument
[protection],theyonlysurviveinuneconomicconditionsortakeadvantage
oftheirmonopolisticposition;ontheotherhand...thelevelofprotection
has gone beyond the necessary limits to help firms at their initial stages.
Thus,inefficienciesandlackofahealthycompetitionhavepersisted.
(Bueno,1970,pp.2223)
Thishadgraveconsequencesinthemedium-term,becausegrowthcouldonly
comefromthedomesticmarketwhicheventuallywouldbelimited,especially
duetolowwagesandtheskewedincomedistributioninheritedfromprevious
decades (Hansen, 1971, pp. 747). And naturally, because even if final goods
wereproducedinternally,theystillrequiredsomeinputsandalmostallcapital
goods from abroad. Therefore, the goal of reducing imports or at least the
importcoefficientwasnotasattainableasitlookedatfirst.
27
Again with the benefit of hindsight, it is clear that another relevant con-
sequence of expanding the protectionist strategy to other industries was that
such a process was more difficult because the market structure of such indus-
trieswasmoreoligopolistic,andbecausetheyrequiredmuchmorecapitalper
job created and a larger market to produce competitively (Boltvinik and Her-
nandez Laos, 1981, pp. 494 and 503). Although the size of the market might
not be a restriction to establish one particular plant, the protection needed
wouldhavetobeforalongerperiodoftimeandagreaterdegreethanhadbeen
the case of the consumer-goods industries. By the same token, these new
industries would have to be export-oriented from the beginning to be able to
continuegrowingoncethedomesticmarketwasfullysaturated.Naturally,the
costofthisprojectwouldbesubstantialandwouldhavetobeabsorbedbythe
economy as a whole via fiscal deficits, external debt and less job-creation per
peso spent. Given the level of self-sufficiency of a large part of the industrial
market, and the high cost of continuing the import-substitution strategy of
development,itseemsclearthatitwaspreciselyatthebeginningofthe1960s
that such a strategy should have been reviewed and trade policy changed to
begingraduallyliberalizingtheprotectedindustriestoforcethemtocompete.
But the mood was not ready for that change. After the GDP growth dimin-
ishedatthebeginningofthe1960s,presumablydueinparttoMexico'spolicies
towards Cuba, the greatest worry was how to resume growth. Mexico rein-
forced its protectionist policy, a decision shared by both the government and
the private sector, including the labour unions. Neither of these particular
groups would benefit in the short run if the economy was liberalized, and
they were convinced with reason that the strategy had so far worked so well
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 195
that there was no cause for them to change. So the process of protecting
businessmen and markets from foreigners and from imports continued. The
strategy was expanded to include intermediate and capital goods, and when
privatecapitalwasnotavailableortheprivatesectordidnotwanttoparticip-
ate in certain industries the government restarted an old scheme to insure
domestic production of most goods: the creation of a growing number of
stateenterprises,ofteninjointventureswithprivatecapital,tocover`strategic'
sectorswhereprivateprofitswererelativelylowandwhichmaturedmostlyin
the very long run. Public enterprises were created in electric power, steel,
fertilizers, petrochemicals, auto-industry, among others, mostly financed by
externalpublicdebt(NacionalFinanciera,1973,pp.21216).
Asanticipated,theimportsofmanufacturedgoodscontinuedtogrowmore
rapidly than exports, and consequently the trade deficit continued to widen,
from US$457 million in 1960 to US$1038 million in 1970. The accumulated
deficitinthedecadeamountedtoUS$4billion.Thisfactwasaggravatedbythe
agriculturalsectorwhichbythemiddleofthe1960shadlostitsroleasaforeign
exchangeproducer,asitstagnatedasaresultoflessinvestmentinagriculture,
both public and private. In fact, the share of public investment in agriculture
declinedgraduallyfromthelate1940suntiltheendofthe1960sattheexpense
of investment to stimulate industry, education, health and urbanization pro-
jects. Also, the policy of low guaranteed prices for agricultural producers in
order to maintain low wages in the cities reduced profit rates in agriculture,
thusdiscouraginginvestment.
28
Finally,themajorconsequenceofcontinuingthestrategywastheincreasing
relianceonexternalborrowing.Fromthelate1950s,theeconomywasunable
to save enough to finance its investment, although the gap was modest com-
paredtowhathappened duringthe 1970s(Table7.4). However,itis truethat
during the 1960s, the country had to obtain external financing in order to
invest the amount required. The private sector was able to save more than it
was investing and the government had to borrow to finance its deficits. By
1970, the public sector was running a deficit of approximately 3.8 per cent,
while the economy as a whole needed around 1.3 per cent of GDP of net
external borrowing. This fact is usually overlooked, essentially because the
standingdebtduringthe1960swasrelativelysmall,butincreasedatanannual
average rate of 15.9 per cent and almost doubled in terms of GDP during the
decade.
29
Therefore,therewasnorealprobleminservicingthedebt,although
thosepaymentsreached23.4percentofexportsin1970,or2.4percentofGDP
(Cardenas, 1996, p. 85). In any case, the problem was that by following the
same development strategy, which implied that imports would continue to
increase and that there would be significant fiscal costs, it was not clear
where the resources would come from to finance growth in the long term. It
seemedasiftheauthoritiesandtheothermaineconomicactorsthoughtthat
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table7.4 Mexico:sourcesoffinancingoftheeconomy,195070(GDP
percentages)
195056 19576 196 2 370
Totalinvestment 15.2 15.5 18.3
externalsavings 0.8 1.8 2.1
domesticsavings 14.4 13.7 16.2
Privateinvestment 9.7 9.9 11.5
privatesavings 9.0 8.9 12.1
balanceofprivatesector 0.7 1.1 0.6
Publicinvestment 5.4 5.5 6.9
publicsavings 5.4 4.8 4.1
balanceofpublicsector 0.0 0.7 2.8
Source: Cardenas,1996,tablesI.2,A.2andA.4.
the capacity to borrow was unlimited, but of course it was not. Insufficient
economicgrowth,althoughstrongbyinternationalstandards,impliedagrow-
inglabourmarketimbalancewheresupplyexceededdemand,andsocialfrus-
trationendedup,asinvariousothercountries,instudentmovementswhichin
Mexico'scasewerebloodilyrepressedin1968.
Themodeldidnotimmediatelycollapse,becauseitwassustainedbyexternal
financingandbytheoilboomduringthesecondhalfofthe1970s.Bytheend
of the 1960s it was clear that major economic liberalization was needed, but
nobodywantedtocarryitout.Neithertheprivatesectornorthegovernment
wanted to change the status quo. The first, simply because that would imply
more competition and the loss of economic advantage and high profit rates;
the second, because it would lose discretionary power to drive the economy
and therefore also lose political control over the private sector. But social
demands, highly exposed by the recent student movement, along with the
steady increase in thelabour force,urgently called for high ratesof economic
growth.
ThecollapseofBrettonWoods,alongwiththeoilembargoin1973,created
unstable financial conditions on a global scale that affected Mexico to some
extent. Coupled with those conditions, the need to continue growing, espe-
cially after a slow year in 1971 which made the government very nervous,
became as urgent as in the early 1960s (Sol s, 1981a, pp. 478). The new
Echeverr a government engaged in an ambitious programme called desarrollo
compartido sharing development in contrast to the so-called desarrollo est-
abilizador stabilizing development of the previous decade. The govern-
ment's response entailed much greater public spending in subsidies, more
state enterprises and infrastructure largely financed by public indebtedness
internalandexternalandmoneyprintingwhichchangedtheroleofthestate
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 197
evenmoreandeventuallyledtoconfrontationwiththeprivatesector(Rubio,
1989,pp.98100).
The government's goals were to stimulate economic growth while simulta-
neously improving income distribution. But this additional objective was not
supportedwithafundamentalinstrument,namelyfiscalreform,thatactually
was intended but aborted in December 1972 because of opposition from the
private sector's elite (Buffie, 1990, pp. 41819). As the spending build-up was
not matched by a similar increase in revenues, inflation erupted after more
than15yearsofpricestability.Theensuingovervaluationoftheexchangerate,
only sustained by external borrowing and to a much lesser extent foreign
investment,worsenedthecurrentaccountdeficit.
The strategy worked in terms of growth, since real GDP increased at an
average annual rate of 6.7 percent during the 197181 period. Therefore, the
additional dollars made it possible to finance the same development strategy
which in fact needed larger amounts of foreign exchange to maintain the
balance of payments in relative equilibrium. In fact, the stock of the external
publicdebtincreasedatanannualrateof25.7percentinUSdollarterms.
Butadditionalforeignborrowingdidnotpreventadevaluationofthepesoin
August1976.Afterashort-livedbutsignificantfinancialcrisis,
30
whichneeded
helpfromtheIMF,therecentoildiscoverieschangedthepicture.Thescarcity
offoreignexchangesuddenlyvanishedasoilprovenreservesskyrocketedfrom
6.3 billion barrels in November 1976 to 16 billion at the end of 1977 to 40
billion barrels a year later (Szekely, 1983, p. 95). That potential wealth was
mirrored by the unlimited availability of resources from the eurodollar mar-
kets,veryliquidasanintermediaryintheso-calledrecyclingprocessofpetro-
dollars.
Theexpectationwasthatanòilsurplus'wouldmaterializeafterafewyears,
fueled by a high platform of crude exports. One of the first objectives that
PresidentLo pezPortillohadforthatsurpluswastofinancea`NationalFundfor
Employment', with jobs as a medium to reduce inequality and poverty. How-
ever, the expectations on, and uses envisioned for, that surplus would soon
explode,andinthemosteuphoricmomentsoilwasseenasthefinanciallever
to attain higher stages of development. Public spending continued with
renewedimpetus,subsidizingconsumptionoffoodstuffs,energyandespecially
the agricultural sector.
31
The purest crystallization of these wild expectations
would arguably be the National Plan for Industrial Development PNDI
presented in 1979,
32
whose objective was for the Mexican economy to
reach an average GDP real growth rate of 10 per cent in the 198090 period.
The PNDI merely promoted the heavy state intervention that was taking
place already. The role of the state was quite clear in this process, and was
to be even more grandiose than the oneplayed during theEcheverr aadmin-
istration.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Someoftheleadingbranchesofindustrywerealreadyinthehandsofstate
enterprises, like oil and electricity and to a lesser extent steel. Consequently,
public investment as a percentage of GDP, including other activities like con-
struction, increased from 5 per cent to 12 per cent, while private investment
remained practically thesame ataround13 percent of GDP(Nacional Finan-
ciera,1984,p.55).Thedaysofacrowding-ineffectofpublicinvestmentwere
long gone and there was actually some crowding-out, as the government
entered new economic activities. Naturally, industry experienced an acceler-
atedpathduringthe1970s,reachinganamazingannualaveragerateofgrowth
ofalmost11percentinrealterms.Theprocessofacceleratedindustrialization
wasextendedtomoresectors,withgovernmentsupportandincreasinglywith
stateenterprises,whichdidnotalwaysoperateefficiently.Morediversified,the
industrialsectoradvancedintotheproductionofawidersetofgoods,includ-
ingequipment,machineryandrawmaterials.Therelativeproductionofcon-
sumer goods diminished at the expense of intermediate and capital goods.
While in 1960 consumer goods represented 58 per cent of manufacturing
production,by1970itdiminishedto50percentandby1981itreachedonly
42 per cent. By contrast, intermediate goods' participation increased steadily
from 28 per cent in 1960, to 32 per cent in 1970 and 36 per cent by 1981.
Similarly,capitalgoodsincreasedtheresharefrom13percentin1960to18per
centin1970andfinallyto22percentin1981(Figure7.3)(Villarreal,1988,pp.
3723). Table 7.5 shows the relative contribution of the ten most important
branches of industry in those three years. The level of protection, now
extended to all sorts of goods, constituted a heavier burden to domestic pro-
ducers,sincethelevelofeffectiveprotectiondeclinedwhenintermediateand
capitalgoodsreceivedsometariffprotection.
Notwithstanding the enormous increase in public spending and the fiscal
deficit,pricesdidnotincreaseasexpectedbecausetradebarriersweretempor-
arilylifted,whilesomekeypublicpriceswithaheavyweightintheconsumer
price index were rarely raised, and never enough to recuperate the level that
inflation had eroded (Figure 7.2). But in any case, the overvaluation of the
exchange rate reappeared in the last third of the 1970s. Merchandise imports
increased at an annual rate of 25 per cent from 1976 to 1981, while exports,
especiallyoil,grewevenfastertopayfortheincreaseddebt-servicing,theresult
of not only more loans, but also the sudden increase in world interest rates
pushedbytheanti-inflationarypoliciesoftheUSFederalReserve.Thecollapse
of the balance ofpayments wasonly amatter of time, for itdepended on the
fate of the international oil market. In the summer of 1981, the first signs of
weaknessappearedandtheoilboomstartedtoturngloomy.Moreloanshadto
becontracted,increasinglyofashort-termnature,butthatwasnotsufficientto
preventalongoverduedevaluationofthepesoinearly1982.However,popul-
ist policies continued for several months, like the emergency wage increase
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 199
0
20
40
60
80
100
Consumer goods
Intermediate goods
Capital goods
1960 1970 1981
Figure7.3 Mexico:evolutionofindustrialstructure,196081(%)
decreedinMarchofthatyear,andespeciallysubsidiesandpublicspendingin
all sorts of projects, many of them only to maintain employment levels. The
debtcrisiswasaroundthecornerandactuallyeruptedinAugust.
Theexchangeratecrisisof1976 showed someofthestructuralproblemsof
thedevelopmentmodel.However,thealmostimmediateavailabilityofforeign
exchangeputaveiloveritssurfaceandthingsremainedthesame.TheLo pez
Portillogovernmentsimplyignoredsuchrestrictionsandtheoilandborrowed
resourcespaidforallinefficienciesandsubsidies.Thedebtcrisishadtoappear
tomakeeverybodyrealisethatthedevelopmentmodelwasnolongerworking.
Infact,thepoliticalstructurewasnotworkingeither,andittookseveralyears,
andahighsocialandeconomiccost,toacknowledgethereality.
* * *
IndustrializationinMexicoduringthetwentiethcenturyhasbeenstimulated
bymarketforcessuchasprices,byexternaldemandcausedbyexternalfactors
likewars,andbyinternaldemand.Inthislastinstance,populationgrowthand
urbanization,alongwithstateintervention,havebeenthemajorforcesbehind
the process of accelerated industrialization. The public sector has built infra-
structureandcreatedexternalities,hasestablished,acquiredandfinancedstate
corporations,andhasimplementedfiscalandtradepoliciestostimulateindus-
try and isolate the economy from external competition, first to diminish
importsandtotrytoavoidabalance-of-paymentscrisis,andthentogenerate
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table7.5 Mexico:positionofthe10mostimportantindustrial
branches,196081(GDPpercentages)
Industrialsector %
1960
1.Textiles 5.32
2.Meatandmilk 5.20
3.Nixtamalflour
a
4.68
4.Wheatflour 4.67
5.Steel 4.47
6.Oilanditsindustry 4.01
7.Garments 3.87
8.Shoeindustry 3.83
9.Non-metallicmineralproducts 3.27
10.Otherfood products 3.11
1970
1.Garments 5.32
2.Steel 4.52
3.Textiles 4.43
4.Meatandmilk 3.74
7.Wheatflour 3.68
8.Nixtamalflour
a
3.52
9.Shoeindustry 3.01
10.Paperandcardboard 2.99
1981
1.Steel 4.48
2.Carindustry 4.31
4.Clothing 3.99
5.Weavingandthread 3.88
6.Machineryandnon-electricequipment 3.55
7.Pharmaceuticals 3.41
9.Wheatflour 3.00
10.Meatandmilk 2.96
a
Localflourfortortillas.
Source: Villarreal(1988),tableLVII,p.375.
a larger domestic market and substitute imports, while simultaneously redu-
cing external dependence. In retrospect, it seems that the domestic economy
cannot be completely isolated from the world economy and that the latter's
fluctuations tend to affect the domestic market in various ways. The policy
instruments that the government has wielded in the past have not always
proven effective to sustain economic growth, partly because the instruments
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 201
themselvesarenotsufficientlyeffective,andpartlybecauseothergovernment
goals have stood in the way of utilizing more wisely such policy instruments.
The abuse of certain policy instruments, like the exchange rate and trade
restrictionswhichhavebeeneasypreytopoliticalpressure,aswellasthelack
of considerationof thelong-term viability of the economic strategy followed,
haveimpededovertheyearsMexico'seconomicdevelopment.
Notes
1 By domestic return value I mean that percentage of export value that remains in
the domestic economy in the form of wages and salaries, the power bill, domestic
inputs, taxes, and so on. It was calculated that around 66 per cent of export value
remained in the country during the 1920s. See Cardenas, 1987, tables 2.3 and 2.4,
pp.25,34.
2 AdetaileddiscussionmaybefoundinCardenas(1987),pp.4955.
3 Also,merchantsthemselveshadtoperformtransactionswithrediscountedbills.See
Cardenas(1987),pp.758.TheactualfiguresmaybefoundintableA2.8,p.214.
4 Implicit tariff rates declined as a consequence of increases in the price level of
imports in pesos, while the official specific tariff rate was kept fixed. Between the
firstandthesecondhalfofthe1930s,theimplicitadvaloremratedeclined24.1per
cent.SeeCardenas(1987),table4.5,p.107.
5 From 1935 to the date of the devaluation, the exchange rate became increasingly
overvalued which slowed down, and sometimes somewhat reversed, the process of
importsubstitution.
6 Seethecorrespondingarticleinthepreviousvolumeinthisseries(Thorp,1999).
7 ThesefiguresarequitesimilartothoseobservedinlargecountriesinLatinAmerica,
suchasBrazilandArgentina,whereitisconsideredthatimportsubstitutionwasan
importantfactorduringthe1930s.
8 ThatwasthecaseofpracticallyeveryotherLatinAmericancountry.Seeforinstance
thechaptersreferringtoColombia,ChileandArgentinainthisvolume.
9 This process continued during the following two decades. See Nacional Financiera,
1973,pp.2126.
10 AdetailedanalysisoftheselawsmaybefoundinMosk(1957),pp.63-7.
11 The`BraceroProgramme'allowedMexicanmigrantlabourerstoworkintheUSAfor
alimitedperiodoftime.
12 The currentaccountdeficitreached US$307million in thosetwo yearsbecausethe
country receivedforeign investment andloans,as well asrevenuesfromthe young
touristindustry,totallingUS$225million.SeeCardenas,1994,p.101.
13 In December 1943, the government decreed an increase in the import tariffs of
600 products, in a range from 10 per cent to 50 per cent, but it was not executed
apparently because of US pressure, and Mexico's concern that it would lose its
right to import certain capital goods during the war. In 1944 the government
made a list, valid until December 1945, of goods that would be subjected to
import quotas, but none was put into effect (Mosk, 1957, p. 70; Izquierdo, 1970,
pp.2478).
14 ThiswasagainstthetradeagreementthattheMexicangovernmenthadattemptedto
renegotiate, unsuccessfully, since mid-1945. The US government accepted it as a
temporarymeasuretoalleviatethepressuresonthebalanceofpayments.Theagree-
mentwasfinallyabrogatedinJuly1950.SeeTorres(1984),pp.22632.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
15 The difference between the first and the second half of the decade is due to the
devaluation of the exchange rate in 1954, which made foreign goods even more
expensiveandthusfurtherstimulatedthesubstitutionofotherimportedgoods.
16 We used the Jorgenson and Griliches methodology to calculate factor and residual
productivitychanges(Cardenas,1978,secondpart,table6).
17 ThefiguresrefertoMexicoCitywages(Bortz,1988,p.266).
18 Velazquezwas,untilhisdeathinJune1997at97yearsofage,absoluteleaderofthe
CTMandthemostimportantMexicanlabourleaderinthecountry'shistory.
19 Forexample, by1961NacionalFinancierahad provided loans to533 enterprises;it
was a minority shareholder in 29 of them, and majority or whole owner in 12: a
fertilizerproducer(FertilizantesdeMexico),twosteelmills(AltosHornosdeMexicoand
Siderurgica Nacional), two paper producers (Compana Industrial de Atenquique and
Fabrica de Papel Tuxtepec), two textiles factories (Ayotla Textil and Operadora Textil),
twosugarmills(IngenioRosalesandIngenioIndependencia),twowoodmills(Chapasy
triplay and Maderas Industrializadas de Quintana Roo), and a fish packaging plant
(Refrigeradora del Noroeste). With the sole exception of the fertilizer plant, all had
beenoriginallysetupbyprivatebusinessmen,andboughtfromthemwhenthreat-
enedwithbankruptcy(OrtizMena,1962,pp.623).
20 ThissectioniswidelybasedonCardenas(1996),chs2,3.
21 Forinstance,MexicoCityincreasedinsizeby71percentduringthedecade,from5.2
millioninhabitantsin1960to8.9millionin1970(NacionalFinanciera,1978,pp.4,
5,11).
22 Figuresinparenthesisshowaveragerealratesofgrowth,196371.
23 NacionalFinanciera(1978,pp.434),deflatedbytheGDPimplicitpriceindex.
24 This is an important point raised in the discussion by Katz (1976). Indeed, very
differentfirmswereabletocohabitinthoseyearsbecauseofprotection.
25 AntonioOrtizMena,thenMinisterofFinance,describedthegovernment'sposition
in the following terms: `...Mexico needs to develop... it can do it faster with
external help. There are two ways to obtain help: direct investments and credits.
Credits are undoubtedly the best of the two methods, because once they had been
paid,theindustryistotallyours'(OrtizMena,1965,pp.8267).
26 SepulvedaandChumacero(1973)quotedbyIzquierdo(1995),p.110.
27 In a contemporary article, Carlos D az-Alejandro (1965) showed that under various
circumstancesthedeepeningoftheimport-substitutionpolicytoincludeintermedi-
ate and capital goods was likely to be import-intensive, especially when industrial
growthwasveryrapid.
28 Public investment devoted to agriculture declined from 17.1 per cent in the
194956 period to 11.2 per cent in 196370, and the real price of corn, beans and
wheatdeclinedover20percentinthe1960s.SeeCardenas(1996),pp.757andtable
II.3.
29 The public external debt in 1959 was US$649 million, 5.5 per cent of GDP, and
increasedtoUS$4262.7millionby1970,12.0percentofGDP.Whenthedebtcrisis
broke out, the standing external debt had risen to US$58 874.2 million which had
helpedfinancechronicfiscalandcurrentaccountdeficits.
30 Thiscrisishasusuallybeenoverlookedintheliteratureprobablybecauseitwasvery
short-lived,asrescuefundsarrivedsoonandtheoilreserveswerequicklyrecognized
bytheinternationalcapitalmarkets.NegreteCardenas(1999)
31 RubioandGilD az(1987,table8,p.31)presentstartlingestimatesoftheresources
that were necessary to provide the increased demand for subsidized public goods
duringthe197782period.Accordingtotheircalculations,thepublicexternaldebt
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
EnriqueCardenas 203
wouldhavebeenUS$28billionless thantheactualamounthadthoseinvestments
nottakenplace.
32 AbriefsummaryofthePNDIispresentedinBrailovsky(1980),pp.285326.
References
Aguilar Cam n, H. and Meyer, L. (1989) A la sombra de la Revolucion Mexicana, Mexico:
CalyArena.
BancodeMexico(1932)InformeAnual1931,Mexico.
BancoNacionaldeMexico(1931)ExamendelasituacioneconomicadeMexico,Mexico,nos
73and74(OctoberNovember).
Boltvinik,J.andHernandezLaos,E.(1981)Òrigendelacrisisindustrial:elagotamiento
del modelo de substitucio n de importaciones. Un analisis preliminar', in R. Cordera
(ed.),Desarrolloycrisisdelaeconom amexicana,Mexico:FondodeCulturaEcono mica.
Bortz,J.(1988)LossalariosindustrialesenlaCiudaddeMexico,19391975,Mexico:Fondo
deCulturaEcono mica.
Brailovsky,V.(1980)ÙsodeunModeloIntersectorialDinamicodelaEconom aparala
Programacio n Industrial', in N. Lustig (ed.), Panorama y Perspectivas de la Econom a
Mexicana,Mexico:ElColegiodeMexico.
Brothers,D.andSol s,L.(1967)EvolucionfinancieradeMexico,Mexico:CentrodeEstudios
MonetariosLatinoamericanos.
Bueno,G.(1970)`Lasperspectivasdelapol ticadeldesarrolloindustrialenMexico',inL.
Sol s(ed.),Laeconom amexicana.Vol.I,Analisisporsectoresydistribucion,Mexico:Fondo
Buffie,E.(1990)ÈconomicPolicyandForeigndebtinMexico',inJ.Sachs(ed.),Develop-
ing Country Debt and Economic Performance, Vol. 2, National Bureau of Economic
Research,Chicago:TheUniversityofChicagoPress.
Cardenas,E.(1997)ÀMacroeconomicInterpretationofNineteenthCenturyMexico',in
S.Haber(ed.),HowLatinAmericaFellBehind.EssaysontheEconomicHistoriesofBraziland
Mexico,18001914,Stanford,California:StanfordUniversityPress.
(1996) La pol tica economica en Mexico 19501994, Mexico: El Colegio de Mexico/
FondodeCulturaEcono mica.
(1994) La hacienda publica y la pol tica economica 19291958, Mexico: Fondo de
CulturaEcono mica/ElColegiodeMexico.
(1987) La industrializacion mexicana durante la gran depresion, Mexico: El Colegio de
Mexico.
(1978) El crecimiento economico de Mexico, 19501975, Mexico: licenciatura thesis,
ITAM.
Cardenas, E. and Manns, C. (1987) Ìnflation and Monetary Stabilisation during the
Mexican Revolution', in P. Bardhan, A. Fishlow, and J. Behrman (eds), International
Trade, Investment, Macro Policies and History. Essays in Memory of Carlos D az Alejandro,
Amsterdam:North-Holland.
Carreras de Velasco, M. (1974) Los mexicanos que devolvio la crisis 19291932. Mexico:
Secretar adeRelacionesExteriores.
D az-Alejandro, C.F. (1965) Òn the Import Intensity of Import Substitution', Kyklos,
no.3.
Hansen R. (1971) The Politics of Mexican Development, Baltimore, Md.: Johns Hopkins
UniversityPress.
HernandezChavez,A.(1993)ÈstadoNacionalista.Sureferentehisto rico',inJ.O.Rodr -
guez(ed.),TheEvolutionoftheMexicanPoliticalSystem,Wilmington,Del.:SRBooks.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Izquierdo, R. (1995) Pol tica hacendaria del desarrolloestabilizador, 19581970, Mexico: El
ColegiodeMexico/FondodeCulturaEcono mica.
(1970)ÈlproteccionismoenMexico',inL.Sol s(ed.),Laeconom amexicana,Vol.I,
Analisis por sectores y distribucion. Mexico: Fondo de Cultura Econo mica. Jorgenson,
D.W. and Griliches, Z. (1967) `The Explanation of Productivity Change', Review of
EconomicStudies,pp.24983.
Katz,J.M.(1976)Importaciondetecnolog a,aprendizajeeindustrializaciondependiente,Mex-
ico:FondodeCulturaEcono mica.
Mosk,S.A.(1957)IndustrialRevolutioninMexico,Berkeley:UniversityofCaliforniaPress.
NacionalFinanciera(1984)Laeconom amexicanaencifras,Mexico.
(1978)Laeconom amexicanaencifras,Mexico.
(1973)`Lapol ticaindustrial',inL.Sol s(ed.),Laeconom amexicanaVol.I.Analisispor
sectoresydistribucion,Mexico:FondodeCulturaEcono mica.
(1955) Convenio de estabilizacio n monetaria,in El Mercado de Valores, no. 49, 5
December.
NegreteCardenas,S.(1999)`MexicanDebtCrises:ANewApproachtotheirGenesisand
Resolution',PhD,UniversityofEssex,England.
OrtizDietz,H.(1985)ElInversionistaMexicano,Mexico,MexicoDataBank.
Ortiz Mena, A. (1965) `Los Creditos Exteriores y el Desarrollo Industrial', conference
beforemembersofConcam n,reproducedinElMercadodeValores,no.47,22Novem-
ber.
(1962) `Discurso en la sesio n inagural de la XXVII Convencio n Nacional Bancaria
(1961)', reproduced in Pol tica Financiera, Seleccio n de Estudios Latinoamericanos
(SELA),ed.PESA.
OrtizMena,R.et al., (1953)Eldesarrolloeconomicode Mexicoysucapacidadparaabsorber
capitaldelexterior,Mexico:FondodeCulturaEcono mica.
Rubio, L. (1989) Èl sector privado en Mexico: pasado y futuro', in J. Elguea (ed.), La
econom amexicanaysusempresarios,Mexico:UniversidadAnahuacdelSur.
Rubio, L. and Diaz, G. (1987) A Mexican Response, New York: Twentieth Century Fund
Paper,PriorityPress.
Sepu lveda,B.andChumacero,A.(1973)LainversionextranjeraenMexico,Mexico:Fondo
ShveidFinkelstein,F.(1972)ImportSubstitutionIndustrialisationinMexico,thesissub-
mittedforthedegreeofB.Litt.HertfordCollege,OxfordUniversity.
Sol s, L. (1981a) Economic Policy Reform in Mexico. A Case Study for Developing Countries,
NewYork:PergamonPress.
(1981b) La realidad economica mexicana. Retrovision y perspectivas, revised edition,
Mexico:SigloXXI.
Szekely,G.(1983)LaEconom aPol ticadelPetroleoenMexico,197682,Mexico:ElColegio
deMexico.
Thorp,R.(ed.)(1999)LatinAmericainthe1930s:TheRoleofthePeripheryinWorldCrisis,
2ndedn,London:Macmillan.
Torres,B.(1984)HacialaUtop aIndustrial.HistoriadelaRevolucionMexicana194052,vol.
21,Mexico:ElColegiodeMexico.
Villarreal,R.(1988)Industrializacion,deudaydesequilibrioexternodeMexico,Mexico:Fondo
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
8
IndustrializationinVenezuela,
193683:TheProblemofAbundance
*
PabloAstorga
Introduction
ThefirstimpressionanextraterrestrialagentapproachingtheEarthwouldhave
ofVenezuelaisthatthecountrydiffersfrommostofitscontinentalneighbours
inatleasttwoaspects,oneduetonature,theothertonurture;ithasoilanditis
notgoodenoughatplayingfootball;ortoputitpositively,itisnottoobadat
playing baseball when compared, for instance, to Brazil. Therefore he would
wonderwhyacountrythathassomanyrootsincommonwiththerestofLatin
Americahasdevelopeddifferentskillsinsports.Hewouldnoticethatthepoor
competitiveleveloffootballisnotduetolackofaficionados;thatthedemand
sidedoesnotseemtobetheproblem.Withmorecarefulobservationbasedon
a long-term view of this century, he would notice also that the country is a
latecomerintermsofthedevelopmentofitsindustrialcapacity,butthatatthe
endofthecenturyitshowsasimilarstructureandproblemscomparedtoother
countriesintheregion.Thenhewouldwonderwhyacountrysorichinoilhas
cometothesameresultintheeconomy,ifnotinsports.Thisistheconcernof
this chapter: Why did an abundance of foreign currency generate much the
same problems as scarcity of it did? Why did different circumstances lead to
similarresults?Wewilladdressthiscentralquestionbylookingattheeconomy
andthehistoryoftheindustrializationofthecountry.
TheconventionalwisdomaboutindustrializationinVenezueladescribesthe
countryasalatecomerandasanunsuccessfulstoryofISIimport-substitution
industrialization presenting many of the ills normally attributed to the
* Ihave greatlybenefited from commentsand observations from R. Thorp, V.FitzGerald,B.Mom-
mer,D.B.UrbanejaandJ.A.Ocampo.Thanksalso toM.Deasforringingsomebells.Ialsowantto
thanktheparticipantsoftheworkshopsheldinOxfordandPaipaonSeptember1996 andMay1997.
During the first stages of the researchI received generous help and information from A. Baptista, S.
Barrios,R.delaCruz,A.Frances,H.Go mez,C.Machado,S.Moncada,H.SontagandH.Viana.Thanks
alsotoJ.Brillembourg,A.Celis,J.ColesandJ.Zarragaforinterviewsgiven.
205
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
206 IndustrializationinVenezuela,193683
majorityofexperiencesintheregionsuchaslowproductivitygains,excessive
diversification,andexcessiveprotectionleadingtoahighlyinefficientindustry
still very dependent on imports and foreign technology. However, what is
surprisinginthiscaseisthatthepresenceofoilmeantthattheresultsobtained
were not driven by a balance-of-payments constraint at least not until the
1980sasinmostothercountries(Macario,1964;Bruton,1967).InVenezuela
themainmotiveforceofindustrializationwasadeliberatedevelopmentpolicy
andinthatrespectitwasmorearesultofactionbythestatethanareactionto
externalcircumstances.
1
So,inthiscasetheÀchillesheel'oftheinward-look-
ingmodelhastobefoundelsewhere.
Thestructureofthischapterisasfollows:itopenswithahistoricalaccount
ofindustrializationinthecountrywhereperiodsareidentifiedaccordingtothe
presenceofsignificantpoliticaloreconomicchanges.Beginningin1920atthe
dawnoftheoilera,itthenmovesontotheperiod193683whichisthemain
focus of this study. The second part explores some basic long-term economic
issuesinordertoplacethedevelopmentofindustryinabroadercontext.Three
issues are addressed: (i) the links between oil and the non-oil economy and
changesinthestructureofgrowth;(ii)therecordoflabourproductivityduring
theperiodunderanalysis;and(iii)theprocessofcapitalaccumulation,physi-
cal and human, that gave support to the industrialization effort. Finally a
sectionofconclusionsispresented.
Giventheimportanceofoiltothedevelopmentofindustry,wewillstartby
offering some basic information about the importance of oil for the
economy. In 1925 oil surpassed coffee as the main export product;
2
since
then it has represented on average 90 per cent of total exports. The capacity
of the country to import increased considerably:
3
In 1950 exports bought 40
times more imports than in 1920, and in 1980 with the second boom
associated with price rises the purchasing power of exports was four times
greaterthantheoneregisteredin1950.Onthefiscalside,oilreplacedcustoms
as the predominant source of revenues for the government after the Second
WorldWar.Sincethenitscontributiontothefiscalreceiptshasbeenaround75
percent.
Although Venezuela still continued to be an economy with a high level of
exportconcentration,thechangetooilastheprincipalproductnotonlyledto
fundamental transformations regarding import capacity and fiscal revenues,
butalsotomorequalitativeaspectssuchasatypeofproducthighlyintensive
in the use of capital, and the ownership of the export sector which was
now exclusively in foreign hands. As a consequence, the fight for the appro-
priation of oil rents became one of the central issues of state action. To the
extentthatamajorproportionofrentswasextractedfromtheoilcompanies,
there was an increase in the amount of resources that could be channelled to
diversifytheinternaleconomy.Inthiswayacloselinkdevelopedbetweenoil
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 207
revenues, the state, and business opportunities of national private capital;
industry grew highly dependent on, and vulnerable to, conditions in the oil
market.
Briefhistoryofindustrialization
4
Wedividethisbriefaccountofindustrializationintothreemainsections.The
firstpresentstheantecedentscoveringtheperiodfrom1920to1935.Itshead-
ingCropsandOilconveystheideathattwosectorswereprovidingthesource
ofgrowth;traditionalagriculturalexportsdominatedbycoffeeandtoalesser
extentcocoa,andthenewlydiscoveredoil.Itisalsoaperiodoftransitionfrom
coffee to oil, two activities that at some point were competing for labour, but
which mostly clashed in a less direct way through movements in the foreign
exchangerate.Theheading,byincludingonlytwoactivities,alsopointstothe
absence of manufacturing as a significant source of growth, so in a sense this
first part will deal more with the forces retarding industry than with those
encouragingitsgrowth.
This period, dominated by export-led growth, is followed by a period of
change where the political discussion revolved around the need to diversify
the non-export economy. This is entitled To Sow the Oil, a metaphor for
industrialization. The traditional export economy fades into the background,
eclipsedbytheshiningoil,andtheactivityofsowinglosesitsnaturalconnec-
tionwiththecrop.Thethirdsection,IntheShadeofOil,coverstheprocessof
consolidation of manufacturing in the country. What was sown and growing
still needed to be protected, but the nature of the protection given, and its
permanence, generated an industry highly dependent on oil revenues. The
historical overview ends in 1983, the year that marks the beginning of a new
period characterized by falling oil revenues, the debt crisis and balance-of-
paymentproblems.
Cropsandoil,192035
Theearlyyears
The general picture of Venezuela in 1920 was of a rural country where 75 per
centofthepopulationwasilliterateandhadalifeexpectancyof33years.Soit
was not a society prepared to take the necessary steps that material progress
required (Baptista, 1984). This situation soon started to change thanks to the
developmentoftheoilindustryanditsinitialinvestmentintheareaofinfra-
structure, mainly directed to building a communication system. Backward
linkagesofthiskindwereprominentinareasneartotheoilwellswhichwere
untilthenunderdeveloped.Thethesisofoilasanenclavesectorhastodomore
with its contribution to employment,
5
but there were also important draw-
backs.Theretainedvalueofoilexportswaslowcomparedtothegainsobtained
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
bythecompaniesandtheexplorationandexploitationofoilhadanimportant
environmentalimpact.
6
Inthe manufacturing sector, despite thelack of official statistics during the
period, there is evidence of a relatively important development in the textile
andrefineryindustries,andalsoawell-establishedfoodandbeveragesindustry.
The bulk of manufacturing activity was concentrated in Caracas and Mara-
caibo, and with the exception of the refining industry it grew in the shade of
a tariff wall and yet was still highly dependent on imports of raw materials,
machinery and equipment. The following quote of P. L. Bell, an American
envoy of the US Department of Commerce, depicts in detail the situation in
manufacturing at the beginning of the 1920s characterized by an incipient
processofimportsubstitutioninlightindustries:
7
...The chief manufacturing industry in Venezuela is that of the cotton
factories, of which there are three in Caracas, two in Valencia, one in
Maracaibo, and one in Cumana...they produce about 50 per cent of the
rough common cloth used in the country. The total investment is over
$2000000...Next to the cotton mills, the most important manufacturing
industry is that of cigarette making...There are three large breweries in
Caracas, Maiquet a and Maracaibo...Two factories making other goods
compete directly with the imported articles, these are the paper factory in
Maracay, which uses domestic raw material, and the paper factory at Car-
acas,calledtheFabricaNacionaldePapel...ThemoderndairyplantatMar-
acay is turning out an average of 300000 pounds of butter a year...The
plantisalsoequippedforcheesemakingandcondensed-milkmanufacture,
and it has its own can-making shop...Practically all of the shoes worn in
Venezuela are made in the great number of small local shops...All the
cheap grades of toilet soap, laundry soap, etc, are made in Venezuela by
themanysoapfactorieswhichusethedomesticcottonseedoil,coconutoil,
etc,producedinthecountry.
(Bell,1922,pp.11516)
There was an incipient development of manufacturing based on a process of
import substitution in light industries. Absent in the preceding description is
the refining activity which, given the relevance of the exploitation of oil on
largescaleforthemodernizationofthecountry,wasthefirstnaturalcandidate
within the industrial sector to be developed. However, it was only after the
SecondWorldWarthatthisactivitybecameimportantinVenezueladespitethe
country being the leading exporter of oil during the 1930s.
8
The main oil
companies operating in the country built the first two large refineries in the
smallislandsofCuracaoandArubaduringtheperiod191729.Thedecisionto
establishtheserefineriesoutsideVenezuelanterritoryisanimportanteventin
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 209
the development of manufacturing because, despite its capital intensity in
refiningandweaklinkages(Harris,1971),itdeprivedthecountryofanimport-
antsourceofinvestment.
9
Themainreasonbehindthedecisionofoilcompa-
nies had to do with the risk factor induced by political instability. From the
pointofviewoftheoilcompanies`refiningeggs'weretobekeptinaseparate
basket from `producing eggs'.
10
In that respect the Dutch colonies were a less
riskyplaceandofferedagoodlocationanddeeperwaters.Onthenationalside
little was done in order to produce a result more favourable for the country's
industrial development. The lack of pressure was due to the weak bargaining
position of the Venezuelan government at that time, and the lack of legal
instruments (McBeth, 1983), although there areclaims indicating the opposi-
tionofGo meztotheinstallationofrefineriesinVenezuela.
11
Incomparativetermssomeindustriesinthecountrywerenotfarbehindthe
level of development reached in other economies. There is evidence that the
amount of textile machinery installed by the end of the 1920s was larger in
Venezuelathaninanyothercountryofcomparablepopulationsize,although
itlaggedfarbehindthoseofBrazilandMexico(Karlsson,1975).However,there
are several indicators pointing to a general picture of an underdeveloped
industry. For example, the consumption of electricity in the country in 1937
wasoftheorderof20kW/hourspercapita,belowotherlatecomerssuchasPeru
(50),Colombia(34),andfarbelowcountriessuchasChile(302).
12
Theofficialviewwasthatthedevelopmentofthecountrywasonlypossible
with the participation of foreign capital and population. Despite the impor-
tance that oil acquired as a source of foreign exchange and government rev-
enues, the sector was seen as an enclave alien to the country. Agriculture was
still considered the sector where the future of the country would lie.
13
In
contrast to the oil industry the regime did not succeed in attracting foreign
investmentintherestoftheeconomy;alowrateofprofitwhichwasbelowthat
offeredbyothercountriesintheregionhasbeenadvancedasthemainreason
forthelackofinterest(Sullivan,1976).Thislowreturnpointstothelimitation
oftheinternalmarketandthefirstsymptomsofanovervaluedexchangerate.
Effectsofthe1929Depression:arichstateinanimpoverishedsociety
TheDepressionseverelyaffectedcoffeeandcocoavialowpricesandincreased
protectionism in traditional foreign markets. Oil production and prices
declined too but to a lesser extent and made a quicker and stronger recovery.
Maracaibowasperhapsthemostaffectedcity.Beingatthetimeanimportant
portforcoffeeexportersandforimportstotheAndeanregion,itsufferednot
only from the slowdown of oil production but also from the coffee crisis.
ImportsthroughtheportofLaGuairadropped,buttoamuchlesserextent.
14
The consequences of the crisis dominated the 1930s, and led to some policy
responses such as the introduction of a preferential exchange rate, known as
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the Convenio Tinoco, in order to protect coffee exporters against the strong
bolivar. The contraction in demand and the move towards revaluation also
affected manufacturing industry, but in this case the government was less
concerned about taking measures in order to alleviate the crisis in the sector.
Thetextileindustryexperiencedadeclineinitsactivitythroughoutthedecade
andfacedanincreaseinforeigncompetition,whichwasincontrasttomostof
thecountriesintheregion.
15
However, when placed in the Latin American context Venezuela appears as
oneoftheless,ifnottheleast,affectedcountriesintheregion.Thecountry's
termsoftradeanditscapacitytoimportthatis,thepurchasingpowerofits
exportsduringtheearly1930swerenotsoseverelyhitasinothercountries,
and the oil market was one of the first to recover from the depression.
16
The
countrywasnotonly an exceptioninthe intensityof theimpactof thecrisis
butalsointhecombinationofpoliciesadoptedinresponsetoit.Whilemostof
the governments in the region devalued their currencies and implemented
expansionaryfiscalpolicies,inVenezuelathethesisofrevaluationprevailed
17
the country was accumulating reserves and the government was running a
fiscalsurplusfrom1932to1935.
18
During193031 therewas afiscaldeficit ofBs.50.7 millions thelargest so
far which was mostly due to the cancellation of the external debt (Bs23.8
millions equivalent to 0.63 million) as part of the commemoration of the
centenary of Bolivar's death and the decline of receipts from import duties
(Veloz,1945).Thereaftertherewereconsecutivesurplusesuntil1936.Receipts
in real terms dropped 12 per cent in 1930 but then quickly recovered and by
1934theyweremorethan25percenthighercomparedtotheamountreceived
in 1928. The internal debt (amounting to Bs20.7 millions) was cancelled in
1935. There was a growing paradox of a rich government or actually a rich
governor in an impoverished country. General Go mez was the principal
landowner in the country and the most powerful businessman (Sullivan,
1976, p. 266). After his death the majority of his properties were confiscated
and thereby contributed to the beginning of the participation of the state in
production activities. To summarize, the crash of 1929 was not used as an
opportunitytoacceleratetheprocessofimportsubstitutionanddevelopment
ofindustry,aswasthecaseinothercountriesoftheregionsuchasArgentina,
Brazil,ChileandColombia.
Tosowtheoil,193647
Aroleforthestate
After the death of Go mez in December 1935 the economy was in a bad state
andthepoliticalsituationwasmarkedbyspreadingdiscontent.Hissuccessor,
GeneralLo pezContreras,putinplaceafirst,modest,programmetoencourage
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 211
industrial development in the non-oil economy. The need for state interven-
tion was acknowledged, given the reluctance of domestic capital to venture
intoareasbeyondthesafeandhighlyprofitableonesofimporttrading,bank-
ingandreal-estatespeculation.Followingthesameobjectivein1937anindus-
trial bank was created (Banco Industrial de Venezuela) with 60 per cent of the
capital subscribed by the government. The government also implemented
policiestoprovide,throughtheprivatebankingsystem,specialcreditfacilities
forthoseexistingandprospectivefirmsmeetingpublicpriorities.
19
Althoughtherewasgeneralagreementontheneedforstatesupporttoindus-
tryandafavourabletradepolicy,therewassomecontroversyabouttheextent
andtypeofindustriestobeencouraged,whethertodevelopagro-industryorgo
beyond what the comparative advantages would dictate. Beyond differences
aboutthesortofindustrytobepromoted,therewerecommonproblemstobe
addressed in order to make progress in developing any industry. The most
urgent were the limits imposed by the reduced size of the market and an
effectivedemandinadequatetojustifyinvestmentinmanufacturing,thelack
ofatrainedlabourforce,andtheinhibitingeffectofastrongcurrency.
Thisperiodisveryvigorousinthecreationofinstitutionsthatwouldshape
the modern state and provide the main instruments for its intervention. The
mostimportantintheeconomicspherewere:theLeydeAranceldeAduanasof
1936, and the Ley de Impuesto sobre la Renta of 1942, which was followed the
next year by a new hydrocarbons law. The Banco Central de Venezuela was
created in 1939 and a law to regulate the banking system was approved in
1940.Regardingsocialinstitutions,alawregulatinglabourconditionsLeydel
Trabajo came into force in 1936, and four years later an education law was
passed and social security was introduced. Finally an agrarian reform law was
passedbyCongressin1945.
20
The effort to stimulate new sources of national production was to some
extent counteracted by the 1939 `trade policy agreement' with the USA.
21
The terms of thetreaty responded to theneed toguarantee the sales ofVene-
zuelanoilinitsmainmarketbyavoidingtheimpositionofimportcontrolsby
the US government.
22
In exchange for this concession the government held
backfromimposingrestrictionsonaconsiderablenumberofUSexports,most
of them of products manufactured. Through this agreement Venezuela gave
dutyconcessionson88itemscovering35percentofthetotalimportsfromthe
USA. Reciprocally, 90 per cent of the total Venezuelan exports to the USA
consisting mainly of oil, coffee and cocoa were covered by the agreement.
23
Therewasonthisoccasionacleartrade-offbetweenoilandlocalindustry,and
thisagreementhadimportantconsequencesfortheinstrumentsofprotection
tobeusedbythegovernmentduringthenexttwodecades.Themainmechan-
ismwastheclausuladeescapethatallowedforquantitativerestrictionsonthose
itemswithsignificantlocalproduction.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Lo pezContrerasfollowedamoreaggressivepolicyregardingtheincreaseof
the country's refining capacity to serve export markets. In this respect new
concessionsweremadeconditionalontheprocessingofpartoftheproduction
inVenezuela.Bythebeginningof1943theobjectivesofhissuccessorGeneral
Medinabecamemoreambitiousandhecontemplatednotonlythewideningof
the existing refinery capacity, but also moving the refineries located in Aruba
andCuracaotoVenezuelanterritory(Machado,1990).Finally,in1943theoil
companies agreedtoinstallinthecountry,withinatime nogreaterthanfive
years, a complete refining plant with a capacity of no less than 40000 barrels
perday.
SecondWorldWaranditsaftermath
Thewarbroughtaboutashortageofimportsthatexposedtheexternalvulner-
abilityofthecountry.Thesupplyside,ingeneral,didnotreactandasituation
of excess demand created serious inflationary pressures. The institutional
responsewasthecreationoftheJuntadeAbastecimientoandtheJuntaNacional
ReguladoradePrecios.AscanbeseeninFigure8.1,manufacturingoutputhada
poor performance growing at a 0.7 per cent annual compound rate between
1939and1944.
Total imports, excluding those made by foreign-owned industries, fell from
Bs.326.2millionsin1939toBs.215.7millionsin1942(adropof33percent),
whereas the oil and gold companies reduced their imports by 36 per cent
during the same period from Bs.82.7 millions in 1938 to Bs.53 millions in
1942. In contrast with what happened to imports, oil exports saw only a
significant decline in 1942 when they registered Bs.440 millions, lower than
100
1 000
10 000
100 000
T
h
o
u
s
a
n
d
s
B
s

o
f

1
9
6
8

1
9
3
6
1
9
4
0
1
9
4
4
1
9
4
8
1
9
5
2
1
9
5
6
1
9
6
0
1
9
6
4
1
9
6
8
1
9
7
2
1
9
7
6
1
9
8
0
1
9
8
4
1
9
8
8

Years
Figure8.1 Venezuela:valueaddedinmanufacturing,193689
Source: Baptista(1997).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 213
thefiguresofBs.469andBs.673millionsfor1939and1941respectively.
24
The
limitations of importing capital goods made the creation of new productive
capacity difficult. This was reflected in the fall of the value of total gross
investmentduringtheperiod.
Yetdespitethisevidenceofgeneraleconomicstagnation,thewarwasatime
ofopportunitiesforthecreationofnewfirmsandtheexpansionofproductive
capacityinmanufacturing.Thisperiodwitnessesthebeginningofsomeofthe
economic groups that were to dominate the economic scene for the years to
come.AnexampleistheMendozagroup,foundersofProtinal,afirmproducing
animal food, in 1941 and the cement factory Vencemos in 1943, the Delfino
groupwiththeenlargementofCementosLaVega,andalsotheCervecer aPolar
establishedin1941.
25
The stimulus given to trade within the region during the war was also an
occasionforthecountrytocomparethedevelopmentofitsindustrywithother
Latin American countries.
26
The vulnerability of the country exposed by the
interruption of trade produced an urge to diversify the economy. Despite its
late start Venezuela soon began to catch up at a fast pace. There was an
agreement on the need to `sembrar el petroleo', that is, to use the proceeds
comingfromittodevelopnationalproductiveforcesandtransformthecoun-
tryintoamoderneconomy.ToachievethisendtheCorporacionVenezolanade
Fomento(CVF)wascreatedin1946 bytheJuntaRevolucionariapresidedoverby
Ro mulo Betancourt and soon became the mainspring of economic develop-
ment in the country.
27
Its nature, structure and functioning were heavily
influenced by the experiences undergone in Chile with CORFO (created in
1939) and in Mexico with the Nacional Financiera Mexican Society of
IndustrialFinancefoundedin1934.TheCorporacionhadbroadpowers,includ-
ingthecapacitytoestablishandoperatebusinesses,acquiresharesinexisting
enterprises, and make loans for up to 20 years and the authority to make
surveys. During its first years of operation the major proportion of its
resources was devoted to the expansion of food production and electric
powergeneration.
Anotherimportanteventintheprocessofindustrialdevelopmentduringthe
periodwastheparticipationoftheInternationalBasicEconomic Corporation
(IBEC) in Venezuela, a profit-seeking initiative organized in 1947 by Nelson
Rockefeller to foster modernization in Latin America. The Betancourt govern-
mentgavesupporttothisinitiativeinthecountry.Itbeganwiththeorganiza-
tion of four projects in the areas of agro-industry, exploitation of fisheries,
production of dairy products, and distribution of consumer goods based on
the idea of a supermarket chain.
28
The IBEC contributed 50 per cent of the
capitalthemoneycameoutoftheprofitsoftheoilcompaniesandtheother
halfwaslefttotheCVF,undertheagreementthataftertenyearsfromthestart-
upofafirmitscapitalshouldbecomeVenezuelan,eitherpublicorprivate.In
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
1948thenewly-electedpresidentRo muloGallegosAccionDemocratica(AD)
wasoverthrown.Thenewmilitaryregimewasnotanenthusiastfortheproject
andin1950itwasevidentthatitwantedtoterminateitsinvolvement.In1951
thepresidentoftheCVFannouncedtheendoftheagreementwiththeIBEC.
Intheshadeofoil,194883
The effortsof industrialization initiatedin the1940s were tocontinue during
the coming decades at an accelerated pace. Table 8.1 includes information
about manufacturing excluding refining.
29
When the first industrial census
was made in 1936 the traditional industries of food and textiles represented
more than 60 per cent of industrial employment and capital in non-refining
manufacturingandaccountedformorethanhalfofitsoutput.
30
Lessthan50
yearslaterthissectorexperiencedasignificantreductioninitssize,andin1981
accounted for less than 20 per cent of employment and capital, and around
Table8.1 Venezuela:sharesofselectedindustriesinmanufacturing(excludingrefining)
(percentages)
F,B&T Tex. Paper Chem. Non- Bas. Met.- Total
met. met. trans. (Bs)
Output 1936 52.3 10.6 0.1 4.5 3.8 0.3 2.4 157
1953 41.6 8.1 1.2 8.5 9.7 0.4 3.7 1229
1961 41.0 15.6 3.3 8.3 6.5 0.9 11.1 3917
1971 29.2 12.1 4.3 10.2 6.7 6.0 17.7 9034
1981 27.2 8.1 3.9 10.8 6.8 7.8 19.6 19485
Labour 1936 59.3 11.9 0.2 2.5 4.9 0.2 1.9 47400
1953 48.9 8.4 1.0 4.1 6.8 0.3 3.8 95518
1961 27.2 23 2.8 5.0 6.6 1.2 14.8 149976
1971 24.7 15.4 3.4 7.5 6.9 4.6 16.8 237313
1981 18.1 11.2 2.9 7.6 6.6 7.2 18.4 456155
Capital 1936 43.2 9.6 1.3 295
1953 53.0 9.0 1.4 5.9 12.7 0.5 2.3 1499
1961 28.2 9.1 3.2 10.6 7.5 26.5 6.5 6407
1971 23.7 9.4 4.8 7.9 9.5 23.3 10.8 5508
1981 18.8 3.8 2.8 6.0 10.4 34.7 13.9 18664
Notes: Outputvalueadded;Labournumberofemployees;CapitalFixedassets.
ThebracketsinthefollowingarethecorrespondingnumbersoftheInternationalStandardIndustrial
Classification,ISIC:
F,B&T:food,beveragesandtobacco(311314);Tex.:textilesandclothing(321);Paper:paperand
paperproducts(341);Chem.:chemicalindustry(3512);Non-met.:non-metallicminerals
(361,362,369);Bas.met.:basicmetals(37172);Met.tran.:metal-transformingindustries(3814);
Total:manufacturingoutputexcludingrefiningindustry.Thefiguresusedtoobtainthepercentages
ofcapitalandvalueaddedin1936 and1953areincurrentBs,thereafterareinBsat1968prices.
Sources: Karlsson(1975)for1936 figures(basedonthefirstindustrialcensus)and1953;OCEI,
EncuestaIndustrial(industrialSurvey)for1961,1971,1981.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 215
27 per cent of total value added. During this time the country went through
aprocessofrapidintroductionanddevelopmentofnewindustrialactivities.
The`take-off'andthestrategicindustries
The military dictatorship of Perez Jimenez (195258) focused its economic
actiononthedevelopmentofthoseindustriesthatwereconsidered`strategic'
forthecountry,namely,steel,petrochemicalsandhydroelectricpower.
31
Dur-
ingthisperiodpublicworks,inparticulartheexpansionofthecommunication
system,becameapriorityforthegovernmentwhich,underwhatwasknownas
`thebulldozerpolicy',allocatedduringthedecadeonaverage32percentofthe
totalbudgettothatend(KornblithandMaingo n,1985).Therewasasignific-
antreductioninthebudgetandfunctionsoftheCorporacion.Amongthemain
changes in its orientation were cuts in those funds directed to direct invest-
ment,withconcentrationofitseffortsonpromotingtheprivatesectorthrough
credit supply.Officially the decision for thechange in direction was basedon
argumentsrelatedtoefficiency,theoverlappingoffunctionswithotheragen-
cies, and the need to reformulate what was thought to be an over-ambitious
project scattered over a great variety of fields. Apart from these technical
reasons, there was a strong political motive behind this curtailment of func-
tions which is connected to the identification of the CVF with the previous
radicaladministrationofAD.
Manufacturingoutputgrewatanaverageannualrateof12percentthrough-
outthe1950s.Yetthisspectaculargrowthwaspartlyduetotheverysmallscale
ofthesectorin1950whenitwasresponsibleforonly11percentofGDP15
percentifoilisexcludedfromGDP.By1953therewerealreadythefirstsignsof
theestablishmentofindustriesinintermediategoods,particularlycementand
chemicals,representingapproximately10percentofemploymentand20per
cent of capital and output of non-refining manufacturing. The traditional
industriesfoodprocessing,beveragesandtextilesgrewlessrapidlybutthis
growthwasstillsignificantandtheyremainedthebiggestintermsofemploy-
mentor value added. Atthe coreof theseresults were traderestrictions, soin
thosecaseswhereprotectionornaturaladvantageswerenotveryhighsuchas
powderedmilk,cigarettesandtextilestheincreaseddemandwascoveredby
imports(ECLA,1966a).
32
Also,complementingtheagreementreachedin1943
during the negotiations of the new Hydrocarbons Law, the oil companies
undertook a major expansion of domestic refining capacity.
33
Venezuela's
refining capacity increased tenfold between 194560, from 100000 barrels
per day in 1945 to 1.4 million barrels per day in 1960 (Banco Central de
Venezuela, 1966). As a result of these investments, by 1960 30 per cent of oil
extractedinthecountrywasrefinedlocally.
Thecontributionofforeigncapitalinthisindustrial`take-off'coincidedwith
thespreadofmanufacturingTNCs
34
throughoutthecontinent.By1957there
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
were25subsidiariesofUSTNCsoperatinginmanufacturing,someofthemin
joint ventures with national firms (Mikesell, 1962). A second contribution of
foreign capital, more in the form of human capital and entrepreneurial skills,
camefromthenumerousgroupsofimmigrantsthatestablishedthemselfinthe
countryafterthewar.Agoodexampleofthecontributionofimmigrantstothe
developmentofindustryistheNewmanbrotherswhoventuredintothepaint-
makingindustrywithPinturasMontana,theseminalfirmforGrupoCorimon, a
leadergroupinthesectorandcurrentlyoneoftheprincipaleconomicgroups
inthecountry.
In 1952 an incident occurred illustrative of the way the breakdown de-
velopedbetweenthelocalprivatesectorandthestateintheareaofproduction,
characterizedbytheexclusionoftheformerfromthemineral-basedindustries.
A group of Venezuelan entrepreneurs led by Eugenio Mendoza and Alberto
Vollmer envisaged the creation of a private steel industry in the Guayana
region.
35
The group was called the Iron Syndicate and wanted to invest
Bs.200millions(approx.US$65million)inbuildingasteelmillwithacapacity
to produce 150000 tons/year 30 per cent of the national demand in 1953.
Thiswasthemostimportantinvestmenteverconceivedbynationalcapitalists.
TheUSSteelCorporationjoinedtheSyndicateprovidingthetechnicalexpert-
ise. Originally, the military government offered its support to the Syndicate,
but soon after announced that the steel programme would be directed by the
state. The government considered the steel sector too important to be left in
thehandsofprivatecapitalandsofinallyoptedforcontractingInnocenti,an
Italiancompanytoconstructaplantwithacapacityof600000tons/yearfour
timeslargerthantheSyndicate'splan.
36
Industrializationunderdemocracy
After the overthrow of Perez Jimenez the newly-elected AD administration
underBetancourtadoptedimport-substitutingindustrialization(ISI)asanoffi-
cialpolicy.Protectionistmeasuresthatbeforewereprovidedfordomesticfirms
on a case-by-case basis now became the rule, and credits were easily available
on generous terms. The CVF was given a renewed mission to support the
strategyofimportsubstitution.Thecommonproceduretogetaccesstoofficial
support was the presentation of a project to substitute imports. Then the
approvaloftheMinistryofDevelopmentwasenoughtoraiseatraderestriction
onthoseproductstobesubstitutedandtobegrantedasoftcreditbytheCVF.
Manufacturing continued its expansion. According to an official survey of
the572manufacturingestablishmentsinoperationwith50ormoreemployees
in1966,39970percentwerefoundedbetween1950and1966and29351
per cent during the period 195060.
37
In terms of the location of industry
therewasanimportantdevelopmentwiththesurgeofValenciaMaracayasthe
mainindustrialbelt,partlyduetothefactthatitwasthemaindestinationarea
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 217
for direct foreign investment. By 1968, of the 170 medium and large firms
installed in Valencia, foreign capital participated in 43 of them, representing
nearly70percentofthecapitalofthewhole(Rangel,1971,p.149).
38
The 1960s witnessed a clear movement towards capital-intensive industries
with important developments in basic metal industries, particularly with the
beginningofoperationofSIDOR
39
in1962,andtheincreaseintheactivityof
themetal-transformingsectorledbytheautomobilesector.Theeconomicrole
ofthestatekeptgrowingparticularlyintheareasofplanningandproduction.
In 1960 the Corporacion Venezolana de Petroleo (CVP) was established with the
aim of initiating Venezuelan direct involvement in all the stages of the oil
industry.
40
Also in the same year the Corporacion Venezolana de Guayana
(CVG)wascreatedinordertodevelopthenaturalresourcesandhydroelectric
potentialofthisregioninthesouth-eastofthecountry.
41
Theinward-lookingstrategyproducedhighandsustainedratesofgrowthof,
onaverage,6 percentperyearduringtheperiod196072.Regardingtheresults
obtained in terms of the reduction of import dependency in the country, the
import coefficient
42
for manufacturing (excluding oil refining) experienced a
significantreductionbetween1950and1968fromavalueof46.5percentto
25.5 per cent. The substitution took place mostly in light industries and to a
lesser extent in the intermediate sector. However during the 1970s and 1980s
thepaceofreductionofimportscametoahaltinmostindustries,andthereis
evenevidencepointingtoincreases intheimportcoefficientoffinaldemand
(32percentin1978)andintheshareoftheimportcomponentofinputs36
percentin1978comparedto31.6 percentin1970.
43
The metal-transforming and transport sectors were at the centre of the
strategy of industrial diversification. The automobile industry originated in
Venezuela in 1948 with the installation of the first foreign assembly plant
and developed slowly until 1962 when the government, for the first time,
formulated a conscious industrial policy aimed at regulating the assembly
activity and encouraging the development of a national auto-parts industry.
However, this was one of the sectors that despite several plans and targets
regarding the incorporation of locally-produced components and exports,
had little success in achieving either import substitution or exports. By the
end of the 1980s, on average, 75 per cent of each vehicle produced was still
importedandtheamountofexportsmadebythesectorwasinsignificant.
Thenewoilboomandthestateentrepreneur
The oil price booms of 1973 and the beginning of the 1980s provided the
governments of Carlos Andres Perez (197478) and Luis Herrera (197983) of
ADandCOPEI
44
respectivelywithmassiveresourceswhichwerepartlydirected
towardsindustrialdevelopmentaswellastowardsthecreationofphysicaland
human capital. As part of a strategy of export diversification based on the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
increase in the productive capacity of energy-intensive industries, the public
sector made important investments in aluminium, steel, energy generation,
andalsointhepetrochemicalarea.Thebasicmetalindustriessawtheirshareof
total capital increase from 23 per cent in 1971 to 35 per cent in 1981, yet
without generating much employment and with output still waiting to reach
itspotential(seeTable8.1).
Theparticipationofpublicinvestmentintotalinvestmentwentupfrom24
per cent in 1970 (Bs.12.3 billions of Bs. of 1984) to 67 per cent in 1982 (42.9
billions of Bs. of 1984), and of this last figure 81 per cent was channelled
through non-financial state enterprises.
45
The nationalization of the iron
industry in 1975 and the oil industry in 1976 dramatically increased the
participation of the state in the economy. The public sector not only became
the main producer in the economy but also the principal source of employ-
ment.By1980,outofatotallabourforceof4.5millionapproximately20per
cent were employed by the public sector.
46
Within the manufacturing sector,
withashareof18percentofthelabourforceinthatyear,thoseemployedin
non-financial state enterprises represented approximately 20 per cent of the
totalemployedinmanufacturing.
Another important policy decision concerning industry occurred in 1973
when the country joined the Andean Pact adopting its Decision 24 a policy
thatrestrictedforeigninvestmentandthepolicyofindustrial programming
bysector.Accordingtothispolicy,governmentplannersofthedifferentmem-
ber countries negotiated and eventually agreed which industries would be
allocated to each of the countries. The consequences of these two policies
have been criticized (Frances and Na m, 1994) as being a factor that retarded
theindustrialexpansionofthecountryduringthe1970sandearly1980s.This
period was also very prolific in the creation of institutions. Several public
financial entities were established by the Perez administration during 1974
and1975inordertocomplementtheCVFinpromotingindustry:Corpoindus-
tria, for small and medium-sized firms, the Fondo de Credito Agropecuario, for
agriculture, and the Fondo de Credito Industrial for industry. However, at this
pointthemultiplicationoforganizationsinvolvedwasaccompaniedbyalack
ofpropercoordinationamongthem,underminingtheoverallefficiencyofthe
industrialpolicy(BitarandMej as,1984).
Thiswasjust oneof themanifestations ofageneral phenomenonof astate
not able to administer efficiently and with good judgement the vast
resources that were accruing to it. This unexpected windfall also became a
powerful magnet to corruption and fraudulent schemes designed to finance
politicians, union leaders and businessmen. At that point getting access to
credits was more related to having the right political connections than to
productive ability or the quality of the projects under consideration. The
succession of scandals and the realization of the exhaustion of the ISI finally
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 219
ledtotheinvestigationoftheCorporationin1983andtoitseventualliquida-
tionin1990.
As a result of this ambitious programme of investments, in 1978 the eco-
nomywasatfullcapacityanddisplayingclearsignsofoverheating.Theman-
ufacturingsectorexperiencedadeclineinactivitydue,toagreatextent,tothe
limited possibilities of substituting imports and to the appreciation of the
bolivar.Duringtheperiod197781itgrewatanannualrateof3 percent,far
below the figure of 8 per cent recorded during 196076.
47
In 1979 the new
administrationofHerreratookmeasurestocooltheeconomyandimplement-
edatimidprogrammeoftradeliberalization.Thecontractionaryeffectofcuts
in expenditure was greatly amplified by a drastic drop in total private invest-
mentwhosevaluein1982washalfthatobtainedin1979,andaquarterofthe
investmentinmachineryandequipment.
48
Anexpansionarypolicystartedin
1982 in an external context that soon turned unfavourable due to the debt
crisisanditsaftermathandaworseningoftheoilmarket.
49
Theconfluenceofthesefactorsledtoadeteriorationinthefiscalandcurrent
accountswhichunderminedthecredibilityofsustainingtheexchangerate.In
February 1983 despite the desperate measures to stabilize the situation up to
the coming election and after a loss of more than US$12 billions in interna-
tional reserves in the preceding 14 months, the government reluctantly
decided to abandon the fixed exchange rate system and adopt multiple rates.
Asaresultthebolivarwasdevaluedby30percentafternearly20yearsoffree
convertibilityandfixedparitywiththeUSdollar.Fiscalpolicyturnedcontrac-
tionaryandtheforeignconstraintstartedtobite.
The deterioration in external circumstances and the difficulties of meeting
thedemandsoftheISImodelsettheconditionsfortheintroductionofpolicy
changes aimed at improving the efficiency of production and promoting
export activities. The 1983 devaluation and a policy of direct subsidies to
non-traditional exports were two measures that generated incentives in this
direction. However, the level of protection remained without major changes,
anditwasonlyattheendofthe1980sthataprogrammeoftradeliberalization
tookplace.
Someaspectsofthelongterm
Theaimofthisfinalpartistostudysomeaspectsofthelongtermandtoplace
the development of industry in a more general context. We begin with a
discussionaboutthelinkbetweengrowthinmanufacturingandtheperform-
anceoftheoilsector.Afterpointingouthowoilincomedeterminesthelevelof
industrialactivityintheshortrun,attentionisdirectedtothechangesinthe
structure of production in the country. Next we look at the record of labour
productivity in manufacturing and other sectors of the economy. The last
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
section is devoted to the presentation of secular trends in the development
ofinfrastructureandhumancapitalanditscontributiontothedevelopmentof
industry.
Theperformanceofmanufacturinganditslinkwithoil
In Venezuela oil and industry have a close, but difficult, relationship. As we
alreadymentionedduringthehistoricalsummary,thereweresignificanttrade-
offsbetweenthedevelopmentofoilandtheprospectsofindustrialization.So
guaranteeingmarketsforoilexportswasagoodargumentforsigningthe1939
Treaty with the USA, but the reduction in tariffs on US imports was not good
newsforlocalindustrialists.Also,theexchangerateprovidedthegovernment
with an instrument to extract part of the rents from the oil companies but at
the expense of the tradable sector. Another policy dilemma that oil brings
about is the tendency towards appreciation of the real exchange rate and its
effectsonthelevelofactivityofthenon-oiltradablesector.Ineconomieswith
a low level of development, a strong currency hinders the potential develop-
ment of activities in this sector (inhibiting effect), whereas in an already
industrializedeconomyitworksagainstthecompetitivenessofthoseindustries
producingtradablegoods(de-industrializationeffect).
`Sowingtheoil'wasawaytoreachabalance,topullbothsectorsinthesame
direction with expenditure as the mechanism of transmission. However, it
createdasituationwhereindustrywashighlydependentupon,andvulnerable
to,oilshocks.Thebasicproblemisthatsincetheimportcontentofproduction
ishigh,thenon-oilindustrialoutputdoesnotcreateitsowndemandandatthe
sametimetheindustryreliesonthestateoilcompanyPDVSAortheforeign
companiesbeforenationalizationfortheprovisionoftheforeignexchangeto
finance import needs. The cycle is completed through the public spending of
oil revenues that increases internal demand enabling the ISI industries to sell
their products profitably. As a consequence of the fact that the economy
functions in this way, changes in the price of oil explain much of the short-
termfluctuationsinindustrialactivity.Inwhatfollowsthefocuswillbeplaced
ontheseculartrendofindustrialoutputandwillcontinuethediscussionabout
theinhibitingeffectoilmighthaveonthetradablesector.
Structuralchanges:hasdiversificationbeenachieved?
We will use the model of the `Dutch disease'to address the question of struc-
turalchange.
50
Thismodelpredictsthatapositiveexternalshockhastwomain
results:first,iftheextraincomeisspentinthelocaleconomythepriceofnon-
tradablegoodsrelativetotradablegoodswillrise.Thischangeinrelativeprices
is called the appreciation of the real exchange rate. By definition, in the new
equilibrium domestic welfare rises but at the expense of a reallocation of
production out of the tradable sector into the non-tradable sector. Second,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 221
the marginal product of labour rises in the booming sector as well as the
demandforlabourwhichinducesamovementoflabouroutoftherestofthe
economy a resource-movement effect.
51
Regarding negative shocks, it is
expected that the reduction in income and consumption will eventually lead
to excess supply of both tradable and non-tradable goods producing a real
depreciationfollowedbyanexpansionofthetradablesectorandacontraction
ofthenon-tradablesector.
Beinginterestedmostlyinmanufacturingafter1936 wewillconcentrateon
the consequences of the spending effect that is, appreciation of the real
exchange rate and changes in the structure of production which are the rele-
vantfeaturesoncetheboomingindustryisconsolidatedandstopsdemanding
production input from other sectors of the economy. Nevertheless, the
resource-movement effect was important during the installation of the oil
industry in the country during the 1920s. In what follows we discuss the
changesinthestructureofgrowthaswellastheevolutionofpricesfortradable
and non-tradable goods. Table 8.2 presents information about changes in the
compositionofthenon-miningGDPduring193689accordingtotheperiods
adopted in the historical account of the first part. It includes the annual
averageratesofgrowthofeachsectorandtotalGDP,aswellasthesectorshares
innon-miningGDP.
As can be observed in Table 8.2, during 1948 to 1958 the basic activities
producinggoodsforthehomemarketagricultureandnon-oilmanufacturing
accounted for only 16 per cent of non-mineral GDP. Agriculture which had
alreadyfelttheimpactoftheDepressiononitsexportssawitssharedropping
from an already low level of 13 per cent to 7 per cent.
52
The tendency in
Table8.2 Venezuela:GDPbyeconomicsectors;annualaveragegrowthandshares
193
growth
647
share
194
growth
858
share
195
growth
973
share
197
growth
483
share
198
growth
489
share
Non-min.GDP
Agriculture 0.6 13.1 5.5 7.3 5.7 7.9 2.8 6.1 3.0 7.3
Manufacture 6.7 7.8 11.4 8.2 7.5 13.4 5.1 15.0 1.8 18.6
Construction 14.3 16.6 7.7 21.2 4.9 14.3 0.1 16.4 7.5 7.6
Services 10.8 62.4 6.5 62.9 5.1 62.8 3.5 59.7 2.4 62.6
Mining 11.7 23.1 6.8 25.5 3.7 23.0 2.0 13.4 4.7 12.4
TotalGDP 9.6 100.0 6.9 100.0 4.9 100.0 2.3 100.0 1.8 100.0
Note: Growthannualaveragegrowthrate;shareannualaverageshare.Thefiguresfor
Manufactureexcluderefining.Miningalsoincludesoilrefining.Servicesistheresidualthatresults
aftersubtractingagriculture,manufacture,mining,constructionandelectricityandwaterfromtotal
GDP.Allcalculationsarebasedonconstantfiguresat1968prices.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
manufacturing was in the opposite direction showing an improvement in its
economic importance with respect to the previous period 8.21 per cent
compared to 7.7 percent and with the highest rate of growth (11 percent).
During197483itsaverageshareofnon-miningGDPwentupto15percent,
more than doublingthe levelreached in the1940s, and thesector grewatan
annual average rate of 5 percent, lower than before but still higher than that
shownbytotalGDP2.31percent.Venezuela,byprotectingthemanufactur-
ingsector,wasabletoavoidthemainimpactofthediseaseonmanufacturing,
and,morethanthat,todevelopitsindustrydespitethespendingeffectandan
overvalued currency.
53
The evidence also points at a high share of services
which is a common feature of mineral-exporter economies. This is a con-
sequence of the way in which the rent from the oil sector flows through the
economy passing first to the government and then out via public works and
publicservices,creatingdemandforimportsandotherservicesbeforereaching
thesecondarysector.
Movementsinrelativeprices
We now turn to the study of the behaviour of relative prices, and begin by
consideringtheevolutionoftherealexchangerate(RER)definedasthenom-
inal exchange rate adjusted by the differences between external and internal
prices.
54
This is presented in Figure 8.2. The RER shows a downward trend
starting in the 1920s up to the Second World War which is consistent with
the presence of the revaluationist policy followed by the government, either
passiveoractive,combinedwiththespendingandresource-movementeffects
broughtaboutbyoilproduction.Thenaperiodofamorestableexchangerate
follows coming to an end with the devaluation of 1961 in response to a
significantdropinoilrevenuesafterthereopeningoftheSuezCanal.Despite
beingaperiodwithvastexpenditureoninfrastructure,thepressureoninternal
priceswaspartlyalleviatedbyagenerousamountofimportsandanexpanding
supplyoflabourduetothearrivaloflargenumbersofimmigrants.In1973the
RERmovedupwardsmostlyduetotheincreaseinexternalinflationfollowing
theoilshockofthatyear.
By1976 thistrendwasreversedandarapidappreciationstarted,lastingupto
1983.ThisdownwardmovementintheRERwastoagreatextenttheresultof
the internal spending of the 1974 and 1980 oil-boom windfalls and a tight
situationinthelabourmarket.After1983thereisagainachangeindirection
and the rest of the decade is dominated by a RER depreciation. As can be
observed in Figure 8.2, during much of the period under analysis inflation in
thecountrywaslowerthanthatabroad.Theavailabilityofoilrentsplayedan
important role in two respects: first it provided a generous currency flow to
maintainastablefixedexchangeratethatworkedasastrongnominalanchor,
and second it supplied the resources to pay for widespread subsidies in the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 223
10
100
1000
I
n
d
e
x

1
9
6
8

=

1
0
0

RER
1
9
2
0
1
9
2
4
1
9
2
8
1
9
3
2
1
9
3
6
1
9
4
0
1
9
4
4
1
9
4
8
1
9
5
2
1
9
5
6
1
9
6
0
1
9
6
4
1
9
6
8
1
9
7
2
1
9
7
6
1
9
8
0
1
9
8
4
1
9
8
8

Years
Figure8.2 Venezuela:realexchangerate,192086
production sector alleviating pressures that otherwise would have led to an
increaseinprices.
55
Anassessmentoflabourproductivity
Estimates for labour productivity in manufacturing, agriculture, mining and
theeconomyasawholearepresentedinFigure 8.3.Wearemeasuringlabour
productivityinanygivensectorsimplyastheratiobetweenitsoutputandthe
economicallyactivepopulationinthatsector.Therefore,becausethefiguresdo
notexcludetheunemployedtheinformationgivenisonlyanapproximation
of the true value of labour productivity. Thus, the series presented should be
taken only as an indicator of changes in trend in the long term and not as
reliablefiguresforproductivityinoneparticularyear.
56
SomepatternscanbeobservedinthevariablesincludedinFigure8.3.Agri-
culturalproductivitywhencomparedwithothersectorsshowsasteadierprocess
ofgrowthatthecloseofthe1970s.Thebehaviourofthisvariableisfasterpaced
duemainlytoasignificantmovementoflabouroutofthesectorandanincrease
inthemechanizationofactivities.Aswouldbeexpected,theminingsectorwith
its natural high capital intensity shows the highest level of productivity and,
with the exceptions of the interruptions caused by the Depression and the
SecondWorldWar,itsperformanceisquitestableduringthewholeperiod.
Manufacturing exhibits an upward trend in its productivity up to the mid-
1970s when it becomes stagnant.
57
According to the evolution of its output
andlabour, theabsence of growth in outputappears as themain cause ofthe
slowdown in productivity during the recent period. The high rates of
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
1 000
10 000
100 000
1 0000 00
Total
Agriculture
Mining
Manufacturing
1925 1935 1945 1955 1965 1975 1985
Figure8.3 Venezuela:labourproductivitybysectors,192590
Semi-logscale.Unitsinthousandsofbolivarsat1968prices
Sources: Baptista (1997) for output. For the economically active population: ECLAC (1966) for
estimates in 1925, 1930, 1935 and 1945; Bairoch et al. (1968) for figures in 1941, 1950 and 1961
(based on population censuses); ILO for 1971 and 1981 (based on censuses) and 1992 (based on
household surveys). Prior to 1941 employment in the mining sector is equated to that of the oil
industry.Thefiguresforthelabourforceintheoilsectorfrom1922to1935arefromAranda(1990),
otherwiseBaptista(1994).
productivity experienced during the 1950s are consistent with significant
developments in infrastructure and the modernization of manufacturing that
tookplaceduringtheperiod.Thisisfollowedbyaperiodwhichcoincideswith
the official implementation of the ISI where productivity grew at a more
moderatepace.Figure8.3showsthatbythemid-1970stherewasaslowdown
in productivity growth. This result has been attributed to the confluence of a
favourable exchange rate and a relatively high income in the context of a
highly protected and small internal market. This resulted in the adoption of
capital-intensivetechniqueswhereeconomiesofscalearelarge;
58
andthereis
underutilization of capacity, and overdiversification in the type of goods pro-
duced (Hausmann and Marquez, 1983).
59
In addition, studies of the business
sector point to the adoption of strategies of vertical integration and rapid
diversification inresponsetouncertaintyinsupplyqualityandopportunity
and the incentives of ISI (Frances and Naim, 1992). These factors tend to
producepooradvantagesfordivisionoflabourandspecializationandincon-
sequencehaveanegativeeffectonproductivity.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 225
Insupportofindustrialization
This final section focuses on three aspects concomitant with industrialization
and the productivity record. The first has to do with the development of the
infrastructure particularly investment aimed at improving communications
and expanding the supply of energy. The second deals with human capital
andtheimprovementofthequalityofthelabourforcethroughinvestmentin
educationandtheaccelerationofaccumulationthroughimmigration.Afinal
aspect is the creation of a market for manufactured goods which is usually
relatedtotheprocessofurbanization.Figure8.4providessomeindicatorsreflect-
ingtheevolutionofthesethreeaspectswhichsupporttheprocessofindustriali-
zation.
Thefirstplotgivessomeinformationonthedevelopmentofinfrastructurein
thecountrybylookingatthecapacityofthecountrytogenerateelectricity.
60
The output of electricity per capita experienced an impressive increase of the
orderof30timesbetween1940and1988,fromlessthan100kilowatthoursper
persontomorethan3000in1988.IntheLatinAmericancontext,duringthe
same period, the country went from being one of the lowest producers of
electricity at the end of the 1930s in relation to the size of its population
tooneofthehighestattheendofthe1980s.
.05
.1
.15 Primary enrollment p-c
1
2
3
Output of electric energy p-c
1920 1940 1960 1980 2000 1920 1940 1960 1980 2000
300
200
100
0
10
20
Inmigration
Oil rent p-c Expenditure on education p-c
1920 1940 1960 1980 2000 1920 1940 1960 1980 2000
Figure8.4 Venezuela:insupportofindustrialization
Note: Outputofelectricityisexpressedingigawatthourspercapita(p-c).Themeasureofeducation
is obtained by dividing the enrolment in primary education by total population. The units of
expenditure in education are Bs. of 1968 spent per person. Due to a change in the fiscal year the
valuefor1962correspondsonlytoJul.Dec.whichexplainsthedropobservedinthisyear.Expendi-
ture figures are obtained after deflating current spending by the CPI with1968 as the base year and
thendividingbytotalpopulation.Thefiguresofoilrentareinrealtermsdividedbytotalpopulation
inmillions.Migrationfiguresareobtainedbydividingnetmovementsbypopulation.
Sources: Figures on electricity output and enrolment are from Mitchell (1993); expenditure in
education are from Kornblith and Maingo n (1985) up to 1979 and IMF thereafter; and those for
migration,oilrent,andtotalpopulationfromBaptista(1997).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Asecondarea where thestate has consistentlyconcentrated alarge propor-
tionofitseffortsandresourcesduringthecenturyisthatofpublicworks,and
in particular the improvement of communications. This was crucial at the
beginning of the century to change the communication structure inherited
from the colonial period which responded to the needs of an agro-export
economy.Thehigherdegreeofintegrationoftheterritoryallowedthegovern-
ment to increase its military and political control, and at the same time
brought about an enlarged internal market. For instance, as early as 1910
there was legislation that committed the government to allocate 50 per cent
ofthebudgettoroads.Aspartofthiseffortoflinkingthecountry,in1924the
Trasandina was inaugurated, a road of more than 1000km that linked the
Andean region with Caracas. A journey that used to take months could now
bemadeinweeks.
61
With the spread of lighting and public works also came the expansion of
enlightenment, and education underwent a major transformation in the se-
condhalfofthecentury.AsshowninthesecondplotofFigure8.4,duringthis
periodtherewasarapidaccelerationofschoolenrolmentinprimaryeducation
relative to population, which reached stability after the 1960s.
62
Universities
mushroomedduringthe1960sand1970sandby1975thenumberofstudents
registeredinhighereducationwasaboutninetimesthefigureof1960.Itrose
from 26000 students in 1960 (1.8 per cent of total enrolment) to 222000 in
1975(6.8percentoftotalenrolment).
Thisexplosiveincreaseinenrolmentwasmatchedbyasignificantchangein
the resources the central government directed to education third plot in
Figure 8.4 both in absolute and relative figures. The average share of educa-
tion from total expenditure for the period 194058 was of the order of 6 per
cent, which contrasts with an average of 13 per cent during the next two
decades. Funding for higher education rose from 3.8 per cent of the total
education budget in 1946 to 42 per cent in 1978 (Kornblith and Maingo n,
1985).Inabsolutefiguresthechangeisalsoimpressive.Measuredinbolivarsat
1968 prices, the government in 1940 spent the equivalent of Bs.14 per capita
on education out of a total expenditure of Bs.209 per capita. By 1950 this
figurehadmorethandoubledtoBs.34,andthenmorethandoubledagainina
decade toreachBs.83 in 1960for atotalexpenditure ofBs.1054 perperson.
Theupwardtrendcontinueduptothebeginningofthe1980swheneducation
expenditurereachedapeakofBs.286 percapitain1981andBs.2032intotal.
Thereafteritexperiencedarapiddeclineandby1986 itwentdowntoBs.196,a
figuresimilartothatof1974.
A second source for the transformation of human capital in the country
was external migration. As canbe observed fromthe fourth plot ofFigure 8.4
the net flow of immigrants shows a strong correlation with the oil rent in
per capita terms. Thus another contribution of oil to the development of the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 227
country came in the form of providing a `population multiplier' and a rapid
increaseintheaccumulationofhumancapital.Theplotalsoshowstwobursts
of international immigration, the first occurred during the 1950s, mostly the
resultofEuropeanimmigration,andasecondone,ofaminorscalerelativetothe
sizeoftotalpopulation,duringthe1970sdominatedbyimmigrantsfromother
countriesoftheregion,particularlyfromColombia.
63
Theoldproblemofattract-
ingimmigrationceasedtobealimittodevelopmentinashortperiodoftime,yet
itwastodevelopattheendofthe1980sintoaproblemdifficulttolimit.
64
Finally,anotherphenomenonthatdeveloped,closelylinkedtothecreation
of industry, was the move to the cities. In 1935 only 35 per cent of the
population was located in towns of more then 2500 inhabitants; in 1980, 75
percentofthepopulationwasdefinedasurban(Baptista,1997).Theurbaniza-
tionprocesshasimportantlinkswithindustrialization.Forinstance,fromthe
demand side urban population provides the market for industrial goods, and
fromthesupplysideitisasourceofdemandforjobopportunitieswhichcalls
for policies to develop manufacturing.
65
In Venezuela up to the 1960s, the
urbanization process was concentrated in the growth of the populations of
Maracaibo and Caracas, cities where oil revenues are produced and spent
respectively. The more recent figures, although still showing great concentra-
tionofthepopulationinfewcities,pointtotheriseofalternativeurbanareas
such as Valencia, Maracay and Barquisimeto which are more linked to recent
developmentsofindustry.
66
Conclusions
IntheLatinAmericancontexttheVenezuelanexperienceisinterestingbecause
the abundance of foreign exchange, which would be expected to be a clear
positivefactor,inpracticegeneratedmechanismsthateventuallyhinderedthe
rationalityoftheprocessofindustrialization.First,itcreatedpressurestowards
an overvalued currency and the need to protect the whole tradable sector.
Second, itcreatedastrong politicaleconomylinked tothetransferofoilrent
to the private sector which was at the root of distortions and inefficiencies.
Finally,duetothecloseconnectionbetweenoilrevenue,thestateandbusiness
opportunitiesafter1936,theindustrythatgrewafter1936wascharacterizedby
a high degree of structural dependency on the level of external revenues and
theamountofpublicspending.
Thereisanotheraspectofthecombinationofoilandabundancethatemerges
fromthisexperience.Oilcomesfirst.Thefightfortheoilrentisazero-sumgame
whereboththestateandtheprivatesectorwanttomaximizeprofitsandrents
respectively.Sothereceiveroftherentisagainsttheproducerofit,andbothare
engaged in a relationship that lacks the cooperative spirit of development
basicallyanon-zero-sumgame.Underthelogicofbargaining,thepublicsector
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
hasanadvantageifitfacesforeigncapitalistswithoutinternalpoliticalsupport.
Therefore, there is a strong argument for excluding local capitalists from the
rent-generatingsector.TheVenezuelancaseisconsistentwiththislogic.
Now we consider abundance. The wealth obtained in this way generated
strong governments which took exclusive responsibility for some productive
activities, at the same time as it promoted development in others with the
participationoftheprivatesector.Butthemagnitudeoftheresourcesathand
andthecomplexityoftheproblemsfacedsurpassedthecapacityofthepublic
sector to manage the process of development. Eventually, the result was dis-
appointinggiventheexpectationscreatedbythesuddenwealth.
Looking on the bright side, without denying the current weaknesses and
limitations of the level of development achieved, the country has succeeded
in a short period of time in building a valuable infrastructure and, to a great
extent,hasovercomedeficienciesinhumancapitalformation.Togainanidea
ofthemagnitudeofthetransformationVenezuelahasexperienced,letusbegin
byrepeatingsomeindicatorsoftheconditionsprevailingin1936:75percent
of the population was living in rural areas, life expectancy was only 39 years,
and non-refining manufacturing accounted for just 6.7 per cent of non-oil
GDP. Fifty years later we find a country with a life expectancy of 70 years,
withalowrateofilliteracy,andwheretheimportanceofnon-refiningmanu-
facturingonnon-oilGDPisoftheorderof18percent.Besides,themorerecent
experienceoffersevidenceofamodernmanufacturingsectorwhereasignific-
ant group of firms are beginning to show theircapacity to compete in export
markets.
Finally,wewouldventuretoofferaninterpretationofthewayIthinksport
and industrialization are related. What we say is based on a very simple, and
perhapsinaccurate,versionofthesportinghistoryofthecountry,butperhaps
itwillbeenoughtomakemypoint.Forthesakeofargument,andthehistory
ofindustrialization,letuslocatetheoriginsofbaseballwiththearrivalofthe
oilcompaniesandthedevelopmentoftheproductioncapacityinoil.Baseball
teamsgrewgraduallyandnativeplayersdevelopedtheirskillsandassimilated
the technology of the sport. At some point there was an opportunity to send
playerstobetrainedintheUSAanditwasalsopossibletointegrateimported
playersintothelocalteams,wherecompetitiveVenezuelanplayerswereinthe
majority. The result: VenezuelaBrazil (100). What about football? Although
its initial introduction into the country may be attributed to the Jesuits, its
development as a mass sport is associated with the migration after the 1940s.
Thestageofexpansionoftheclubscoincidedwiththeincreaseintheoilrents,
sointhesamewaythatitwascheaptoimportgoods,sotherewerefewlimits
to importing players from abroad. Foreigners became the majority, and little
room was left for the development of the skills of local players. The result
BrazilVenezuela(100).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 229
Notes
1 What is expected is that deliberate planned ISI and that motivated by foreign
restrictions bring about bias in opposite directions. As Hirschman (1968, p. 91)
puts it `...of the four motive forces behind ISI balance of payment difficulties,
wars, gradualgrowthof income, anddeliberatedevelopment policy only the first
leadstoabiasinfavourofnonessentialindustries[thetendencytogiveimportance
towhatisunimportant]...deliberateddevelopmentislikelytoproduceexactlythe
oppositebias;andtheremainingtwoareneutralwithrespecttotheluxurycharacter
oftheindustry'.
2 During18801913,VenezuelawassecondonlytoBrazilasaworldexporterofcoffee.
To a great extent the surplus was not invested in manufacturing or other develop-
ment activities. Most of the coffee earnings went towards financing the import of
consumergoods,largelyofaluxurykind,bythelandlords.SeeRangel(1969).
3 Also knownas the purchasing power of exports. It results from dividing the export
valuebytheimportpriceindex.
4 Foramoreextensivetreatmentofthehistoryofindustrializationinthecountrysee
the following: Araujo (1964, 1969), Bitar and Troncoso (1983), Betancourt (1967),
ECLA (1966a), Co rdova (1963), Sonntag and De la Cruz (1982), Jongkind (1981),
Karlsson(1975),Purroy(1982),Rangel(1958,1968,1971),MazaZabala(1964,1976).
ForaviewoftheeconomichistoryinthiscenturyseeAranda(1990),FigueroaBrito
(1974)andGonzalezDeluca(1994).
5 Thefollowingimpressionofacontemporaryoffersanillustrationofdifficultiesfaced
during the initial steps: `Development since that time [1912] has been greatly
retarded, principally by the lack of transportation facilities. In the solution of the
problemofgettingmachinerytothewellsiteandofgettingtheoiltomarketliesthe
keytotheoilindustryinVenezuela.Therearenoroads,oreventracks,inthoseparts
ofthecountrywhereoilisfound'(Bell,1922,p.94).
6 Forexample,thewaterandcoastalareasoftheMaracaibolakewerepollutedandthe
exploration activities led to conflicts with indigenous groups that resulted in the
exterminationofsomeofthem(Sullivan,1976,p.259).
7 Bell also wrote a monograph on Colombia a year later. These works offer a unique
opportunity to make a comparative study of the economic development of both
countriesatthestartofthetwentiethcentury.
8 Inthelate1930sVenezuelawastheworld'sleadinglow-costoilproducerproviding
one-third of total net oil exports. Yet the country was the only major oil producer
thatlackedadomesticexportrefiningindustry(Karlsson,1975).
9 What was at stake? According to Bell àbout US$2000000 has been spent on this
refinery,Americanskilledlabourbeingemployedinitserectionandpracticallyallof
themachineryandmaterialsusedinconstructioncomingfromtheUS.About1100
nativelabourershavebeensteadilyemployedsince1914ontheproject'(Bell,1922,
p.446).Tohaveanideaofwhatthesefiguresrepresentedinlocalterms,thevalueof
the investment was equivalent to the capital invested in the cotton industry, the
largest in the country at the time. Regarding employment, the oil industry in
Venezuelausedatotalof3463personsin1922(Aranda,1990,p.88).
10 When the decision to build the refinery in Curacao was taken in 1916, although
General Go mez had already consolidated his regime and given political stability
to the country, the past history of internal turmoil and trouble with foreign
investorswas still fresh in people's minds. For instance, as recently as 1902, during
the government of Cipriano Castro, German and British battleships blockaded
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the Venezuelan coasts as an action to demand the service of the country's external
debt.
11 According to Vallenilla (1973, p. 38), Go mez did not want refineries in Venezuela
becauseoffearthattheoilregionwouldbecomepoliticallytoopowerful,fuellingthe
aspirationsforindependenceofsomegroupsintheZuliastate,themostimportant
oilregion.Atthesametime,theinstallationoftherefineriesmeantthecreationofa
centrewithahighconcentrationofworkerswhichcouldhavecreatedproblemsfor
thegovernment.OnthisissueseealsoQuiroz(1987).
12 ThesourceforthefiguresonelectricityconsumptionisWilkie(1974).
13 ThefollowingfragmentofGo mez'speechbeforeCongresson19April1929reflects
this predominant perception: Òil production rose to 107235054 barrels in 1928,
placingVenezuelainsecondplaceintherankingofoilproducts.Certain,oilisbeing
exportedbyforeigncapital:ithadtobethatwayorelsewewouldresignourselvesto
leavefor deadthatwealthin the bowelsofthe earth.However, ourgeneralwell-being
depends on agriculture. In order to promote it I proposed in my speech of 1928 the
creation of the Banco Agr cola y Pecuario' (Memoria del Presidente, 1929, author's
translation,emphasisadded).
14 TheeffectsofthecrisesandtheregionaldifferencesaredescribedinGonzalezDeluca
(1994,pp.21898).SeealsoMcBeth(1983)fortheimpactontheoilindustryandthe
coffeeareas.
15 The crisis in the textile industry is documented in Gonzalez Deluca (1994) and
Karlsson(1975).
16 See Bulmer-Thomas (1994, table 7.2, p. 197) for a comparative evaluation of the
impact of the Depression and evidence for terms of trade and purchasing power of
exportsofseveralLatinAmericancountries.
17 Therewasaimportantdebatein1934betweenAlbertoAdriani,insupportofdeva-
luation,andVicenteLecuna,thepresidentofthemainbank(BancodeVenezuela),
whowasinfavourofastrongbolivar.SeeRangel(1968,pp.20727)andBaptistaand
Mommer(1987).Laterin1937therewasanofficialrevaluationofthebolivar.Before
thefiscalreformof1942andthe1943HydrocarbonsLawcurrencyrevaluationwas
one of the main mechanisms the government had to extract rents from the oil
industry. Because the companies paid salaries and for local purchases in local cur-
rency, a revaluation generated an increase in the number of dollars to be spent to
acquirethebolivarsneededtomakepaymentsinthecountry.SeeMayobre(1944).
18 Another aspect in which Venezuela did not follow the general trend is that J. V.
Go mezwasoneofthefewheadsofgovernmenttosurvivetheDepression.
19 In1936 theExecutiveendorsedupto50percentoftheloansgrantedto industrial
ventures by private commercial banks and other financial institutions. See Barrios
(1996,p.65).
20 SeeDavila(1988)forastudyofthisperiodofinstitutionalbuilding.
21 This treaty was part of a wave of reciprocal trade agreements signed by most Latin
AmericancountrieswiththeUSAduringthe1930s,followingafundamentalshiftin
UStradepolicyaftertheGreatDepression.SeetheUSTariffCommission(1945)for
thetradeagreementswiththeUSA,andECLA(1956)forananalysisoftradepolicies
duringtheperiodintheregion.
22 The following quote stressed the perception of external vulnerability at the
beginning of the 1930s: `The development of the Venezuelan oil industry, and the
experience of the 1932 US tariff on foreign oil revealed to the Venezuelan Govern-
ment its vulnerability and lack of real control of the industry, and its depend-
ence on the continual growth of production, which in turn depended on the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 231
requirementsoftheoilmarketoftheindustrializedUSandEurope'(McBeth,1983,p.
205).
23 The negotiation of the treaty and its implications is well-documented in Machado
andPadro n(1987).
24 TradefiguresaretakenfromMinisteriodeFomento(1948).
25 TheoriginsofthemaineconomicgroupsinVenezuelauptothe1960saredescribed
inRangel(1971).
26 TheopinionofMrR.Aristiguieta,afutureCVFdirector,isillustrativeofthemoodof
thattime(1938):`Whatweneedistoproducebyourselvesthegreaterpartofthose
articles that we now import and whose purchase is ruining us in an accelerated
form...At the presenttime, Latin America manufactures the majorityofthe goods
itconsumes.WhenisVenezuelagoingtojointhismovement?Whenweunderstand
the problem and have a definite orientation for our industrialization' (quoted in
Barrios,1996,p.60).
27 It had its origin in the Junta para el Desarrollo de la Producion Nacional, a non-
profit-making agency established by the government in 1944 to directly provide
soft credits to agricultural, cattle-raising and industrial activities. See Machado and
Arenas (1995) on the creation of the CVF, and Barrios (1996) for an account of the
roleoftheCVFthroughoutits37yearsofactivelife(194683),inparticularaboutits
participationinthepromotionofthesugar,cementandelectricitysectors.
28 OfallfourprojectsonlythefooddistributioncompanyCADAsucceeded.Therestof
them were an economic failure, partly because they were too idealistic in their
conception and not well-conceived in terms of the realities of the country for
example, in the project for the development of agriculture it was assumed that
there would be support for research in tropical crops; however, they did succeed in
introducing new techniques and a different approach to the creation and develop-
mentofindustriesinthesectorscoveredbytheinitiative(interviewwithJonathan
Coles, 15 September 1996). For an account of IBEC in Latin America see Broehl
(1968),andGonzalezDeluca(1994)fortheVenezuelanexperience.
29 Being a capital-intensive sector, the relative size of the refining total investment in
manufacturinghasbeenverysignificantthroughouttheperiodthatis,35percent
in1961.Itsshareofmanufacturingoutputhasgravitatedtowards11percentsince
1950forexample,12percentin1960and10percentinthe1980s.Intermsofthe
labour force, refining has never exceeded 8000 employees which makes it a very
small contributor to employment in manufacturing that is, 1.2 per cent in 1980
correspondingto5714employees.OutputfiguresaretakenfromBCV(1990b),and
thoseofcapitalandemploymentfromOCEI'sEncuestasIndustriales.
30 However, in 1936 out of the 8025 manufacturing establishments counted in the
census,therewereonly 160 firmswith morethan20employees.The artisansector
accountedfornearly60percentofoutputinthesector(Co rdova,1963).
31 SeeRodr guezCampos(1983),Rinco n(1982)andCastillo(1990)foramoredetailed
treatmentoftheeconomicprogrammeofthePerezJimenezregimeanditsideologi-
calunderpinning.
32 In 1952 there was a revision of the trade treaty with the USA. As a result, import
tariffswereraisedonanumberofproductssuchaschemicals,biscuits,juices,canned
fish and tyres. A number of these products were already being produced in the
countrybysubsidiariesofUStransnationalcorporationssuchasGoodyear,Firestone,
Nabisco,andYukery(Arenas,1990).
33 The Hydrocarbons Law of 1943 was instrumental in augmenting the government's
control of, and participation in, foreign-owned industry and increasing the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
processing of crude oil within the country. The agreement about refining was not
part of the final document of the law, but as a result of the negotiations with the
government the companies agreed to increase significantly the proportion of the
crudetoberefinedinthecountry.SeeMachado(1990)foradetaileddescriptionof
the negotiations surrounding the 1943 Law, and in particular on the issue of the
refineries.
34 Acronymfortransnationalcorporations.
35 Already in 1947, during the first government of Ro mulo Betancourt, the CVF in
association with Nelson Rockefeller envisageda plan for the constructionof a steel
plant on the banks of the Orinoco river, which would supply the increasing local
demand. This plan contemplated a partnership between the state and foreign and
nationalprivatecapital.However,theprojectwasabandonedafterthemilitarycoup
of1948(Moncada,1995).
36 See Mart n Frechilla (1994, pp. 22031) and Moncada (1995) for an account of this
caseofconflictofinterest.
37 The figures are taken from Karlsson (1975, pp. 21819) who uses unpublished data
fromCordiplan'sSecondIndustrialSurveyof1966.
38 Apparentlythiscourseofeventswasdifficulttoforeseethreedecadesbefore.Inthe
opinion of Bell: `The Puerto Cabello and Valencia district is not highly developed
industriallyanddoesnotpromisetobecomeso'(Bell,1922,p.261).
39 AcronymforSiderurgicadelOrinoco.
40 Also on this occasion national private capital was excluded, although there are
exceptions of private, national oil-related firms. The first is the Mito Juan company
foundedin1965byHumbertoPenalozaanoilengineer,formeremployeeofCreole
whose operation was limited to exploiting marginal wells. Also between 19712,
R.Tudela,aVenezuelanbusinessman,establishedafirmtooperateintheoilsector.
Bothinitiativescametoanendwiththenationalizationoftheoilindustryin1976.
41 TheCVGcomprisesseveralsubsidiaries;amongthemostimportantaretheOrinoco
Steel Corporation (SIDOR), the Caroni's Hydroelectric Corporation (EDELCA), and
thealuminiumsmelterCaroni'sAluminiumCorporation(ALCASA).
42 Defined as the weight of imports in the internal final demand for manufactured
goods.
43 FigurestakenfromBitarandTroncoso(1983,tableA-4,p.260)basedonCordiplan
andBCVstatistics.
44 A Christian Democrat party which, together with Accio n Democratica (AD), dom-
inatedthepoliticalsceneduringthe1970sand1980s.
45 Atthebeginningofthe1980stherewere239non-financialfirmsinVenezuelainthe
public sector, involved in a variety of activities such as iron extraction, steel and
aluminiumproduction, communication, electricity, airlines, water, tourism, and so
on.ForananalysisofthestateenterpriseinthecountryseeKelly(1984).
46 Itis important to notethatdirectemploymentin the oil industry accountsfor less
than1percentoftotalemployment.
47 An analysis of this situation of low growth in the middle of the oil boon is in
HausmannandMarquez(1983).
48 The figures for the total and machinery, at 1984 prices, were Bs.68.4 billions and
Bs.16.9billionsin1979,andBs.33.9billionsandBs.4.1billionsin1982respectively.
Source:Baptista(1989)basedonBCV.
49 At the end of 1982 the external debt owed to commercial banks amounted to
US$27.2 billions. This figure was not high relative to the country's export
capacity. In 1981 the debtexport ratio was 134, well below the Latin American
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 233
averageof270;however,42percentofthedebtwasdueforrepaymentwithinayear.
Thus, Venezuela's problem was its debt structure rather than debt total, see Cline
(1984).
50 The model considers the case of a positive shock in the booming sector resulting
from a once-and-for-all exogenous technical improvement, a windfall discovery of
newresources,oranexogenouspermanentriseintheexternalpriceofthebooming
product relative to all other prices. The shock is treated as permanent. The basic
referencesareCordenandNeary(1982)andCorden(1984).Peltzer(1944)provides
anearlyanalysisofthe`Dutchdisease'inVenezuela.
51 If the booming sector is an enclave and does not compete with other sectors for
factors of production, the resource boom affects the domestic economy solely
through the spending effect which operates in the same manner as an exogenous
transfer.
52 Actually,thebestcandidateforthelaggingsectorinthiscaseisthetraditionalexport
agriculturethathasnotbeenabletorecoversincethecrisisinthe1930s.
53 Theabundanceofforeignrevenuescreatedbyoilexportshastwocontraryeffectson
tradepolicy.Ontheonehand,itrelaxestheneedtousetariffsasasourceforfiscal
revenues. On the other hand, it generates a pressure for the overvaluation of the
currency. In this context there is a tendency to increase protection in order to
compensatethetradablesectorforitslossincompetitiveness.
54 The formula we are using to calculate the RER is ePi=P, where e is the nominal
exchangerate,PiisthepriceindexofanaverageofVenezuelanimportsindollars
notincludingtheeffectoftariffsandPstandsfordomesticpricesmeasuredbythe
consumerpriceindexinthemetropolitanareaofCaracas(Baptista,1997).Infact,a
better measure for the effect of changes in relative prices on resource allocation is
given by the ratio of tradable goods to non-tradable goods, but this is a more
information-demanding index, so good estimates of it are only available for the
more recent period. However, the ratio of the price of tradable to non-tradable
goodsfortheperiod196886isconsistentwiththepatternsobservedintheRERof
Figure8.3here(Astorga,1991).
55 Thistooktheformofdirectsubsidiesfortheproductionofgoodsintheprivatesector
inawiderangeofagriculturalproducts,lowpricesinintermediarygoodsproduced
by state enterprises such as steel products, and other goods produced by the public
sectorsuchascommunications,electricityandpetrol.
56 ThefigurespresentedinFigure8.3areexposedtoseveralbiases;twoofthemare:(i)
by using economically-active population as a proxy for labour we are not dis-
countingtheunemployedandinconsequenceourfiguresunderestimatetheoutput
per worker employed; (ii) due to the presence of a significant non-monetized
economy during the first half of the century the figure of output, particularly in
agriculture, might appear lower than it should have been. We are not considering
moreappropriatemeasuresoflabourproductivityfortherecentperiodbecausethat
would undermine the consistency of the series needed to show the long-run tend-
ency.
57 In this section we look at the results of productivity only at an aggregate level.
Escobar (1984) in a study of the performance of manufacturing by sectors during
the 1970sfound agreat dispersionin rates ofgrowthof labourproductivity. So the
conclusionsbasedontheanalysisofthewholeofmanufacturingmaybemisleading
ifconsideredasrepresentativeofparticularindustries.
58 According toBaptista(1997)thecapitaloutputratioofthenon-oileconomyregis-
teredvaluesclosetofivepointsduring197789,considerablehigherthananaverage
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
of 3.5 for the period 196076. Back in the 1950s Furtado saw the bias towards the
adoption of capital-intensive techniques as one of the peculiarities of Venezuelan
development,andattherootofthedivergencebetweenproductivityandwagesand
acauseoftheslowabsorptionofthepopulationintoproductiveactivities(Furtado,
1960).
59 Theautomobileindustryisarepresentativecaseofoverdiversificationofproduction
andunderutilizationofcapacity.Sinceitsinstallationatthebeginningofthe1960s
no singlecompanyhas everreached an annual productionrun greater than 45000
units in an industry where efficient plants operate at production runs between
200000 and 300000 units. The problem of economies of scale is compounded by
therelativelylargenumberofassemblers14operatinginthecountryby1989and
bythediversityinthenumberofmodelsofvehiclesproducedanaverageoffiveper
assembler.
60 Dataonelectricalenergyprovidesinformationondifferentaspects.Thetrendcom-
ponent of the series tends to be linked both to a process of industrialization and
urbanization,whereasfluctuationsaroundthistrendareassociatedwithchangesin
industrialoutput.Inaddition,per- capitaestimatesconveyinformationaboutqual-
ityoflifeandwelfare.
61 See Mart n Frechilla (1994) for the history of construction in Venezuela between
190858.AlsoMachadoetal.(1981).
62 There is a jump in the figures for enrolment in primary education around 1960
coinciding with the overthrow of the military in 1958. According to the official
figures, total enrolment increased 40 per cent in the years between 1958 and 1960
from1.045millionstudentsto1.451million.Althoughwithdemocracyeducation
becameapriority,thefiguresseemtooverstatetheimmediateeffectofthischangein
policy.AllfiguresquotedcomefromBCV(1978).
63 Accordingtoinformationprovidedbypopulationcensuses,thenumberofforeigners
living in the country was 49928 in 1941 representing 1.4 per cent of the total
population the same number (50000 foreigners accounting for 2 per cent of
totalpopulation)estimatedbyonesource(Ramos,1923,p.341)in1905.Itreached
526 188 persons by 1961 representing 7 per cent of the total population, nearly
half of them from Spain, Italy and Portugal. The census of foreigners in 1980
registered 2.3 millions corresponding to 15 per cent of the total population. The
immigration process in Venezuela is revealed in detail in Berglund and Hernandez
(1985)fromwherewehaveextractedtheinformationrelativetothestockofforeign
population.
64 Thefollowingquoteillustratestheneedforattractingimmigrantsintheearlyyears
ofthiscenturyandalsothechallengethecountryfacedfeedingthem!:`Spontaneous
immigrationislacking,althoughitisnecessarytodeveloptheagricultural,mineral,
andotherresourcesofthecountry...thequestionoffoodforEuropeanimmigrants
isserious,astherobustEuropeanrequiresfourtimesasmuchasdoesthenativeVenezuelan
oftheagriculturaldistricts,andtheamountsofwheat,oils,wines,etc.towhichthe
European is accustomed, are not easily procured in Venezuela' (Bell, 1922, p. 17,
emphasisadded).
65 See Bairoch (1988) for the characteriztics of urbanization and its links with the
economyindevelopingcountries.
66 In1920thejointpopulationofCaracasandMaracaiboaccountedfor35percentof
total urban population. In 1961 this percentage rose to 45 per cent and by 1990
dropped to 33 per cent. See: Greenfield (1994) for populations of both cities and
Baptista(1997)forurbanpopulations.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 235
References
Aranda,S.(1990)LaEconom aVenezolana,Caracas:EditorialPomaire.
Araujo,O.(1969)SituacionIndustrialdeVenezuela,Caracas:UniversidadCentraldeVene-
zuela.
(1964) `Caracterizacio n histo rica de la industrializacio n en Venezuela', Econom a y
CienciasSociales,vol.6(4),pp.527.
ArcilaFar as,E.(1961)HistoriadelaIngenier aenVenezuela,Caracas:ColegiodeIngenieros
deVenezuela,anocentenario18611961.
Arenas, N. (1990) La Denuncia del Tratado de Reciprocidad Comercial entre Venezuela y
EstadosUnidos,Caracas:CentroVenezolanoAmericano.
Astorga,P. (1991)`TheDutchDiseaseModelandtheVenezuelan Experienceduringthe
70s',MScthesis,UniversityofOxford.
Bairoch,P.(1988)CitiesandEconomicDevelopment:FromtheDawnofHistorytothePresent,
London:Mansell.
Bairoch, P. et al. (1968) The Working Population and its Structure, Bruxelles: Institut de
Sociologie,Universite LibredeBruxelles.
BancoCentraldeVenezuela(1990a)LaEconom aContemporaneadeVenezuela,edicio n 50
aniversario,Caracas.
(1990b)SeriesEstad sticasdeVenezuela,edicio n50aniversario,Caracas.
(1978)LaEconom aVenezolanaenlosU os,Caracas.
ltimosTreintayCincoAn
(1966)LaEconom aVenezolanaenlosU os,Caracas.
ltimosVeinticincoAn
Baptista, A. (1997) Bases Cuantitativas de la Econom a Venezolana: 18301994, 2nd edn,
Caracas:EdicionesFundacio nPolar.
(1989) Bases Cuantitativas de la Econom a Venezolana: 18301989, 1st edn, Caracas:
EdicionesMar adiMase.
(1984) `Mas alla del optimismo y del pesimismo: las transformaciones funda-
mentalesdelpa s',inM.Na mandR.Pinango(eds),op.cit.,pp.2041.
Baptista, A. and Mommer, B. (1987) El Petroleo en el Pensamiento Economico Venezolano,
Caracas:EdicionesIESA.
Barrios, S. (1996) The Modern Interventionist State in Venezuela.The Case of the Venezuelan
DevelopmentCorporation,Berkeley,California:UniversityofCaliforniaPress.
Bell, P.L. (1922) Venezuela: A Commercial and Industrial Handbook, Washington DC: US
DepartmentofCommerce.
Berglund, S. and Hernandez, H. (1985) Los de afuera, Caracas: Centro de Estudios de
PastoralyAsistenciaMigratoria.
Betancourt,R.(1967)Venezuela,Pol ticayPetroleo,Caracas:EditorialSenderos.
Bitar,S.andTroncoso,E.(1983)ElDesaf oIndustrialdeVenezuela,BuenosAires:Editorial
Pomaire.
Bitar, S. and Mej as, T. (1984) `La Industrializacio n en Venezuela', in M. Na m and R.
Pinango(eds),op.cit.,pp.10221.
Broehl,W.G.(Jr)(1968)TheInternationalBasicEconomyCorporation,USBusinessPerform-
anceAbroad,Vol.13,Washington,DC:NationalPlanningAssociation.
Bruton, H. (1967) `Productivity Growth in Latin America', American Economic Review,
December,pp.1099166.
Castillo, O. (1990) Los Anos del Bulldozer. Ideolog a y Pol tica 19481958, Caracas: Tro py-
kosCENDESUCV.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Cline,W.(1984)Èstructura,or genesyadministracio ndeladeudaexternadeVenezuela',
inBancoCentraldeVenezuela(1990a),op.cit.,Vol.IV,pp.556.
Co rdova,A.(1963)`Laestructuraecono micatradicionalyelimpactopetroleroenVene-
zuela',Econom ayCienciasSocialesvol.1,Caracas,pp.728.
Corden, W.M. (1984) `Booming Sector and Dutch Disease Economics: Survey and Con-
solidation',OxfordEconomicPapers,vol.36(3),pp.35980.
Corden,W.M.andNeary,J.P.(1982)`BoomingSectorandDe-industrializationinaSmall
OpenEconomy',EconomicJournal,vol.92,pp.82548.
Davila, L.R.(1988) ElEstadoylasInstitucionesenVenezuela(19361945),Caracas:Acade-
miaNacionaldelaHistoria.
ECLAC(1966a)ElDesarrolloIndustrialdeVenezuela,SantiagodeChile:UnitedNations.
(1966b) The Process of Industrial Development in Latin America, New York: United
Nations.
(1956)TheStudyofInter-AmericanTrade,NewYork:UnitedNations.
Escobar, G. (1984) `Gerentes, obreros y maquinas: la productividad industrial', in M.
Na mandR.Pinango(eds),op.cit.,pp.396420.
Figueroa Brito, F. (1974) Historia Economica y Social de Venezuela, Vols I, II, Caracas:
EdicionesUniversidadCentraldeVenezuela.
Frances,A.andNa m,M.(1994)`TheVenezuelanPrivateSector:FromCourtingtheState
to Courting the Market', in L. W. Goodman et al. (eds), Lessons of the Venezuelan
Experience,WashingtonDC:TheWoodrowWilsonCenterPress.
Furtado,C.(1960)Èldesarrollodelaeconom avenezolanaenelu ltimodecenio',Bolet n
EconomicodeAmericaLatina,vol.v,no.1,marzo,SantiagodeChile,ECLAC.
GonzalezDeluca,M.E.(1994)LosComerciantesdeCaracas,Caracas:CamaradeComercio
deCaracas.
Greenfield, G.M. (1994) `Venezuela', in G.M. Greenfield (ed.), Latin American Urbaniza-
tion.HistoricalProfilesofMajorCities,London:Greenwood.
Harris, W.G. (1971) `The Impact of the Petroleum Export Industry on the Pattern
of Venezuelan Economic Development', in R.F. Mikesell (ed.), Foreign Investment
inthePetroleumandMineralIndustries,Baltimore,Md.:TheJohnsHopkinsUniversityPress.
Hausmann, R. and Marquez, G. (1983) `La crisis econo mica de Venezuela', Caracas:
CuadernosdelCENDES,vol.1.
Hirschman,A.O.(1981)EssaysinTrespassing:EconomicstoPoliticsandBeyond,LosAngeles:
CaliforniaUniversityPress.
(1968) `The Political Economy of Import-Substituting Industrialization in Latin
America',QuarterlyJournalofEconomics,vol.lXXXII,no.1,Feb.pp.132.
International Monetary Fund (various years) Yearbook of International Financial Statistics,
WashingtonDC:IMF.
InternationalLabourOffice(variousyears)YearbookofLabourStatistics,Geneva:ILO.
Jongkind, F. (1981) Venezuelan Industrialization, Dependent or Autonomous?, Amsterdam:
CEDLA(CentrodeEstendiosdeDesamollodeLatinoAmerica).
Karlsson,W.(1975)ManufacturinginVenezuela:StudiesonDevelopmentandLocation,Stock-
holm:LatinamericaInstitutetiStockholm.
Kelly,J.(1984)`Lasempresasdelestado:dellugarcomu n',inM.Na m nalsentidocomu
andR.Pin ango(eds),op.cit.,pp.12250.
Kornblith, M. and Maingo n, T. (1985) Estado y Gasto Publico en Venezuela, 19361980,
Caracas:EdicionesdelaBiblioteca,UniversidadCentraldeVenezuela.
Kornblith,M.andQuintana,L.(1984)`Gestio ncentralyfiscalizacio ndelpoderpol tico
en los gobiernos de Cipriano Castro y de Juan Vicente Go mez', Politeia, Caracas:
UniversidadCentraldeVenezuela,vol.10,pp.143219.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PabloAstorga 237
Luzardo,R.(1963)NotasHistorico-Economicas19281963,Caracas:EditorialSucre.
Macario, S. (1964) `Protectionism and Industrialization in Latin America', Santiago de
Chile:EconomicBulletinofLatinAmerica,vol.lX,no.1,pp.61101.
Machado de Acedo, C. (1990) La Reforma de la Ley de Hidrocarburos de 1943: Un impulso
hacialamodernizacion,Caracas.
Machado de Acedo, C., E. Pacheco and E. Plaza (1981) Estado y Grupos Economicos en
Venezuela,Caracas:EditorialAteneodeCaracas.
MachadodeAcedo,C.andPadro n,M.(1987)LaDiplomaciadeLopezContrerasyelTratado
de Reciprocidad Comercial con los Estados Unidos, Caracas: Ministerio de Relaciones
Exteriores.
MachadodeAcedo,C.andArenas,N.(1995)`Lacreacio ndelaCorporacio nVenezolana
deFomento(1946)',Caracas,CENDEStemasparaladiscusion,vol.32.
Mart nFrechilla,J.J.(1994)Planes,PlanosyProyectosparaVenezuela:19081958,Caracas:
UniversidadCentraldeVenezuela.
Mayobre,J.A.(1944)`LaparidaddelBol var',inBancoCentraldeVenezuela(1990a),op.
cit.,Vol.I,pp.4384.
MazaZavala,D.F.(1964)Venezuela,unaEconom aDependiente,Caracas:InstitutodeInves-
tigacionesEcono micas,UniversidadCentraldeVenezuela.
Maza Zavala, D.F. et al. (1976) Venezuela, Crecimiento sin Desarrollo, Mexico: Editorial
NuestroTiempo.
McBeth, B.S. (1983) Juan Vicente Gomez and the Oil Companies in Venezuela, 19081935,
Mikesell,R.F.(1962)UnitedStatesPrivateandGovernmentInvestmentAbroad,Universityof
Oregon.
MinisteriodeFomento(195763)AnuarioEstad sticodeVenezuela,Caracas.
(1948)AnuarioEstad sticodeVenezuela,Caracas.
MinisteriodeHacienda(1936)MemoriadelMinisteriodeHacienda,Caracas.
Mitchell, B.R. (1993) International Economic Statistics: The Americas 17501988, London:
Macmillan.
Moncada,S.(1995)ÈntrepreneursandGovernmentsinVenezuela,19441958',doctoral
thesis,UniversityofOxford.
Na m,M.andPinangoR.(eds)(1984)ElCasoVenezuela.UnaIlusiondeArmon a,Caracas:
EdicionesIESA.
OficinaCentraldeInformac on(OCEI)(variousyears),EncuestasIndustriales,Caracas.
Peltzer,E.(1944)`Laindustrializacio nenVenezuelayelaltotipodecambiodelBol var',
inBancoCentraldeVenezuela(1990a),op.cit.,Vol.I,pp.85108.
PresidenciadelaRepu blicadeVenezuela(1929)MemoriadelPresidente,Caracas.
Ramos,A.(1923)OCafe,RiodeJaneiro:Pap.SantaHelena.
Purroy,M.I.(1982)ElEstadoylaIndustrializacionenVenezuela,Valencia:VadellHermanos.
Quiroz, I. (1987), `La pol tica petrolera venezolanafrente a la refiner a de Curazao:Una
Perspective Histo rica', in A. Serb n (ed.), Venezuela y las relaciones internacionales en la
Cuenca del Caribe, Caracas: Instituto Latinoamericano de Investigaciones Sociales
(ILDIS).
Rangel, D.A. (1971) Capital y Desarrollo. La Oligarqu a del Dinero, Caracas: Editorial
Fuentes.
(1969) Capital y Desarrollo. La Venezuela Agraria, Caracas: Universidad Central de
Venezuela.
(1968) El Proceso del Capitalismo Contemporaneo en Venezuela. Caracas: Universidad
CentraldeVenezuela.
(1958)LaIndustrializacionenVenezuela:Caracas:UniversidadCentraldeVenezuela.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Rinco n,F.(1982)ElNuevoIdealNacionalylosPlanesEconomico-MilitaresdePerezJimenez,
Caracas.
Rodr guez Campos, M. (1983) Venezuela 19481958, El Proceso Economico y Social de la
Dictadura,Caracas:AlianzaGraficaEditorial.
Sonntag, H. and De la Cruz, R. (1982) Èstado e industrializacio n en Venezuela'. Revista
MexicanadeSociolog a,pp.90539.
Sullivan, W. (1976) `Situacio n econo mica y pol tica durante el per odo de Juan Vicente
Go mez, 190835', in SullivanW. and Izard M., eds, Pol tica y Econom ca en Venezuela,
18101976,Caracas:Edicio ndelaFundacio nJohnBoulton.
US Tariff Commission (1945) Reports on Economic Controls and Commercial Policy in the
AmericanRepublics,Washington,DC:USTariffCommission.
Vallenilla, L. (1973) Auge, Declinacion y Porvenir del Petroleo Venezolano, Caracas: Monte
A
vila.
Veloz, R. (1945) Econom a y Finanzas de Venezuela. 18301944, Caracas: Impresores Uni-
dos.
Wilkie, J.W. (1974) Statistics and National Policies, Los Angeles: Latin American Center
Publications,UCLA.UniversityofCaliforniaatLosAngeles.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
9
ColombiaintheClassicalEraof
Ìnward-LookingDevelopment',
193074
Jose AntonioOcampoandCamiloTovar
*
Modern economic development, which had started to take hold in Colombia
withtheexpansionofthecoffeesectoratthebeginningofthecentury,speeded
up in the 1930s, when adjustment to external shocks generated by the Great
Depression worked as a catalyst for the spread of the dynamics of the coffee
sectortotherestoftheeconomy.Thereafter,anduntiltheoilcrisisof1973,the
economy experienced continuous growth accompanied by sweeping changes
in the production structure. These changes were backed by internal market
integrationandtheactiveinterventionofthegovernmentineconomicaffairs.
Asaresult,Colombiawentfrombeingarural,agriculturalsocietytoanurban,
semi-industrialized one. Although the country followed general regional
trends, it avoided some of the excesses of ìnward-looking development' in
other Latin American nations, especially too much emphasis on import sub-
stitutionandunduegovernmentinterferenceinproductionactivity.Thedevel-
opment process only partially benefited the population, however. Social
inequality, already considerable at the beginning of the period analysed here,
increased,leadingtoanevenlessequitablesociety.
ThischapteranalysesthemainelementsofColombianeconomictransforma-
tionduringtheclassicalLatinAmericaneraofìnward-lookingdevelopment'.
It is divided into four parts. In the first, we present a brief general account of
the economic growth and structural transformation experienced by the
Colombian economy between 1930 and 1974. In the second, we analyse
the process of change in production patterns, particularly the consolidation
of industrialization, and the role of economic policy in this process. The
* We would like to thank Enrique Cardenas, Jorge Katz, Adolfo Meisel, Juan Carlos Ram rez, Rose-
maryThorpandotherparticipantsintheseminarsthattookplaceinOxfordandPaipaColombia
fortheircommentsonpreviousdraftsofthismaterial.
239
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
240 Colombia:Inward-LookingDevelopment,193074
thirdpartexaminesthetransformationoftheagriculturalsector,andtherural
conflicts that accompanied this transformation. Finally, the fourth part sum-
marizeshowtheproductsofdevelopmentweredistributedamongthevarious
socialgroups.
Themajortrends
1
Structuralchanges
Between 1929 and 1974 the Colombian economy grew at a yearly rate of 4.5
per cent 2.2 per cent per capita. This was among the most dynamic growth
ratesintheworldin192950,butitfellbehindintheperiodofaccelerationof
worldeconomicgrowthbetween1950and1973.Thus,whileColombianGDP
percapitagrew1.7percentannuallyin192950and2.2percentin195073,
that of the industrialized countries grew by 1.3 per cent and 3.5 per cent,
and that of the Latin American countries by 1.3 per cent and 2.4 per cent in
those periods (Maddison, 1989). This process was accompanied by important
changesindemography,inthelabourmarketandintheproductionstructure.
Fromthedemographicpointofview,thecountryexperiencedanimportant
transition,characterizedbyareductioninmortalitythatwasonlyfollowedby
a lagged decrease in the birth rate. As a result, population growth accelerated
from 2.2 per cent in 193851 to 3.1 per cent between 1951 and 1973. At the
sametime,thecountryunderwentrapidurbanization.Whilein1938only31
percentofthepopulationlivedinurbanareasand8percentinoneofthefour
maincities,by1973theseproportionshadrisento59percentand25percent,
respectively. The attraction exerted by the cities in terms of job and welfare
opportunitiesplayedanimportantroleinthisprocess,asdidalsotheexodusof
the rural population generated by insecurity in the countryside, especially
between the mid-1940s and the mid-1960s, during the period known in
Colombianpoliticalhistoryas`LaViolencia'.
The process of urbanization and the change in production patterns that
accompanied it had major effects on the labour market, not only in terms of
job-creation but also of labour relations. Thus, the share of primary sectors
in job-creation decreased from 62 per cent in 1938 to 35 per cent in 1978, as
secondaryactivitiesandespeciallyservicesbecamethemainsourcesofemploy-
ment generation. Indeed,while the share of secondary sectors in job-creation
increasedfrom17percentto22percentinthosefourdecades,thatofservices
increasedfrom21percentto43percent.Withrespecttolabourrelations,the
proportionofwage-earnersinnon-agriculturalemploymentincreasedfrom58
percentto67percent,whileinagriculturalactivitiesitremainedaround44per
cent.Thus,theincreaseintherelativeweightofnon-agriculturalemployment
andofwage-earnersinurbanactivitieswasreflectedinanincreaseintheshare
ofwage-earnersintotalemployment,from51percentto60percent.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2
4
1
Table9.1 Colombia:compositionofeconomicactivity(estimatedatconstant1975pesos)
192529 193034 193539 194044 194549 195054 195559 196064 196569 197074
Agriculture
a
47.7 46.2 43.7 37.5 40.5 33.6 31.3 28.9 26.6 23.9
Mining 3.1 3.5 3.6 3.2 2.8 3.2 3.2 3.1 3.0 2.3
Manufacturing 7.8 7.6 10.2 15.7 14.8 17.4 19.4 20.6 21.1 22.5
Construction 3.4 2.1 2.8 4.0 3.4 2.8 3.4 2.9 3.2 3.6
Trade 9.6 15.0 15.1 13.8 11.6 10.4 9.8 9.8 9.9 10.4
Financialservices 3.5 4.0 5.0 5.6 6.5
Transport 2.3 1.9 2.8 4.6 4.9 6.6 6.9 6.9 6.9 7.3
Communications 0.2 0.2 0.3 0.6 0.5 0.3 0.4 0.5 0.7 0.8
Electricity,gasandwater 0.3 0.4 0.6 0.7 0.8
Governmentservices 7.5 7.6 7.4 7.0 6.1 7.3 6.9 7.1 7.0 7.2
Personalservices 18.3 15.8 14.0 13.7 15.5 7.8 7.6 7.3 7.2 7.3
Housingrentals 6.7 6.7 7.3 8.0 7.5
a
Includingfishing,huntingandsilviculture.
Sources: NationalAccountsofECLAC(192550),BancodelaRepublica(195070)andDepartamentoAdministrativoNacionaldeEstad sticasDANE(197074).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Economic growth coincided with radical changes in the production struc-
ture. In particular, the share of agriculture in gross national product (GNP),
whichhadbeen48percentinthesecondhalfofthe1920s,decreasedto24per
cent in 197074 (see Table 9.1). As the relative importance of the agricultural
sector declined, other sectors flourished, in particular manufacturing, trans-
port, public services, telecommunications and, to a lesser extent, trade and
financialservices.Allthesesectorsincreasedtheirshareinnationalproduction
remarkably. Overall, their contribution to GNP increased from 20 per cent in
192930tomorethan44percentinthefirsthalfofthe1970s.
These changes in production patterns were accompanied by a declining
trend in the share of exports in national production, which decreased from
24 per cent to 15 per cent during these years (see Table 9.2). Exports were
dominated by coffee, which accounted for more than 70 per cent of external
sales until the 1950s, the remainder consisting essentially of mining exports,
especially gold, oil and oil derivatives. However, starting in the 1960s the
country's export structure underwent a major transformation characterized
by the emergence of non-traditional agricultural and manufacturing items.
By the first half of the 1970s, they already accounted for more than 40 per
centofexternalsales.Asweshallseebelow,thistransformationwastheresult
ofpoliciesaimedexplicitlyatdiversifyingthecountry'sexportstructure.
Theimportcoefficientalsoexperiencedadownwardtrendduringthesefive
decades, although with a cyclical pattern that reflected the evolution of the
country's external purchasing power. There were also significant changes in
import composition, particularly during the first phases of industrialization.
Thus,importsofconsumergoods,whichaccountedformorethan47percent
ofthecountry'sexternalpurchasesinthe1930s,decreasedtolessthan10per
centoftotalimportsbythemid-1950s.Thereafter,asischaracteristicofsemi-
industrializedcountries,thebulkofexternalpurchasesconsistedofintermedi-
ateandcapitalgoods.Theformerincreasedfrom29percentofimportsinthe
1930stomorethan50percentinthefirsthalfofthe1970s,whileimportsof
capitalgoodsincreasedfrom20percentto38percentoftotalimportsduring
thesameperiod.
Phasesofeconomicgrowth
The lack of dynamism of traditional exports and the slow take-off of new
exportitemsleftthecountry'sexternalpurchasingcapacityand,consequently,
itseconomicactivityheavilydependentonfluctuationsininternationalcoffee
prices during most of the period under analysis. During the Great Depression
thecountrysufferedtheeffectsofbothlowerexportincomeandcapitalflows,
as wellasahigher service ofthe substantialexternal debtaccumulated in the
1920s. This situation generated a significant drainage of foreign exchange
reserves, which led to an external debt moratorium in the early 1930s. This
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
c
2
4
3
Table9.2 Colombia:externaltraderatiosandcomposition
192529 193034 193539 194044 194549 195054
b
195559 196064 196569
Ratios
a
Exports/GDP 24.0 24.4 24.0 20.4 21.6 19.0 17.8 16.5 15.8
Imports/GDP 24.1 13.1 15.9 11.2 16.7 21.4 18.4 16.3 15.5
CompositionofExports
c
Coffee 67.7 60.5 53.6 60.7 72.1 78.7 76.2 68.9 61.0
Gold 3.2 7.5 16.7 17.4 5.6 2.7 2.5 2.8 1.7
Oilandfuel-oil 14.7 20.2 19.4 14.3 14.6 13.6 14.3 16.1 13.5
Non-traditional 14.3 11.7 10.3 7.7 7.7 5.0 7.0 9.1 15.0
PrimaryManufactures
d
3.1 8.7
Compositionofimports
Consumptiongoods 43.5 47.9 41.6 29.1 23.6 14.6 9.4 7.3 7.9
Intermediategoods 21.3 28.9 28.9 45.6 33.7 44.4 50.2 45.2 48.0
Fuels 4.3 3.2 1.9 0.5
Capitalgoods 32.2 19.8 23.6 19.6 35.7 36.5 36.6 44.0 41.9
Others 0.2 0.6 0.6 1.6 1.7
a
Goodsandservices(constant1975prices).
b
Excludes1951forthecompositionofimports.
Goodsandgold.
d
SITCgroups5to8.
Sources: NationalAccounts(seeTable9.1),DANEForeignTradeYearbooks(severalissues)andRevistadelBancodelaRepublica(severalissues).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
moratorium,togetherwiththeexpansionofcoffeeexportvolumes,facilitateda
%

o
f

G
D
P

a
t

1
9
7
5

c
o
n
s
t
a
n
t

p
r
i
c
e
s

recoveryofimportsandeconomicgrowthand,inparticular,aboomofindus-
trialproduction.Thisexpansionsloweddownduetoacollapseofcoffeeprices
inthefirstyearsoftheSecondWorldWarand,evenmore,bythedifficultyof
importing due to the rationing imposed by the USA during the critical war
years. The growth of GDP, which had reached an annual rate of 5.9 per cent
between1931and1939,fellto2.6 percentin193945(Ocampo,1987).
2
The rise in coffee prices in the first decade of the postwar period facilitated
rapideconomicgrowth,atarateof5percentayearbetween1945and1955.At
the same time, the rate of investment in fixed capital, especially imported
machinery and equipment, reached record levels (see Figure 9.1). Directly or
indirectly,ahighproportionoftheseinvestmentswentintothemanufacturing
industry,whichachievedaveryrapidgrowthrateduringthoseyears,aswewill
see. Thefirst decade of the postwar period can, thus, be appropriately charac-
terizedasthephaseofindustrialmodernization.
This phase was interrupted by the fall in international coffee prices that
began in the mid-1950s. In spite of renewed access to external financing,
especially to the resources provided by the USA through the Àlliance for
Progress' and by the new multilateral lending agencies the World Bank and
theInter-AmericanDevelopmentBankthecountryexperiencedmorethana
decade of severe foreign exchange shortage. Nevertheless, the economy's
growth rateduringthese yearsofsevereconstraints wasbyno meansunsatis-
factory,4.4percentayearbetween1955and1967.
12
10
8
6
4
2
0
192529 193034 193539 194044 194549 195054 195559 196064 196569 197074
Figure9.1 Colombia:fixedinvestmentinmachineryandequipment
Source: NationalAccounts(seeTable9.1).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Jose AntonioOcampoandCamiloTovar 245
This phase was followed by the period of greatest economic growth in
Colombia, 196774. During these years, which can appropriately be called
the `golden age', GDP grew at an annual rate of 6.5 per cent. The take-off of
non-traditional exports and the regularization of external capital flows were,
undoubtedly,thebasisoftheexpansion,whichalsocoincidedwiththeendof
the process of rapid structural change that had begun in the 1930s. Thus, the
phaseofacceleratedindustrialdevelopmentendedamidstarelativeabundance
offoreignexchange.Moreover,forthefirsttimeinclosetoacentury,thelatter
situation wasnotareflection of high international coffeeprices. The interna-
tional crisis which began with the oil price shock of 1973 marked the end of
thisexpansion.
During the five decades considered in this chapter, Colombia experienced
only moderate inflation. On average, the inflation rate was 10 per cent per
annumbetween1932,whenthedeflationthatwasafeatureofthecriticalyears
of the Great Depression ended, and 1971, when the country entered a new
stage in its history of inflation. During this period, phases of accelerating
inflation were almost invariably associated with sharp adjustments in the
exchange rate especially in 193334, 195657 and 1962 which affected
thegeneralpricelevelthroughincreasedimportprices,orthroughtheshortage
offoreignproductsduringthecriticalyearsoftheSecondWorldWar,194244.
Nevertheless,intheabsenceofindexationmechanisms,theseshockswereonly
temporary and were thus followed by years of relatively low inflation, asso-
ciatedtoalargeextentwiththenormalizationofimportsupplies.Thispattern
wasbrokenin1972whenaphaseofhigherinflationbegan:itwasmarkedby
the generalization of implicit or explicit indexation mechanisms, the first
examples of which were a system of mini-devaluations, introduced in 1967,
and indexed financial assets (Unidad de Poder Adquisitivo Constante, UPAC),
introducedasasavingsinstrumentin1972.
Emergenceandconsolidationofstateinterventionism
Theprevioustrendswereaccompaniedbyanincreaseinthesizeandachange
intheconceptionandstructureofthepublicsector.Beforethe1930s,govern-
menteconomicinterventionswerelimitedtoafewareas,especiallythedevel-
opment of transport infrastructure, and the idea of the state as a regulator of
economicactivitywasabsent.Inthe1930snew,prolificformsofintervention
were developed within the framework of a new concept of the state as a
regulator of social and economic life. These new forms and concepts did not
emerge as a result of the application of Keynesian and ECLAC (Economic
Commission for Latin America and the Caribbean) theories, with which they
werelaterassociated. Onthecontrary,theyweregenerallydevelopedasprag-
maticanswerstosuccessiveexternalcrisesandthepressureofnewsocialforces
that erupted into the political scene. It must be pointed out, however, that it
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
was influenced by certain juridical and political schools of thought and by
previousorconcurrentdevelopmentsinothercountriesoftheregion.
Duringthe1930s,somepeculiarfeaturesofColombianstateinterventionism
were moulded, particularly the combination of active intervention in the
external sector and in credit markets with great conservatism in fiscal and
monetary policies. New policy instruments emerged as a result of the severe
fiscal and monetary contraction of the early part of the decade, which led to
the adoption of counter-cyclical policies, particularly exchange control, loans
fromtheBancodelaRepu blicatothecentralgovernment,protectionistprovi-
sions andan active reorganizationof mortgageloans(Ocampo andMontene-
gro,1984;Ocampo,1987).Astimewentby,theseinterventionmodalitieswere
complementedbythedevelopmentofdirectimportcontrols,policiestoredir-
ectcredittostrategicsectorsandactiveinterventioninthecoffeesector.
Although the emergence of state interventionism was not the result of a
preconceived plan, the new circumstances transformed the ways of thinking
about the economy. Concepts like interventionism, planning, national eco-
nomyandsocialresponsibility,amongothers,enteredpoliticalandeconomic
debatesinthecountry,andthesenewconceptsplayedanessentialroleinthe
constitutional reforms of 1936 and 1945. The first reform established that
`property is a social function that implies obligations' and that the state
couldinterveneintheeconomy`withtheaimofrationalisingtheproduction,
distributionandconsumptionofwealth,ortogivetheworkerthefairprotec-
tion that he is entitled to'. The reform of 1945 introduced the concept of
planning, establishing that the Congress would draw up `the plans and pro-
grammes to be complied with in promoting the national economy, and the
plans and programmes of all public works'. However, it took some time for
theseconceptstobeappliedinpractice.
Theseedsofinterventionismborefruitduringthepostwaryears,independ-
entlyofthesharppoliticalchangesofdirectionthatthecountryexperienced.
Planning mechanisms and new forms of economic regulation were expanded
considerably. Several international organizations and missions influenced
these developments, particularly the World Bank, the Economic Commission
for Latin America and the Caribbean, and the US Agency for International
Development. These multiple influences, together with the country's own
traditionsinparticularstrongfiscalandmonetaryconservationismresulted
inaneclecticmodel,asweshallseeinthefollowingsections.
The development of the state had two outstanding features: a growing
centralismandthemultiplicationofautonomousstateentities.
3
Asaresultof
thefirstoftheseprocesses,localandregionalgovernmentslostthecentralrole
that they had had until the first decades of the century, consolidating the
primacyofthecentralgovernment.Thesecondprocessledtoamorecomplex
state, with increasingly specialized functions, following the principles of
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
administrative decentralization. Indeed, there was a marked expansion of
state entities. In the 1940s, 18 national public enterprises or establishments
were created, in comparison to the 35 created between 1886 and 1939. The
number increased in the following decades: 32 public entities were created
in the 1950s and 43 in the 1960s. In the 1970s, the process slowed down,
with only 17 new entities being created. The new state design was clearly
brought to fruition with the administrative reform of 1968 and the constitu-
tional reform of that year. The first sought to give more coherence, organiza-
tion and control to the state machine, while the second extended the state's
powers of intervention by establishing that it would be responsible for the
general management of the economy, through planning. Greater decision-
making powers were also given to the executive to facilitate the use of inter-
ventioninstruments.
This process was accompanied by a significant increase in public employ-
ment.Nonetheless,thesizeofthestateandpublicemploymentwererelatively
moderatebyregionalstandardsanditsactivitieswereconcentratedinthemost
traditionalsectorsofstateaction,avoiding,inparticular,thecreationofalarge
publicbusinesssector.
Finally, measured by government expenditure as a proportion of GDP, the
publicsectorgrewmoreorlesscontinuouslyduringthedecadesconsideredin
this chapter. While in the first half of the 1950s it represented 9.7 percent of
GDP,inthefirsthalfofthe1970sithadalreadyreached17.6 percent.
4
More-
over,duringtheseyearspublicinvestmentasaproportionoftotalfixedinvest-
ment grew from 15.9 to 29.6 per cent. This increased expenditure was
channelled, primarily, to social investment education and health and
infrastructuredevelopment.
Fragileinternalmarketintegrationandregionaldiversity
Even though substantial progress had been made in developing a modern
transport network during the 1920s, allowing the country to benefit from a
lateandstillincipientinternalmarketintegration,thiswouldbeoneofthe
main bottlenecks for the country's growth throughout the period of analysis.
Theactiontakeninthisfield,thoughsignificant,wasslowandmetwithmany
difficulties, particularly the topographic complexity of Colombia. Transport
deficiencieswereidentifiedasoneofthecountry'smainproblemsbythefirst
World Bank Mission in 1950, headed by Lauchlin Currie. According to this
mission, transportation problems were so serious that it divided the country
intofourdistinctareas,whichinturnconstitutedseparatetradeandeconomic
entities.Thiseconomicconfigurationmeantthatimportinggoodsfromabroad
wascheaperthantransporting themfromotherregionsofthecountry.There
was exchange of goods only in the case of those products that could sustain
high transport costs salt, sugar, oil and some import and export products.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Thus,thecountryhadonlypartiallyovercomewhatthehistorianLuisEduardo
NietoArtetahadtermedtheàrchipelagoeconomy'.
In 1930 the transport infrastructure was centred on the railway network,
which had grown rapidly in the first decades of the century, reaching
2843km. Thereafter, its growth was practically paralysed, although it contin-
ued to dominate cargo transportation for a while. In the 1950s it would be
overtaken by road transport, considered more suitable for the Colombian
topography. Beginning in the 1930s, the expansion of the transport network
was,thus,focusedonroads.
Between 1930 and 1960 the road network increased six-fold. After 1960,
growth was slower, but quality tended to improve. As we shall see later,
owingtotherapidgrowthoftheroadnetworkitwaspossibletotakeadvantage
oftheinternalmarketintegrationwhichwas,undoubtedly,thefocalpointof
the economic development and industrialization that marked this period.
Nevertheless, the benefit derived from road development was partly under-
minedbythepoorqualityoftheroadsandbytheirhighrateofdeterioration.
Table9.3showsthatpavedroadsalwaysconstitutedasmallproportionofthe
total road network (also, although there are no statistics, they were rather
narrow);before1950,lessthan10percentofthetotalnationalroadnetwork
was paved and in 1974 barely 33 per cent. The rapid rate of deterioration of
theseroads,whichwereimpassableinwinter,andthelowlevelofmaintenance
meant that even newly-built roads were frequently impassable, as Lebret's
Mission(MisionEconom ayHumanismo,1958)pointedout.TheMissionreport
wentsofarastopointoutthat,duetothestateoftheroadsandthetimelostas
a result of breakdowns caused by their appalling condition, haulage costs in
ColombiawerefivetimeshigherthanintheUSA.
Despite these difficulties and thanks to the development of the transporta-
tionnetwork,thecountrywitnessedthemodificationofitsspatialandregional
structureasaresultoftheintegrationofregionalmarkets.Marketintegration
would not have been possible without the parallel development of a telec-
ommunications telephone and telegraph network. Indeed, the number of
telephone lines multiplied by 20 between 1939 and 1974, which also meant
thatcoverageincreasedconsiderably,from9to37linesperthousandinhabit-
ants during the same period. The telegraph system was also especially impor-
tant after the 1929 crisis: the number of telegrams increased five-fold in the
period193145,whilethenumberofwordstransmittedincreasedbyfour.
Until mid-century, the beneficiaries of development were the four most
industrialized cities in the country and the central coffee-producing region.
Thereafter, as the markets became integrated, commercial agriculture and
stockbreeding were extended into new areas. Bogota increased its share in
national production, from 14 per cent in 1950 to 21 per cent in 1973, as did
some regions on the agricultural frontier, in the Caribbean and Orinoquia
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
c
Table9.3 Colombia:evolutionofinfrastructure
Railway Roads Telephonenetwork
network(kms)
Paved National Total Lines Linesper
kms. %ofnational 1000
roads inhabitants
1930 2843
a
5743
1939 3156 13468 41272
1943 3467 9978 16 335 46 939
1950 3526 748
b
13330 96 662 8.6
1954 2986 1350
c
9.7 13898 23295 154309 12.1
1960 3562 3000
d
18.8 15933 36 890 269550 17.5
1962 3435 4700 322100 19.6
1971 3468 5023
e
25.1 20017 44013
g
619040 29.0
1974 3269 6856
f
32.9 20843 49770 866820 37.4
a
AccordingtoMisionEconom ayHumanismo(1958),in1930thenetworkhad3262kms.
b
Asof1948.
Asof1955,takingintoaccountnationalroadsonly.AccordingtoMisionEconom ayHumanismo
(1958)theremainingnationalroadscomprised2045kmsofunfinishedroadsand9777kmsof
tarmacroads.
d
Asof1959.
e
Includingnationalanddepartmentalroads.
f
Including3540kmsoflocalroads.
g
Including7280kmsoflocalroads.
Sources: Contralor aGeneraldelaRepublica(1941,1943and1983);MinisteriodeTransporte(1995);
Sanchezetal.(1996);ECLAC(1965);WorldBank(1950).
areas,whilethecentralcoffeeregionlostthedominantpositionithadheldin
thefirsthalfofthetwentiethcentury.
Regional development had two fundamental characteristics. On the one
hand it maintained a multipolar development pattern which distinguishes
Colombia from the majority of Latin American countries. On the other hand
therewasarelativeconvergenceofdevelopmentlevels,undoubtedlyfacilitated
bytheintenseinternalmigrationwhichcharacterizedtheperiodofanalysis.In
the mid-1970s, although some areas of the country remained marginalized
fromthedevelopmentprocess,regionaldifferenceswererelativelysmallcom-
paredwithusualpatternsinthedevelopingworld.
Economicpolicyandindustrialization
5
Thepatternsreferredtointheprevioussectionsclearlyindicatethattheinter-
nal market was the driving force of structural transformations that took off
following the Great Depression. It can thus be adequately characterized as
ìnward-lookingdevelopment'.Asinothercountriesoftheregion,manyana-
lystshaveassociatedthistransformationwiththeeffectsofeconomicpolicies.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Nevertheless, despite the fact that policies played an important role in the
structuraltransformationandindustrializationprocesses,theywerenotneces-
sarily the decisive factor. Moreover, policies did not always operate against
market forces, as is suggested by the same simplistic analyses, nor did they
followauniformpatternofimportsubstitution.Onthecontrary,otherfactors,
associated tothedynamicsoftheindustrialization andinternalmarketinteg-
ration as well as the incentives created by the cyclical availability of foreign
exchange,wereperhapsmoredecisive.Infact,economicpolicyreinforcedthe
effects of these exogenous factors to a great extent. Moreover, because of the
externalinfluencestowhichthecountrywassubject,thephaseofpureimport
substitutiondidnotlastverylong;itgavewaytoanearlymovementtowardsa
mixed model, which combined such substitution with active export promo-
tion. A detailed analysis of the industrialization process and the policies that
accompanieditallowsustounderstandtheroleplayedbyeconomicpolicyand
otherfactorsinthisprocess.
Characteristicsofindustrialdevelopment
The structural breakdown in the pattern of economic development is evident
in the continuous increase in the share of manufacturing activities in GNP
sincethe1930s.Throughoutthisprocessofindustrialization,themanufactur-
ingsectorunderwentsignificantchangesinitsproductionstructure,aswellas
inthesourcesofcapitalandbusinessleadership.Themaintransformationsin
the productionstructure aresummarized inTable9.4where, following aclas-
sification slightly different from that used by Hollis Chenery in his classic
studiesondevelopmentpatterns,thevariousindustrialsectorsofthecountry
aredividedintotraditional,intermediateandlate.Itcanbeobservedthatfrom
the point of view of trends in the production structure, the manufacturing
sectorwentthroughtwodistinctphasesduringtheperiodunderanalysis.
The first phase corresponds to the 25 years that followed the Great Depres-
sion. During this phase, industrial growth was led by the production of bev-
erages,oilderivatives,non-metallicmineralsand,especially,textiles,whilethe
more traditional industries like food, tobacco and furniture entered a long
periodofrelativelossofimportance.Thesedynamicsectors,whichconstitute
the ìntermediate' branches of the classification in Table 9.4, attained their
highest share in industrial value added in the first half of the 1950s. During
thisphase,theimportanceof`late'industries,someofwhichhadalreadytaken
their first steps during the 1930s and the Second World War, also started to
increase.
6
During the second phase, from the mid-1950s to the end of our period of
analysis, the central feature in the structural transformation of Colombian
manufacturing was the growth of these late branches paper and printing,
chemicalproducts,basicmetalsandmetalmechanicsattheexpenseofboth
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2
5
1
Table9.4 Colombia:compositionofindustrialGDP
a
(asproportionoftotalvalueaddedinthemanufacturingsector)
192529 193034 193539 194044 194549 195054 195559 196064 196569 197074
TraditionalSectors 77.4 76.2 60.8 55.6 51.5 48.4 43.0 39.9 40.4 34.1
Foodstuffs 62.8 66.9 52.1 42.6 39.4 37.5 33.6 31.8 33.5 27.8
Tobacco 14.6 9.4 8.7 7.3 7.0 6.5 5.0 4.3 4.1 3.7
Wood/wood-based 5.8 5.1 4.4 4.4 3.8 2.8 2.5
products
IntermediateSectors 13.7 16.7 30.5 35.3 37.9 38.4 36.9 37.0 35.5 36.2
Beverages 4.2 4.9 8.4 8.5 10.9 12.3 9.6 9.1 8.6 8.7
Textilesandapparel 5.0 7.9 15.1 18.9 18.1 16.8 16.9 17.2 16.6 17.2
Oilandderivatives 1.7 1.6 2.7 2.7 3.5 3.7 4.7 5.3 5.4 5.4
Non-metallicminerals 2.8 2.3 4.4 5.2 5.4 5.6 5.6 5.4 4.9 4.8
LateManufactures 8.9 7.1 8.7 9.0 10.6 13.2 20.1 23.1 24.1 29.7
Paperandprinting 1.8 2.2 3.1 3.5 4.1 4.7 6.2
Chemicalsandrubber 7.4 5.7 6.7 5.8 6.4 7.7 8.9 9.2 9.1 10.8
Basicmetals 1.5 1.3 2.0 1.5 2.0 1.0 5.6 6.6 6.4 6.5
Metallicproducts, 1.4 2.2 3.2 3.8 6.2
machineryandequip.
a
Forthe192539period,proportionoftotalvalue-addedexcludingwood,wood-basedproducts,paperandprinting.
Sources: SeeTable9.1.OriginaldatafromECLACandBancodelaRepublicahavebeenconvertedto1975prices,weightingtheproductionofthesubsectors
includedbythevalueaddedofthosesectorsaccordingtothe1975DANEAnnualManufacturingSurvey.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
the traditional and intermediate industries. Although some of these branches
achieved levels ofdevelopment comparable tothose inothercountriesof the
samesizeandpercapitaincome,othersdidnot;thisisespeciallytrueformetal
andmetal-mechanicindustries(Syrquin,1987).
Thestructural transformation ofthemanufacturing sectorcoincided witha
substantialtransformationofcraftsmanshipandsmallindustries.Thedecrease
in the share of more traditional craft activities, especially textiles basically
woollenandsisal,afterthedisappearanceofcottonhandicraftfabricsmainly
carried out by women, had begun before the 1930s. It reached a low point
around mid-century. After that, as Berry (1983) has pointed out, small-scale
industrial production changed significantly, from a competitive to a more
complementarystructuretothatofmanufacturingproduction.Thistransforma-
tionwasveryfastbetweenthe1930sandthe1960s,andwascharacterizedby
a reduction of independent employment and employment in small firms in
the production of apparel and an increase in the relative weight of metal-
mechanicand,toalesserextent,carpentryworkshops.
The structural changes that industry experienced were accompanied by an
important increase in productivity levels, especially labour productivity.
Between 1925 and 1974, industrial production per worker increased at an
annualrateofjustover3percent(seeTable9.5).However,mostoftheincrease
in industrial labour productivity resulted from the greater use of machinery,
especially during the first postwar decade when labour productivity increased
most rapidly. Thus, total factor productivity increased at a moderate annual
rateof1.2percent.
7
During the years of expansion, the progress of industrial production was
closely linked to the ups and downs of the external cycle. This affected not
only industrial growth rates, but, particularly, the sources of growth and the
sector's capital accumulation rates. Thus, after an initial decline in the first
yearsoftheGreatDepression,thesectorgrewrapidlyduringthe1930s(man-
ufacturing GDP increased at an annual rate of 9.6 percent between 1931 and
1941),sloweddownduringthecriticalyearsoftheSecondWorldWar(4.2per
centperannumin194145),speededupagainduringthepostwarcoffeeboom
(anannualrateof9.2percentbetween1945and1956),decreasedconsiderably
during the subsequent years of foreign exchange shortage (6.1 per cent in
195667) and went back to rapid growth during the `golden age' with which
ourperiodofanalysisended(8.0percentin196774).
8
Theclosecorrelationbetweenindustrialdevelopmentandtheexternalcycle
is even more evident for capital accumulation and the sources of demand
growth.Indeed,giventhehighimportcomponentofinvestmentinmachinery
and equipment, the rate of capital accumulation in the industrial sector fol-
lowedthefluctuationsofthecountry'sexternalpurchasingpowerinaprocyc-
licalway.Thus,itwasveryrapidduringthefirstdecadeafterthewarand,toa
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table9.5 Colombia;sourcesofgrowthinthemanufacturingsector
192529 192945 194556 195667 196774
Supply-sidedeterminants
ManufacturingGDP 4.5 7.8 9.2 6.1 8.0
Capital 21.2 2.8 11.4 5.4 6.4
Labour 5.6 3.7 4.0 3.0 6.2
Labourproductivity 1.1 4.0 4.9 3.0 1.7
Totalfactorproductivity 11.7 4.7 0.1 1.4 1.6
Demand-sidedeterminants
Totalmanufacturing
Domesticdemand 8.8 4.0 9.1 3.8 8.1
Importsubstitution 4.3 3.7 0.2 1.6 0.2
Exports 0.0 0.1 0.3 0.5 0.8
Totalgrowth 4.5 7.8 9.7 5.9 8.7
Traditional&intermediatesectors
Domesticdemand 4.0 5.6
Importsubstitution 0.8 0.1
Exports 0.5 0.8
Totalgrowth 5.3 6.5
Latemanufactures
Domesticdemand 3.6 12.1
Importsubstitution 3.7 0.2
Exports 0.4 0.7
Totalgrowth 7.7 13.0
Sources: Authors'estimatesbasedonNationalAccounts(seeTable9.1),manufacturingcensusesand
DANE,AnnualManufacturingSurveysandForeignTradeYearbooks.
lesserextent,duringthe`goldenage'(seeTable9.5andFigure9.1).Duringthe
upswings of the external cycle, internal demand also stimulated industrial
growth. Import substitution played a significant role in the years of foreign
exchangeshortage:the1930sand195667,especially,inthelattercase,inthe
late industries. Indeed, during the latter period, half the growth of these
industries was associated to import substitution. On the other hand, exports
playedasecondarybutincreasingrole,especiallyinthelaststagesofourperiod
ofanalysis.
Theseconsiderationsindicatethatdifferentbutcomplementaryfactorssup-
portedindustrialexpansionindifferentphasesoftheexternalcycle.Ingrowth
phases, the dynamism of internal demand and high investment rates played
theleadingrole.Duringtheyearsofexternalcrisis,importsubstitutionplayed
thisrole,accelerating,inturn,therateofstructuralchange.Asweshallseein
the following section, this behaviour was determined to a great extent by the
incentivesgeneratedbyeconomicpoliciesthroughouttheexternalcycle.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Asidefromtheeffectsoftheinternationalcoffee-pricecycleandthepolicies
to which it led, the process of structural transformation was affected by two
additional forces: internal market integration and the dynamics of the indus-
trialization process itself. In fact, the great industrial backwardness of the
country before the Great Depression was determined in large measure by
poor internal market integration. Nevertheless, as we saw in the previous
section,attheendofthe1920sthecountryfinallyhadthebasisofamodern
transport network. This fact, by itself, had a noticeable effect on industrial
development from the 1930s on, when manufacturing companies could be
designed with the national and not only with regional markets in mind.
Onceindustrialdevelopmentwaswellunderway,itreinforcedinternalmarket
integration and vice versa. Indeed, the former began to flourish thanks to
marketintegration,whilethegrowthoftheroadnetworkwaspossiblebecause
oftheinterregionaltradegeneratedbyindustrialdevelopment.
Another equally important factor in the process of changing production
patternswasthe`linkages'whichcharacterizedtheprocess,bywhichindustrial
growth reinforceditself. The expansion of a consumer goods industry created
demandforintermediateandcapitalgoods,whichsupportedthelaterstagesof
manufacturingdevelopment.Atthesametime,industrycreatedademandfor
agriculturalgoodsthat,asweshallsee,wasthebasisforthetransformationof
non-coffeeagriculture.
Until the Second World War, the main sources of industrial financing were
thereinvestmentofprofitsandnewequity;loansplayedaverysecondaryrole.
However, a percentage of the new equity was an indirect reinvestment of
profits,sincefromveryearlyontherehadbeenatendencytolinkthedistribu-
tionofdividendstothesubscriptionofnewshares(ECLAC,1957;Chu,1972).
Although the reinvestment of profits continued to be an important source of
financinginthepostwarperiod,accountingforbetween35percentand50per
centofthefinancingofindustrialfirms,therelativeimportanceofnewequity
declinedsincethe1960s;towardstheendoftheperiodofexpansion,thelatter
sourcecontributedonlymarginallytothefinancingoftheindustrialsector(see
Table9.6).
Borrowingfilledthegapleftbythissourceoffinancing.Thereplacementof
equity by borrowing was, without doubt, a result of the policy of directing
development credit to sectors considered strategic for the country's develop-
ment. According to some analysts, it could also have been related to the
introduction of the double taxation of profits in 1953. The policy of directed
credithadbeguninthe1930sand1940s,butitwasonlyconsolidatedwiththe
financial reform of 1950 when the banks were obliged to offer medium-and
long-term credit to industry. In the 1960s, it advanced further when finance
corporations investment banks were created, as also were development
fundsadministeredbytheBancodelaRepublica.Thelatterchannelledresources
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table9.6 Colombia:sourcesofcorporatefinancing
Sources UsesofFunds
Internal Increaseof Loans Fixed Inventories Financial
(reservesand share capital
retained capital
profits)
Nationalcorporations
ECLAC 194145 21.7 44.5 30.9 41.1 14.7 44.2
194652 21.9 33.7 35.4
a
42.2 13.4 44.4
Pieschacon 195354 45.8 15.6 38.7 64.6 5.0 27.9
195659 42.4 22.7 34.9 36.3 26.8 39.4
196064 31.3 24.3 44.4 39.3 15.3 45.3
196568 40.3 13.7 46.8 41.5 14.9 41.2
Manufacturingcorporations
Chu 194345 47.0 52.0 1.0
CONPES 195356 50.5 27.5 24.5
DNP
195759 45.1 35.0 20.0 50.1 28.9 21.0
Chica 196064 30.1 27.8 42.1 42.2 16.7 41.4
196568 50.1 10.6 39.4 35.9 17.3 46.8
197174 34.0 6.8 59.2 26.4 22.7 50.9
a
In1952,23percentforindustrialcorporationsvs29percentforcorporationsasawhole.ECLAC
(1957),p.123.
Note: Subtotalsdonotaddupto100forcertainyears,owingtoerrorsoftranscriptionintheoriginal
publicationsortotheinclusionofotheritems.
Sources: ECLAC(1957,Table9.1);Pieschaco n(1973,table3.7);Chu(1972,p.143);Departamento
NacionaldePlaneacio n(1962,vol.II,tablesII-9andVI-20);Chica(198485,tables2628(averageof
alternativesAandB)).
fromexternal credit, frommoneycreation or fromtaxation. The functionsof
the finance corporations, however, went beyond granting credit; like their
public predecessor, the Instituto de Fomento Industrial (IFI), created in 1940,
they were conceived as a mechanism to channel risk capital to industry, that
is,astrueinvestmentbanks(Berry,1983;Sandoval,1983).
Thestate,throughtheIFI,theEmpresaColombianadePetroleos (ECOPETROL)
created in 1951, and foreign direct investment, played a decisive role in
theemergenceanddevelopmentofthe`late'manufacturingindustries.Never-
theless, neither the public sector firms nor the multinationals attained the
relative dimensions they did in other Latin American countries. Towards the
endofthe1960s,Berry(1983)estimatedthatpublicsectorfirmsaccountedfor
only8percentofindustrialproductionandmultinationalsforonly16 percent
ofinvestmentinthemanufacturingsector.Infact,inthecaseofIFI,policywas
aimed from thebeginning atsupporting new companies during their launch-
ing stage, generally in association with private domestic or foreign investors;
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
once these companies were consolidated, the state contribution had to be
transferred to the private sector. This approach, although not adopted in
every case, avoided the Institute's investments becoming the means of con-
solidatingabigpublicbusinesssector.Thus,domesticprivatecapitalcontinued
toaccountforthebulkofproductionandcapitalaccumulation.
Theconcentrationofproductioninthehandsofjustafewcompanieswhich
hadalreadyreachedrelativelyhighlevelsinsomesectorsbeforethecrisisofthe
1930s was accentuated in the following decades. In the 1930s and 1940s,
production became concentrated to a considerable extent in several sectors,
particularly textiles. Given the reduced size of the internal market, the new
sectors that developed from the 1930s onwards were characterized by high
levels of industrial concentration from the start. At the end of the 1960s,
more than half the industrial value added was generated by industries with
moderatelyorhighlyconcentratedoligopolies(Misas,1975).
Parallel to the loss in relative importance of craft-related employment, the
industrial sector gained importance in employment generation until 1974, at
whichtimeitaccountedfor6.3percentoftotalemployment.Asaproportion
of non-agricultural employment, its share was already high in 1945 and stag-
natedaround10percentfromthe1950son.Despiteitsreducedshareinjob-
creation,thehighqualificationsrequiredofitsworkersmeantthatthelabour
supply for the sector was not elastic. On the contrary, the shortage of skilled
labour led to an increase of industrial wages relative to those of unskilled
workers until the 1960s, contributing, in fact, to the deterioration of income
distribution in the country during these years (see the last section of this
chapter).
Thanks to the political support of the governments of the Republica Liberal
(193046), the labour movement was strengthened during the first phases of
ourperiodofanalysis.Industrialworkers'unions,whichbegantoplayalead-
ing role in the overall union movement, benefited particularly from this pro-
cess.However,in1945,stilllessthanafourthofworkersinthemanufacturing
sector were unionized. With the return to power of conservatives (194653)
and the military dictatorship that followed (195357), the rise of the union
movement was interrupted. The first two Frente Nacional governments (1958
66) gave the labour movement new momentum,
9
and as a result industrial
unionsenjoyedaperiodofconsiderableexpansion.Fromthemid-1960stothe
mid-1970s,therateofunionizationinthemanufacturingsectorwasjustover
60 per cent; during the same period this sector played a leading role in the
strikemovement.
10
Theroleofeconomicpolicyinindustrialexpansion
As in most of Latin America, the ruling classes in Colombia interpreted
the events that took place in the 1930s and during the Second World War
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
as demonstrations of the danger of orientating production towards inter-
national markets. Thus, the need to build a solid industrial basis for develop-
ment was raised, but also a modern agriculture that would provide
manufacturing with inputs and would help to generate the necessary foreign
exchange for its development. These ideas, which were not completely
absorbedintothethinkingofthegovernmentsoftheRepublicaLiberal(Saenz,
1992), finally influenced the management of state policies in the postwar
period. Even then, however, the strong influence of agricultural interests in
Colombian politics prevented the adoption of a strong industrial policy
(Go mez, 1976). During the period of expansion, state actions covered several
areas including, as we have already mentioned, directed credit policies and
direct investments through IFI. However, the most decisive action was in the
areaofforeignexchangeandtradepolicies.Contrarytowhatisoftenclaimed,
state action was not intended to eliminate the price signals sent by the inter-
nationalmarketbut,toalargeextent,toreinforcethem.
The main characteristic of foreign exchange policies of this period was the
maintenanceofexchangecontrolssincetheirintroductioninSeptember1931.
Althoughtheseverityofexchangecontrolsvariedovertime,increasingduring
periodsofforeigncurrencyshortageanddecreasingduringyearsofabundance,
theywereonlyabandonedforcapitalflowsbetween1948and1967whenthey
werereplacedbyafreeexchangerateforsuchtransactions.Duringtheseyears,
management of the exchange rate went through three different periods. The
firststartedwithamassivedevaluationin193134,followedby14yearswitha
fixedexchangerate.Duringthesecondperiod,whichlastedfrom1948to1967,
the government imposed frequent steep devaluations every three to five
years and maintained a multiple exchange rate regime. The latter generally
includedfourrates:abasiconeformostimportsandsomeexports,afreerate
for capital transactions and minor exports, and two discriminatory rates for
coffeeandoil.Finally,startingin1967thecountryadoptedacrawlingpegwith
unannounced mini-devaluations and returned to the principle of a single
exchangerate.
11
Regardless of changes of regime, the management of the real exchangerate
had two major features: an upward long-run trend and a cyclical movement
oppositetothatofthetermsoftrade(seeFigure9.2).Thefirstofthesefeatures
supportedstructuraltransformationeffortsinaneconomyinwhichtheinitial
comparativeadvantagesstronglyfavouredonlyafewprimaryexportproducts.
The second, on the other hand, reflected the dual role that the exchange
policies played during the industrialization process, until the 1960s. During
coffee booms it served as a mechanism to transfer resources from the coffee
sector to industry, making investment in machinery and equipment and
the imported inputs required by the new industrial sectors cheaper. On the
other hand, during periods of foreign exchange shortage the exchange rate
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
25
50
75
100
125
150
175
Real import exchange rate
Terms of trade
I
n
d
e
x
1
9
8
6
1
9
8
4
1
9
8
2
1
9
8
0
1
9
7
8
1
9
7
6
1
9
7
4
1
9
7
2
1
9
7
0
1
9
6
8
1
9
6
6
1
9
6
4
1
9
6
2
1
9
6
0
1
9
5
8
1
9
5
6
1
9
5
4
1
9
5
2
1
9
5
0
1
9
4
8
1
9
4
6
1
9
4
4
1
9
4
2
1
9
4
0
1
9
3
8
1
9
3
6
1
9
3
4
1
9
3
2
1
9
3
0
1
9
2
8
Figure9.2 Colombia:termsoftradevsrealimportexchangerate
Sources: Real import exchange rate according to Ocampo (1989b); Terms of trade according to
ECLAC.
encouraged structural change and, in particular, the shifting of domestic
demandtowardsthenewimport-substitutionsectors.
This cyclical pattern was broken during the golden age, when our analysis
ends.Indeed,unlikeinthepreviousperiodsofforeignexchangeabundance,at
the end of the 1960s and during the first half of the 1970s, the real exchange
rateexperiencedasubstantialdevaluation.Asaresultofthis,aswehaveseen,
the rate of capital accumulation did not recover to the typical levels experi-
enced in the previous periods of abundance (see Table 9.5). Nonetheless, this
breakinthetraditionalpatternwasdecisiveinexplainingthegrowthofnon-
traditionalexportsduringtheseyears.
Foreign exchange policies were thus a major determinant of the cyclical
pattern that the industrialization process followed until the 1960s. It served
asthemainelementbywhichpriceincentivesgeneratedoverthecourseofthe
coffee cycle were transmitted to the economy as a whole. This relationship
indicates, moreover, that incentives for import substitution in periods of for-
eignexchangeshortagewereaccompaniedbysimilarincentivesfortheexport
sectors.Inthecaseofcoffee,however,theywereeliminatedbytheadoptionof
a discriminatory exchange rate and complementary tax schemes, which
absorbed the effects of the real devaluation (Ocampo, 1989b). This practice,
which began in the 1930s, was fully developed from the end of the 1950s
onwards (see Table 9.7). Onthe contrary,the incentives generatedby the real
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
c
2
5
9
Table9.7 Colombia:tradepolicyindicators
192529 193034 193539 194044 194549 195054 195559 196064 196569 197074
.6 Taxoncoffeeexports 2.6
a
4.3 13.6 0.0 7.1 12.8 25.9 36 35.3
Incentivesfornon-traditionalexports 39.0
b
36.9 29.1 29.1 23.1
.8 Averagecollectedtariff 30.6 41.5
Proportionofimportsapproved
underpriorlicensingregime
Proportionoflicensesnotapproved
Importdepositsasa%ofimports
22.5 16 11.1 18.3 12.0 14.2 17.8 17.0
21.0 29.5 50.4 78.1 70.0
15.5
c
32.3 8.8
4.8 16.2 21.2 24.9 13.0
a
193234;
b
195354;
195963.
Source: Ocampo(1990).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
devaluation, far from being eliminated in the case of non-traditional exports,
werereinforcedbyseveralpromotionschemes(seebelow).
The use of tariff and non-tariff instruments designed to support the indus-
trialization process complemented but were not a substitute for foreign
exchange policy. Protectionism, which had deep roots in the past (Ospina
Vasquez, 1955; Ocampo and Montenegro, 1984), increased during the years
of ìnward-looking development'. In 1931, an ambitious import-tariff reform
wasintroduced,butitseffectswereerodedbyinflationinthefollowingyears,
given the specific character of the tariffs. At the end of the 1930s a new
instrument began to be used: direct import control. Nevertheless, the most
extremeformsofprotectionismwereonlyadopted,inthecaseoftariffpolicies,
inthereformsof1950and,especially,of1959and1964(Mart nez,1986).The
last two were induced by the coffee-price collapse of the mid-1950s and rein-
forcedthe effects of the realdevaluation that thecountryexperienced during
these years. More importantly, the foreign exchange shortage led the author-
ities to use non-tariff instruments intensively and on an almost permanent
basis (see Table 9.7). The combination of real devaluation, high tariffs and
quantitative restrictions created high theoretical levels of protection during
the 1960s, particularly in the case of traditional industries. Nonetheless, as
pointed out by Hutchenson (1973), at the end of the 1960s this protection
was largely redundant in the traditional industries, and it was only used in a
notablewayinthe`late'manufacturingsectors(seeTable9.8).
Infact, despiterisingtheoreticallevelsofprotection, theprices ofnational,
ascomparedwithinternational,industrialgoodsfellthroughouttheperiodof
industrial expansion (see Figure 9.3). This trend, together with productivity
trends, indicates that despite high protection levels and high industrial con-
centration, the Colombian industrialization process was relatively efficient
during the period of analysis and that this greater efficiency was transferred
to domestic consumers through a decrease in the relative prices of manufac-
turedgoods(Berry,1983;PosadaandRhenals,1988).
Ontheotherhand,iftheproductionstructureofColombiaattheendofthis
periodiscomparedtointernationalpatterns,itbecomesevidentthatColombia
had avoided severe distortions. Indeed, the country preserved a relatively
traditional industrial structure avoiding excessive emphasis on intermediate
and capital goods that, owing to the country's size, were unlikely to become
competitive.Thus,thetraditionalfood,beveragesandtextilebranchescontin-
uedtoberelativelylargeforthecountry'slevelofdevelopment,whilebranches
that developed late, especially related to basic metals and the manufacture of
machinery,continuedtobesmallcomparedtoothercountrieswiththesame
levelofdevelopment(Syrquin,1987).Atthesametime,theshareofagriculture
inproductionwasrelativelylargecomparedtothatincountriesofasimilarsize
andlevel ofdevelopment. Notonly that,but,asweshall seein thefollowing
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table9.8 Colombia:effectiveprotectionaccordingtoISICclassification
ISIC Sector Applied Theoreticalprotection
1969 withaveragecollected
tariff1974
311/312 Foodstuffs 8.9 54.0
313 Beverages 103.3 75.2
314 Tobacco 83.6 97.6
321 Textiles 7.6 16.0
322 Wearingapparel,exceptfootwear 12.7 49.3
323/324 Textiles 4.1 14.8
331 Wood/woodproducts,exc.furniture 1.1 17.8
332 Furniture 10.9 57.8
34 Paper,printingandpublishing 8.0 40.8
351/352 Chemicalproducts 22.6 7.3
353/354 Oilandderivatives 4.1 6.7
355 Rubber 25.8 18.1
356 Plasticproducts 89.2 n.a.
36 Glass,chinaandearthenware 0.0 5.3
37 Basicmetals 32.1 8.8
381 Metalproducts,excludingmachinery 42.9 47.0
andequipment
382 Machinery,excludingelectrical 11.9 27.4
383 Electricalmachineryandsupplies 668.1 28.4
384 Transportequipmentandmaterial 319.0 70.9
395/390 Professional,scientificandother 0.0 24.1
equipment
TotalTraditional
a
6.7 41.7
TotalLate
a
/
b
104.5 28.4
TotalManufacturing
a
53.3 36.5
a
Weightedbyvalue-addedin1979;
b
sectors351/352,355,356,37and38.
Sources: Hutchenson(1973);Garayetal.(1974).
section,fromthe1950son,industrialdevelopmentmadeadecisivecontribu-
tiontothetake-offofnon-coffeecommercialagriculture.
The swing towards more intensive protectionism in the mid-1950s did not
preventthegovernmentfromsimultaneouslyimplementingasetofmeasures
aimed at promoting the diversification of the export base. As we have seen,
preferential exchange rates for minor exports had been awarded for the first
timein1948.Asaresultofthecoffeecrisis,thisincentivewascomplemented
in 1957 by a system of import-duty exemptions for inputs used in export
production, known in Colombia as the Vallejo Plan, with tax exemptions for
minor exports in 1960, with mechanisms to cover the foreign exchange risk
for external financing to non-traditional exports and with the first systems
of directed credit to these activities in 1964. Until 1967, incentives for non-
traditional exports were high, but unstable. Decree 444 of 1967, which
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
100
150
200
250
300
350
Consumer prices
Wholesale prices
I
n
d
e
x

1
9
4
8
1
9
4
9
1
9
5
0
1
9
5
1
1
9
5
2
1
9
5
3
1
9
5
4
1
9
5
5
1
9
5
6
1
9
5
7
1
9
5
8
1
9
5
9
1
9
6
0
1
9
6
1
1
9
6
2
1
9
6
3
1
9
6
4
1
9
6
5
1
9
6
6
1
9
6
7
1
9
6
8
1
9
6
9
1
9
7
0
1
9
7
1
1
9
7
2
1
9
7
3
1
9
7
4

Figure9.3 Colombianindustrialprices/internationalprices
Sources: Domestic consumer prices according to DANE; wholesale prices and average import
exchange rate according to Banco de la Republica; International manufacturing prices according to
WorldBank
established the crawling peg, also stabilized the incentives regime: it replaced
theexistingpreferentialexchangerateandtaxexemptionsbyataxcertificate
(certificado de abono tributario), created the Export Promotion Fund (Fondo de
Promocion de Exportaciones, PROEXPO) and improved the Vallejo Plan. Even
though, on average, subsidies decreased (see Table 9.7), this was widely
compensated for by greater stability and by the real devaluation of the peso
thattookplaceduringtheseyears(D az-Alejandro,1976).
The policy package adopted at the end of the 1950s and improved in 1967
combined protectionism with active foreign exchange management and
export promotion. A last element of this package was economic integration,
but its effects were rather limited. Its main instrument, the Cartagena Agree-
ment signed in 1969, which created the Andean Group, did not overcome
obstacles to the liberalization of Andean intrarregional trade, to the adoption
of a common external tariff and to the design of effective industrial develop-
ment programmes. Even so, during the 1970s Colombian trade with the
Andean countries, especially with Venezuela, grew substantially, although
only partly supported by preferences awarded under the liberalization pro-
grammes. Some firms also benefited from sectoral industrial development
programmes.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Theusualdescriptionofthedevelopmentmodelthatprevailedduringthese
years as ìnward-looking' or ìmport-substitution' does not reflect its global
features. After an initial phase, which can probably be described in those
terms, economic policy shifted towards a mixed model, the main goal being
to overcome the foreign exchange bottlenecks generated by the collapse of
internationalcoffeepricessincethemid-1950s.Moreover,thispackagedidnot
seek to replace but rather to reinforce the effects of changes in the external
variables.Itsadoptionhadverybeneficialresults.Indeed,notonlydiditleadto
a phase of structural transformation of the industrial sector, but also to the
emergenceofawiderangeofnon-traditionalexports.Furthermore,itallowed
manufacturing industry and the total GDP to maintain rather good growth
rates during the years of severe foreign exchange shortage in the 1960s. Also,
from1967on,itenabledtheeconomytoenterthephaseofmostrapidgrowth
sincethe1920s,withoutsuchexpansionbeingdependantonfavourableinter-
nationalcoffeeprices,asithadbeentypicallyforalmostacentury.
Agriculturaltransformationandruralconflicts
Theweakeningofthecoffeeeconomy
12
Thanks to planting in the 1920s and 1930s, coffee production continued to
expandatarelativelyfastrateuntilthefirsthalfofthe1940s,whenthecoffee
crop averaged 5.3 million bags according to FAO estimates, exceeding the
amountregisteredin1932by54percent.Thisrepresentedanannualincrease
inproductionduringthoseyearsof4.4percent,whichwasnotablealthough
lowerthantheannual7 percentincrease thathadcharacterizedthefirsttwo
decades of the century. Nevertheless, the collapse of coffee prices during the
Great Depression led to a slowing down of the rate of planting and affected
the rate of increase in production considerably. Thus, the early 1940s marked
thebeginningofalongphasewhichwouldlastuntiltheendofourperiodof
analysis, during which coffee production lost the dynamism that had charac-
terizeditinthefirstdecadesofthecentury.Indeed,inthefirsthalfofthe1970s
production totalled 7.7 million sacks (see Table 9.9), hardly 45 per cent more
thanachievedthreedecadespreviously;therateofgrowthwasthusreducedto
only1.3percentperyearduringthesethreedecades.
Although the fall in international prices was the factor that initially led to
this slowdown, slow growth in the coffee sector continued even after prices
improvedinthefirstpost-SecondWorldWardecade.Thisindicatesthat,rather
thanexternalphenomena,itwaslinkedmoretointernalfactors,especiallyto
the gradual exhaustion of the agricultural land frontier in western Colombia,
where coffee production had expanded rapidly in the first decades of the
century, and to the deterioration of plantations. Indeed, during the period of
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2
6
4
Table9.9 Colombia:growthofagriculturalsector
Coffee
a
Non-coffeeAgriculture Cattle
Area(thousandofhectares) (Valueofproduction,1975millonpesos)
Millonsof Commercial Traditional Traditional Total Comercial Traditional Traditional Total population production
60kg.bags and andmixed, andmixed, and andmixed, andmixed, (mn) (mn)
ofgreen plantation
c
dynamic
d
undynamic
e
plantation
c
dynamic
d
undynamic
e
coffee
19451949 6.1
b
19501954 6.3 373 351 1215 1939 3276 8206 8296 19778
19551959 6.5 460 379 1253 2092 4644 8792 8329 21765 12895 1759
19601964 7.4 652 401 1176 2229 7648 9570 8546 25764 13854 2020
19651969 7.8 852 465 1220 2537 10999 11598 9477 32074 14799 2315
19701974 7.7 936 690 939 2565 14926 15580 7374 37880 16088 2700
a
Coffeeyears;
b
excluding1945;
c
rice,barley,sorghum,cotton,africanoilpalm,soybeans,sugar,exportbananasandcocoa;
d
potatoes,tobacco,manioc,
plantains;
e
Corn,wheat,beansandbrownsugarloaf.
Source: Ocampoetal.(1987).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
slow growth, productivity per hectare fell at a yearly rate of between 1.0 per
cent and 1.5 per cent, according to the data that can be used as a basis for
calculations.Althoughtheneglecttheplantationssufferedduringtheyearsof
La Violencia, which hit the country's coffee regions severely in the 1940s and
1950s,probablycontributedtothedeclineinyield,thefundamentalfactorwas
the ageing of the plantations. In fact, using traditional technology coffee
plantationsreachmaximumyieldat12to15years,andproductiondecreases
thereafteratanannualrateofaround1percent.
The ageing of the coffee plantations began to be the main concern of the
country'scoffeeauthoritiesattheendofthe1950s.Bythenitwasevidentthat
coffee production in Colombia could not carry on expanding based on
traditional technology. In the 1960s in order to overcome this technological
backwardness, the National Federation of Coffee Growers (Federacion
Nacional de Cafeteros) started a campaign of technical assistance to farmers
through the rural extension services established in all of the country's coffee-
growing areas, in order to disseminate the new farming techniques which
were the result of almost 20 years of research by the National Centre for
Coffee Research (Centro Nacional de Investigaciones de Cafe, CENICAFE).
Increased planting density, the use of an improved (Caturra) variety, and the
systematicuseofchemicalfertilizerswerethemaincomponentsofthis`green
revolution'.
The process of modernization of the coffee sector advanced at a very slow
pace during the first years, since low international coffee prices created few
incentives for investment in coffee production. As a result, technological
innovation only really took off during the coffee boom that began in 1975,
beyondourperiodofanalysis.
Available data indicate that the small and medium-sized production units
thathadmadeadvancesduringthegreatcoffeeprosperityofthebeginningof
thecentury,continuedtoconsolidatetheirgainsuntilthemid-1950s.Bythen,
64percentofthecoffeeareawasconcentratedinabout221000medium-sized
orsmallparcelsoflandofupto10hectares.Fromthatyearonwards,aprocess
ofgradualconcentrationofcoffeeplantationsandproductionbegan.By1970,
plots of 10 hectares or less only accounted for 56 per cent of the area under
coffeecultivation.Nonetheless,coffeegrowingcontinuedtobeoneoftheleast
concentratedactivitiesofthenationaleconomyand,infact,thesmallerunits
successfullyassimilatedthenewtechnologiesdevelopedbytheFederation.
Fromthe1930sonwards,thedevelopmentofthecoffeesectorwasbackedby
a complex institutional structure that regulated its activities. The National
Federation of Coffee Growers had been established in 1927, but it only
started to intervene to any extent in the internal market at the beginning of
the 1930s. These years saw the construction of the first warehouses and the
firstmarketinterventionsinregionswheredomesticmark-upswereexcessive.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
The Federation negotiated and administered the first, unsuccessful, interna-
tional coffee agreement, signed with Brazil in 1936. Thus, beginning in the
1930satraditiondevelopedbywhichthecoffeeproducersandthegovernment
negotiatedtheformulationofinternalandexternalcoffeepolicies,theimple-
mentationofwhichremained,however,inthehandsoftheFederation.Under
thisinstitutionalarrangementithasbeenpossible,inspiteofsomeconflicts,to
manage Colombian coffee policies successfully and with great stability for
morethanhalfacentury.
ThecreationoftheNationalCoffeeFund(FondoNacionaldelCafe),in1940,
to facilitate the participation of Colombia in the Inter-American Coffee
Agreement signed the same year, widened the Federation's field of activity
still further. An additional incentive for its consolidation was the transfer of
partofthetaxesoncoffeeexports,firsttotheFederationandlatertotheFund.
This tax revenue, together with profits from the management of the Fund,
allowed it to considerably expand its participation in internal and external
markets. Internally, the Federation gradually developed a complete system
of financing, marketing, storage, technical assistance and technological
development, based on a network of coffee institutions. In the international
context it played a prominent role in the opening up of European markets
after the Second World War; in the 1960s, the Federation had thus become
the main coffee exporter of Colombia. Thanks to its financial strength, it
was also able to participate, with share capital, in a series of new commercial
and financial firms, notably the Grancolombian Merchant Fleet (Flota
Mercante Grancolombiana) and the Banco Cafetero, created in 1946 and 1953
respectively.
The Inter-American Coffee Agreement, mentioned above, was the first
successful experience of its kind. In particular, it regulated the entry of coffee
into the USA during the Second World War and thus avoided an excessive
depression of international coffee prices while the European markets were
temporarily closed. Shortly after the war was over, it lost its raison d'e tre and
ceasedtoexist.Thefallininternationalpricesthatbeganinthemiddleof1955
gave rise to new attempts at regulation which proved to be much more
lasting. After several partial agreements involving the main producing coun-
triesandsomeconsumercountries,thefirstInternationalCoffeeAgreement,in
which virtually all the producing and the main consumer countries particip-
ated,wassignedin1962.Thisagreementestablishedatradition,whichlasted
just over a quarter of a century,of regulating the international market during
periodsofoverproductionbymeansofasystemofquotasthatslowedthefall
ininternationalcoffeeprices.Itssuccessor,the1968Agreement,wasbrokenin
1972andwasfollowedbytheshortphaseofmarketfreedomattheendofour
periodofanalysis.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Themodernizationofnon-coffeeagriculture
13
Contemporary descriptions of Colombian agriculture at the end of the nine-
teenthcenturyandthebeginningofthetwentiethcenturyillustratethegreat
technologicalbackwardnessofagriculturalproductionatthattime.Themain
agricultural innovation during the first half of the twentieth century was,
without doubt, the introduction of coffee, which allowed the permanent use
oflandwhichuntilthenhadonlybeenexploitedbymeansoftraditionalslash-
and-burntechniques.
Despitetheeffortsofthenationalgovernmentandtheprivatesector,which
involvedtheorganizationofseveralmissionsbyforeignadvisersandtheestab-
lishmentofexperimentalstations,ruralfarms,agriculturalschoolsandthefirst
schoolsofagronomyinthefirstdecadesofthecentury,thetechnologicalback-
wardnessremainedacuteuntilmid-century.Upuntilthen,theseeffortshada
significanteffectonagricultureonlyinsomeregions,usuallynearlargecities.
Thiswasthecase,forexample,forsugarcaneproductionintheCaucaValley,or
cottoncultivationinTolima.Thefiguresshowthatduringthe1930sand1940s
agriculturelaggedbehindtherapidindustrializationthatthecountrywasbegin-
ningtoexperience.Thisfactwasnotablyapparentintherelativeproductivity
rates of the two sectors, with a gap which widened considerably, and in the
increasing imports of agricultural inputs by the expanding manufacturing
industries. In particular, in the 1940s the country imported more than three-
quartersofthecottonrequiredbyitsdynamictextilesector,whichwasatthe
heartoftheindustrialexpansionofthe1930sand1940s.
Moderndevelopmentinthecountrysidetookoffonlyinthe1950s,asalate
response to the new conditions created by the rapid industrialization process
whichgaverisetoincreasingdemandforagriculturalgoodstosatisfytheneed
for raw materials, food for a growing urban population and new exports to
generatethenecessaryforeignexchangetosupplythecapitalgoodsandinter-
mediate inputs required by the more dynamic sectors of the economy. This
stimulated a process of accelerated modernization of the sector between the
1950sandthe1970s,replacingcoffeeasthecoreofagriculturaldevelopment.
This growth and transformation of the sector was possible owing to the
availability of labour and underutilized arable land, to the diffusion of new
technology,toabusinessclassthatvisualizedthepotentialbenefitsofmechan-
ization of the countryside, to the growth and integration of the domestic
market, to the various mechanisms used to protect agricultural production,
and to the opportunities that the export markets started to provide in some
sectors. Between the first halfof the 1950s and the first halfof the 1970s, the
areaundercultivationwithmaincropsotherthancoffeeincreasedfrom1.9to
2.6 millionhectares,whileproductionincreasedby92percent(seeTable9.9);
thatis,atanannualrateof3.3percent.Theincreaseinproductivityoftheland
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
inthisquarterofacenturywasasubstantial1.9percentper year.Alternative
estimates that take into account overall agricultural activity give very similar
figuresfortheincreaseintotalfactorproductivityduringthisperiod.Thus,the
Mision deEstudios delSector Agropecuarioestimated an annual growth of global
productivityof2.4percentbetween1950and1975,withagrowthofagricul-
turalGDPof3.8percent.
14
The largest expansion of the area under cultivation during these years was
that planted with commercial and plantation crops, which constituted the
heart of business agriculture. They accounted for 19 per cent of the area
undercultivation atthe beginning of the 1950s, but thisfigure had increased
to36 percentbythefirsthalfofthe1970s(seeTable9.9).Thiswaspossibleasa
result of the modernization of old crops (rice, cotton, sugar cane for refining
and bananasfor export) andtothe introduction of newcommercial products
(sorghum,soybeansandAfricanoilpalm,inparticular).
Attheotherendofthescale,theareaundercultivationwithtraditionaland
mixed crops typical of the rural economy (corn, wheat, beans and brown
sugarloaf)decreasedinabsoluteterms,andwasdrasticallyreducedasapropor-
tionofthetotalareaundercultivationwithcropsotherthancoffeefrom63
percentto37percentduringthisquarterofacentury.Othertraditionaland
mixedcropspotatoes,tobacco,cassavaandplantainsafteraperiodofslow
growth until the mid-1960s, expanded dynamically during the final phase of
ourperiodofanalysis.
Productivitytrendswerealsodifferentforthevariousproductsandcontrib-
utedtoenhancechangesinproductionpatterns.Thus,whileyieldperhectare
increasedanannual3percentduringthisquarterofacenturyforcommercial
crops, for the remainder it grew at lower and irregular rates. This increase in
productivitywaspossiblethankstothemodernizationoftechniques,inparticu-
lar the use of machinery, chemical supplies and improved seeds, and to the
adaptationofthelandinordertofacilitateirrigation,drainage,ploughingand
cultivation.
In the first postwar years, tractors and other agricultural machinery, which
until then had only been used to an extremely limited extent in Colombia,
started to be imported. At the beginning of the 1970s, mechanized farming
methods were already being used on a quarter of the area under cultivation,
andonamuchhigherproportionoflandwherecommercialcropsweregrown.
Ontheotherhand,theuseofagrochemicals,althoughintroducedratherlate
in the mid-1950s was much more widespread, particularly in the case of
fertilizerswhichwerealreadybeingusedonatleast60percentoflandunder
cultivation in the mid-1970s. The introduction and adaptation of improved
varieties, although it had precedents going back to the beginning of the
century,wasonlyconsolidatedwiththecreationofastableschemeoftechno-
logical research in the 1950s and, especially, the 1960s. Land development
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
came later and it was limited; it was associated with the establishment of
irrigation districts by the Colombian Institute of Agrarian Reform (Instituto
Colombiano de Reforma Agraria, INCORA) and private investment in a few
regionsofthecountrymainlyintheCaucaValley.
Theabove-mentioneddifferencesinarea,yieldandproductionmethodsfor
thedifferenttypesofcropswereareflectionoftheheterogeneoustransforma-
tion of the rural sector. Commercial agriculture used flat land, with relatively
specialized medium-sized and large production units, adopted the `green
revolution', employed wage-earning labour and participated in the market
mainly asaproducerof rawmaterials forindustry.Thepeasant sector,on the
other hand, used hilly land, in small units with a strong bias towards mixed
farming,crucialfordomesticfoodsupplyandwithbigadvantagesintermsof
ensuring the family's survival, but had a shortage of suitable land, business
capacitiesandcapitaltomodernizeproduction.
Although the agricultural expansion displaced cattle-raising from the best
lands, the opening up of new areas on the country's agrarian frontiers meant
that grazing land increased from 16.3 million hectares in 195054 to 22.2
million hectares in 197074, according to Kalmanovitz' (1978) estimates.
Nonetheless, it continued to be an extensive activity with low productivity:
thenumberofcattleincreasedatasimilarratetothatoftheareagrazed,rising
from 12.9 million head inthe second half ofthe 1950s to 16.1 million inthe
first half ofthe 1970s, a yearly increase of 1.5 percent. Veterinary techniques
reducedthemortalityrateoftheherdsandmadepossibleanannualincreasein
production of 2.9 per cent during the same period, a rate, however, that was
lowerthanthatofnon-coffeeagriculture,especiallycommercialcrops.
Fromthe1960s,thedynamismofagriculturalproductionallowedthesector
tocontributesignificantlytotheincreaseinthecountry'sexportbase.Among
theproductsexported,bananasandtobaccohadalong-establishedplaceinthe
country'sexternaltrade,althoughinthecaseofbananaspostwarexportdevel-
opment took place in a completely new region Uraba. Two additional pro-
ducts,sugarandcotton,weretheresultofaprevioussuccessfulphaseofimport
substitution. In the case of sugar, this phase took place in the 1930s. Never-
theless,large-scaleexportsofsugaronlystartedatthebeginningofthe1960s,
when the Cuban revolution led the USA to redistribute the import quota
assigned to Cuba. These exports formed the basis for the later conquest of
otherexternalmarkets.Cottonexperiencedabriefphaseofimportsubstitution
in the 1950s, which, from 1960 on, led to exports based on active sectoral
policies.
This modernization process required very active state intervention in agri-
culturaldevelopment.Startinginthe1930s,thesepoliciesresulted,inparticu-
lar, in the gradual creation of an agricultural intervention scheme that was
muchmoreelaboratethanthatdesignedtopromotethemanufacturingsector.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Theschemeincludedseveralelements.Thefirstwasaspecialsystemtofinance
the sector, which included the creation of three commercial banks the Caja
Agraria in 1931 and the Banco Cafetero and the Banco Ganadero in the 1950s
andacomplexsystemoffinancialregulationsthatgavethebanksaccesstothe
resources of the Banco de la Republica in order to promote agriculture, and
forced all of them to channel part of their credit for the sector. Even
though the origins of these regulations dated back to the 1940s, they were
onlyfullydevelopedsincethe1950financialreform.Thesemechanismswere
perfectedatalaterstagewiththecreationoftheAgrarianDevelopmentFund,
Fondo Financiero Agrario, in 1966 and its transformation into the Fondo de
Fomento Agropecuario in 1973, both as rediscount mechanisms in the Banco de
15
laRepublica.
The pioneering attempts in the area of research and technological advisory
serviceswereboostedinthe1950sbyspecialsupportprogrammesoftheRock-
efellerFoundationandbythecreationoftheDivisionofAgriculturalResearch
of the Ministry of Agriculture. In 1962, the Colombian Agricultural Institute
(InstitutoColombianoAgropecuario,ICA)wassetup.ThegoalsofICAwereonly
achieved fully in 1968, when the research and technology diffusion pro-
grammes were centralized in the Institute. In the area of marketing the
NationalSuppliesInstitute(InstitutoNacionaldeAbastecimientos,INA)wascre-
ated in 1944. After the 1968 administrative reform, it became the Institute of
Agricultural Trade (Instituto de Mercadeo Agropecuario, IDEMA). The Agrarian
Reform Institute (INCORA) should be added to this network and will be con-
sidered in more detail in the nextsection; it was also in charge of land devel-
opmentpoliciesduringourperiodofanalysis.
Totheseeffortsshouldbeaddedthesectoralpromotioncampaignsthatwere
initiatedattheendofthe1940sbyseveralinstitutionscreatedforthispurpose,
notablytheinstitutesforthepromotionofcotton,tobaccoandbarleyandthe
cattle-raising funds. Producer organizations generally managed these pro-
grammes,aroundwhichtheprovisionofmachinery,agrochemicals,improved
seedsandtechnicalsupport wasorganized.Someofthemalsobegantoparti-
cipate in product processing and marketing, and in some cases in research.
Although in most cases they were financed by voluntary contributions from
their members, some of them had dues fixed by law. The earliest and most
important example of such private arrangements was, of course, the National
Federation of Coffee Growers. Starting in the 1950s, many others joined in
(Asocana,Fedearroz,Federalgodon,Fedegan,Fenalce,andsoon).
Many analysts have argued that the exchange rate and external trade pol-
icies,designedtopromotetheindustrialsector,hinderedagriculturaldevelop-
ment.Thisassertionisnotsupportedbythehistoricalfacts,asformerPresident
Carlos Lleras Restrepo persistently noted. In fact, from the end of the nine-
teenth century when protectionism was adopted as the official policy, the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
countryalwaysprovidedtariffprotectiontoagriculturesidebysidetothatfor
new manufacturing activities. The same was true for non-tariff measures,
whichinthecaseoftheagriculturalsectorincorporated,duringourperiodof
analysis, rigorous supervision of import quantities authorized on the basis of
shortfallsinthedomesticmarketinmanycasesundertheimportmonopoly
of IDEMA and several parallel mechanisms to guarantee the purchase of
national harvests. For their part, agricultural machinery and supplies always
receivedthelowestlevelsofprotectionintheColombiantariffstructure.
The results of this can be observed in historical trends. The take-off and
expansion of commercial agriculture, although it occurred rather late, coin-
cidedwiththephaseofintensivestructuraltransformationofmanufacturing.
Aswehaveseeninprevioussections,agricultureinColombiaaccountedfora
normaltolargeshareofnationalproductionforacountryofitssizeandlevel
of development. On the other hand, as was seen in previous sections, it
remained relatively backward with respect to the development of some `late'
branchesofmanufacturingindustry.
Ruralviolenceandagrarianreforms
Thehighdegreeofinequalityandtheconsequentstruggleforlandownership,
conflicts inherent to the colonization process and the state's tenuous hold in
ruralareas,havebeenthecausesofanothersignificantfeatureoftheColombian
countrysidesincethe1930s:persistentruralviolence.Nonetheless,itsmodal-
itiesandthegroupsinvolvedhavechangedsignificantlyinthecourseoftime.
The1920sweremarkedbyincreasingruralstruggles,especiallyinthecoun-
try'scoffee-growingareaswherelargeestateshaciendasprevailed.Thesecon-
flicts continued into the next decade, but were progressively replaced by
confrontations between conservatives and liberals which characterized the
period of Colombian history known as La Violencia. Between the mid-1940s
andtheendofthe1950s,vastruralareaswereinvolvedinanundeclaredcivil
warastheirinhabitantswerepersecutedbythearmyandbyguerrillasofeither
politicalaffiliation.Manypeoplewerekilled,othersweredispossessedoftheir
lands and goods, large sectors of the population migrated and, obviously, the
socialstructureandthepowerrelationshipsintheregionsoftheconflictswere
deeplyaffected.
During the first years of the National Front (195874) the nature of the
violence changed and, in some cases, it was also transferred to other parts of
the country. The new phase of conflict, which lasted until 1965, was chiefly
markedbyruralbanditry.Thiswasaresultofthedisintegrationofsomeofthe
groups involved in the first wave of violence, and the emergence of the first
armed revolutionary movements, formed by rural guerrillas with liberal and
communist affiliations who had been active in the initial phase, and from
urbanstudentmovementsradicalizedbythesuccessoftheCubanrevolution.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Inthesecondhalfofthe1960s,socialtensionsinthecountrysidedecreased
somewhat in view of the success of the National Front two-party model,
attempts at reform with the second agrarian reform act and military action
against the strongholds of bandits and guerrillas. In 1968 the National Peas-
ants' Association (Asociacion Nacional de Usuarios Campesinos, ANUC) was
constituted;LlerasRestrepo'sgovernmentconceiveditasaninstitutionalchan-
nelforthefarmerstoexpressdissent.Thefirstyearsofthe1970swereagainof
turmoil,undernewtypesofconflict:theinvasionofland,communityprotests
andotherformsofsocialpressuretomaketheagrarianreformmoreeffective.
These were led by a radicalized ANUC. The movement's internal divisions,
however,weakeneditsactionfromthemid-1970sonwards.
The unequal ownership of land has been one of the determining factors in
persistingruralviolence.Inthe1960sand1970s,ownersofplotsoflandofless
than 20 hectares represented between 85 per cent and 87 per cent of all rural
landowners,buttheyonlyhadbetween16 percentand18percentoftheland;
attheotherextreme,lessthan4percentofproprietors,whoownedonaverage
over100hectares,held60percentoftheland(Lorenteetal.,1985).
Theeffortsmadebythestatestartinginthe1930storeformthisstructureof
land ownership hadthree characteristics:they were gradual,intermittent and
they were closely linked to efforts to pacify the countryside.
16
In the 1920s,
coffee plantation owners in the east of the country started to come under
increasingpressurefromtheirtenantstodoawaywiththesemi-feudaltenant
system:theywereseekingtoreplacethemixtureoflandleaseandwork,which
wasattheheartofworkingrelationshipsonthehaciendas,foroneofland-lease
only.Thetenantsorganizedthemselvestodemandthesechangesandtheyalso
joined thenewpolitical movements ofthetime, giving anationaldimension
toaregionalconflict.
Theinitialansweroftheliberalregimethatbegantogovernthecountryin
1930wasalimitedagrarian reform,with adequatecompensationtotheland-
owners. However, from the end of Olaya Herrera's administration (193034)
thecontentoflegalreformstartedtobediscussed,anditfinallytookshapein
Law 200 of 1936. This first agrarian reform far from challenged the country's
land-ownership structure. Although it changed that structure permanently in
some regions, particularly the coffee-growing area of Cundinamarca, the
number of peasants helped by it represented only 6.5 per cent of the tenants
and occupants of frontier lands (colonos) recorded in the 1938 population
census. Moreover, when the peasants' movements lost strength, the land-
owners' offensive resulted in a counter-reform, Law 100 of 1944, by which,
onthebasisofadeclarationofthepublicutilityofthesharecroppingcontracts,
the owners were once more granted the type of benefits that had been called
into question in 1936: it was forbidden for leaseholders to sow permanent
crops, the cultivation of subsistencecrops was restricted, and itbecame quick
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
and easy to pay improvements, thus allowing owners to demand that share-
croppersleaveimmediately.
The harsh rural violence which afflicted Colombia in the 1940s and 1950s,
together with mass migration to the cities and the reformist ideas of the
National Front and the Alliance for Progress, were the framework for the
country's second agrarian reform. The basic concept, given concrete form by
Carlos Lleras Restrepo, arose from the need to introduce comprehensive
reformsthatwouldcombinelanddistributionwithcredit,technicalassistance
and the establishment of irrigation districts in order to reduce the serious
socioeconomic tensions in the rural areas and retain a larger proportion of
thepopulationinthecountryside.ThisideamaterializedinLaws135of1961
and 1 of 1968 and in the creation of INCORA as the implementing agency.
During the initial years of the reform, two different periods can be distin-
guished. Between 1962 and 1967, its main features were the organization of
the new Institute and support for production, infrastructure works and the
programme of supervised credit for small farmers. Between 1968 and 1971,
INCORA activities and ANUC mobilization intensified, as also did political
opposition toreform. In anycase, action to redistribute the land was limited.
Duringtheperiodofgreatestactivity,between1963and1971,668000hectares
wentintotheNationalAgrarianFund,thatis74000ayear,aquantitycertainly
insufficienttoalterthestructureofrurallandownershipsignificantly.
The opposition to the agrarian reform finally took form in the Chicoral
Agreement,signedinJanuary1972,whichvirtuallyeliminatedthelandredis-
tribution efforts except in areas where there were serious social conflicts. The
basicinstrumentsofthenewpoliciesweretheLaws4and5of1973andLaw6
of 1975. The first of these laws provided strong protection against agrarian
reformforpropertiesunder160hectaresandintroducednewcriteriatodeter-
mine whether the land was being properly exploited, which made expropria-
tionverydifficult.Simultaneously,themechanismofpresumptiveincomewas
createdtoencouragegooduseoftheland;thetaxreformof1974extendedthis
mechanism to all sectors. Law 5, which established the Fondo Financiero Agro-
pecuario,expandedtheruralpromotioncreditmechanisms.Finally,Law6 was
similar to Law 100 of 1944, since it restricted the rights of sharecroppers over
improvementsmadeonthelandduringtheleaseperiods.
Distributionoftheproductsofdevelopment
Halfway through the twentieth century, when it was possible to assess social
conditions globally for the first time, the country's social indicators were dis-
astrous.Healthandnutritionhadimprovedinthefirstdecadesofthecentury,
as the trends in several indicators reflect. Population growth had accelerated
from an annual 1.5 per cent in the nineteenth century to 2.1 per cent in the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
first half of the twentiethcentury. To agreat extent, thisacceleration was the
result of the reduction in mortality, as is indicated by the notable increase in
life expectancy: from around 30 years at the beginning of the century to 49
yearsinthe1950s.Infantmortality,althoughveryhigh,hadalreadystartedto
decrease. These mortality trends, in turn, reflect not only some improvement
in public health techniques and systems, but also in nutrition, as reflected in
theaverageheightofthepopulation,anindicator whichiswidelyusedinter-
nationallytomeasurethehistoricalevolutionofnutrition(OrdonezandPola-
n a,1992).
Inanycase,thehealthindicatorscontinuedtobetypicalofaverybackward
country;inparticular,infantmortalitywasstillashighas124perthousand.
17
Educationindicatorswereequallypoor:42percentofthepopulationover14
years of age had received no education, and only 49 per cent had had some
primaryeducation(seeTable9.10).Asaresult,theilliteracyrateofthepopula-
tionover10yearsofagewas38.5percent,andashighas50.5percentinthe
rural areas. On the other hand, unlike those for health, the education indica-
tors were not yet showing clear improvements. This only began to happen
around1950whenthelarge-scaleprimary-educationprogrammestookoff.
Table9.10 Colombia:maineducationalandhealthindicators
1951 1964
a
1973
b
Education(%ofpopulationaged15andover)
None 41.9 27.1 20.8
Primary 48.6 58.6 54.5
Secondary 8.8 13.2 21.9
University 0.7 1.1 2.6
Illiteracyrates(%ofpopulationaged10
andover)
Urban 21.3 14.4 9.9
Rural 50.5 40.6 31.3
Total 38.5 26.6 17.7
Lifeexpectancyatbirth(years)
Urban n.a. 59.4 60.9
Rural n.a. 56.0 56.5
Total 48.7 58.6 59.8
Infantmortality(rateperthousand)
Urban
Rural
Total
n.a. 76.4 68.5
n.a. 95.3 91.6
124.0 81.0 74.1
a
1966forlifeexpectancyandinfantmortality;
b
1971forlifeexpectancyandinfant
mortality.
Source: Ocampo(1992).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2
7
5
Table9.11 Colombia:qualityofhousing
1951Census 1964Census 1973Census
Admini- Elsewhere Total Admini- Elsewhere Total Admini- Elsewhere Total
strative strative strative
centres centres centres
Characteristicsofdwelling
Dirtfloor 25.0 66.7 52.7 16.1 57.1 38.6 21.9 51.1 33.7
Makeshiftwalls 75.3 97.9 90.3 29.1 69.9 51.5 19.3 60.6 36.0
Dwellingclassifiedbyavailabilityofservices
Watersupply 66.9 7.4 28.8 65.8 13.3 38.7 86.9 27.6 62.7
Energy 64.3 4.2 25.8 65.4 5.6 34.5 87.0 15.0 57.6
Sanitation 70.2 11.3 32.4 69.1 14.1 40.7 91.6 25.0 68.1
Sewagesystem 72.2 8.1 46.1
Populationindwellingswithsevere 29.9 47.1 38.8 26.7 46.4 34.2
overcrowding
a
a
Morethan3personsperroom.
Source: Urrutia(1990).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Housing conditions were equally unsatisfactory: 53 per cent of houses had
dirtfloorsand90percentofthewallsweremadeofinadequatematerial.Only
just over one-third of houses had some basic utilities. In the rural areas these
indicatorsgreatlyexceededtheurbanaverages,wheretherewasbetteraccessto
publicutilities(seeTable9.11).
Starting in the 1950s all these indicators improved substantially, although
largedifferencesremainedbetweenurbanandruralareasofthecountry.Thus,
between the beginning of the 1950s and the 1970s, infant mortality was
reducedbytwo-fifthsandlifeexpectancyapproached60years.Theproportion
of the population over 14 who were uneducated went down to 21 per cent,
gradually becoming concentrated in the older population groups, while the
percentageofpeoplewithsomesecondaryorhighereducationincreasedto25
per cent. During the same period, the number of houses with unsatisfactory
floors or walls was reduced to one-third, and access to public utilities was
extended significantly in urban areas. Despite this improvement, the rural
areaindicatorscontinuedtobeverypoor(seeTables9.10and9.11).
Severalstudiesindicate,ontheotherhand,thatbetweenthe1930sandthe
1960s there was a marked deterioration in income distribution (Urrutia and
Berry,1984;Londono,1995).Thisprocesswasinterruptedinthemid-1960sor
thebeginningofthe1970sprobablythelatterwhenaperiodofsubstantial
improvement in the distribution indicators began. Thus, towards the end of
ourperiodofanalysis,incomedistributionwasoneoftheworstinColombian
history.AccordingtoLondono'sdata,theGinicoefficientincreasedfrom0.45
in1938to0.53in1971.Thisdeteriorationreflectedbothasubstantialincrease
intheincomeoftherichestdecileofthepopulation(from35to42percent),
andareductionintheincomeofthepooresthalf(from20to16 percent).
The large amount of surplus labour in the countryside, resulting from the
modernization of the agricultural sector, the rural violence and the unequal
distribution of land, were the factors that contributed most to this result.
Everything seems to indicate that at the beginning of the 1960s the real
agriculturalwagelevelwassimilartothatofthe1930s.
In the urban sector, the main beneficiaries of economic development until
themid-1960sweretheownersofcapitalandthemiddleclasses.Duringthese
years, the manufacturing industry, the state and other urban activities gener-
ated great demand for skilled workers, whose real income increased consider-
ably.Fromthesecondhalfofthe1950s,ontheotherhand,therealwagesofall
urbanworkerswentupinasustainedway.
Table9.12confirmsthisinterpretation.Itestimatestherelationshipbetween
averagewagesinvarioussectorsoftheeconomyandincomepercapitainthe
country. As can be observed, the relative position of large manufacturing and
government workers improved until the second and the first halves of the
1960s, respectively. Construction workers, representative of the least-skilled
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table9.12 Colombia:averagewagesforselectedsectors(as
proportionofpercapitanationalincome)
195054 195559 196064 196569 197074
Manufacturing 3.19
a
3.78 4.49 5.07 4.52
Government 5.92 6.85 7.16 6.51 5.19
Construction 1.57 1.93 2.27 2.30 1.63
Agriculture 0.91 0.88 0.96 0.95 na
a
1953.
Source: Ocampoetal.(1987).
workers in urban areas, also improved their relative position, although to a
lesserdegreethanmanufacturingworkers.Therelativestabilityofagricultural
earnings with a slightly declining trend in the 1950s translated into their
worseninginrelationtourbanwages.
These conditions changed markedly in the 1970s. In particular, the educa-
tion programmes started in the 1950s began to have a considerable effect on
the supply of skilled workers, and, at the same time, heavy migration to the
cities significantly reduced surplus labour in rural areas. These facts, together
with accelerating inflation in the first half of the 1970s, translated into a
generaldecreaseinallurbanwagesforthefirsttimesincethewar.
The trends for the distributive indicators are consistent with those for pov-
erty. According to the available statistics, until the 1960s there was little
improvement in poverty indicators, and even the poorest homes experienced
adeclineinincomeperfamilymemberasaconsequenceoftheincreasedrates
of dependency characteristic of a period of rapid population growth higher
numbers of children per household and of women not working outside the
homebecauseofhavingtoraisechildren.Fortheurbansectorasawhole,while
35percentofthepopulationwasbelowthepovertylinein1964,thepropor-
tionhadincreasedto48percentin1973.Percapitafoodconsumption,which
can be considered an adequate indicator of poverty, decreased by some 7 per
centbetween195052and197274(Carrizosa,1987).
Thus,inthemid-1970sthepopulationasawholehadnotbeguntocapture
the benefit from the country's significant advances in terms of economic
growth.Eventhougheducation,healthandhousingindicatorshadimproved,
especiallyinurbanareas,povertyhadprobablyincreased,incomedistribution
haddeterioratedandtherural/urbangaphadmarkedlywidened.
Notes
ForamoredetailedanalysisoftrendsandphasesintheColombianeconomyduring
thisperiod,seeECLAC(1957),DepartamentoNacionaldePlaneacio n(1977),Gaviria
etal.(1971),Kalmanovitz(1985),Ocampo(1987)andOcampoetal.(1987).Muchof
thedatapresentedinthissectionistakenfromthelastpaper.
1
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2 Growth was uneven during the war. The slowdown was strong between 1939 and
1943,whenGDPonlygrewatarateof1.1percent.Thelasttwoyears,however,saw
relatively rapid growth (5.7 per cent annually). Moreover, during the first period,
industrialproductioncontinuedtogrowrapidlyuntil1941.
3 SeeMisio ndeFinanzasIntergubernamentales,1981,chsIIIandXIV.
4 These are figures obtained from Banco de la Republica national accounts and differ
from those that can be estimated from the new series of the Departamento Adminis-
trativoNacionaldeEstad stica.
5 Foramoreextensiveanalysisofindustrialdevelopmentinthepostwaryearsandits
relation to economic policy, see Bejarano (1978), Berry (1983), Echavarr a et al.
(1983),Ocampo(1990)andotherstudiesreferredtointhefollowingsections.
6 Afewsubsectorsalsohadrootsinnineteenth-centuryindustrialattempts.
7 Foralternativeestimates,seeBerry(1983).
8 Thecut-offpointsdiffer,insomecases,fromthoseidentifiedinaprevioussectionon
overall economic fluctuations, to appropriately capture some peculiarities in the
evolutionoftheindustrialproduction.
9 TheFrenteNacionalcoveredfourpresidentialperiodsbetween1958and1974.
10 SeeUrrutia(1969)andMisio ndeEmpleo(1986,ch.6).
11 Foramoredetailedhistoryofforeignexchangepolicies,seeWiesner(1978).
12 For a detailed analysis of topics analyzed in this section, see the work, in three
volumes, of Junguito and Pizano (1991, 1993 and 1997). See also Arango (1982),
Machado(1988),Ocampo(1989a)andPalacios(1983).
13 Foramoreextensiveanalysisoftopicscoveredinthisandthefollowingsection,see
Bejarano(1987),Berry(1995),Fajardo(1983),Kalmanovitz(1978),Londono(1989),
Machado(1986)andPerry(1983).
14 Misio ndeEstudiosdelSectorAgropecuario(1990,vol.I,p.46).
15 In the late 1980s, beyond our period of analysis, this development fund was sepa-
rated from the central bank and transformed into a rediscount (second tier) bank,
FINAGRO.
16 Aside from documents quoted in note 13, for the history of the agrarian reforms,
Machado(1984)isalsoausefulsource.
17 Foramoredetailedanalysisofsocialconditions,seeUrrutia(1990).
References
Arango,M.(1982)ElcafeenColombia,19301958:Produccion,circulacionypol tica,Bogota:
CarlosValenciaEditores.
Bejarano, J.A. (1987) Ensayos de historia agraria colombiana, Bogota: Fondo Editorial
CEREC(CentrodeEstudiosdelaRealidadColombiana).
(1978) Ìndustrializacio n y pol tica econo mica (19501976)', in M. Arrubla et al.,
Colombiahoy,Bogota:SigloXXI.
Berry,A.(1995)`TheContributionofAgriculturetoGrowth:Colombia',inJ.Mellor(ed.),
AgricultureontheRoadtoIndustrialization,Baltimore,Md.:JohnsHopkinsUniversityPress.
(1983) À Descriptive History of Colombian Industrial Development in the Twenti-
eth Century', in A. Berry (ed.), Essays on Industrialization in Colombia, Tempe:Arizona
StateUniversity.
Carrizosa, M. (1987) Èvolucio n y determinantes de la pobreza en Colombia', in J.A.
Ocampo and M. Ram rez (eds), El problema laboral colombiano: informesespeciales de la
MisiondeEmpleo,Bogota:Contralor aGeneraldelaRepu blica-DepartamentoNacional
dePlaneacio nSENA.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Chica, R. (198485) `La financiacio n de la inversio n en la industria manufacturera
colombiana',DesarrolloySociedad,no.1516,SeptemberMarch.
Chu, D.S.C. (1972) `The Great Depression and Industrialization in Latin America:
ResponsetoRelativePriceIncentivesinArgentinaandColombia,19301945',doctoral
thesis,YaleUniversity.
Contralor aGeneraldelaRepu blica,(1941,1943and1983)Sintes sEstad sticadeColom-
bia,Bogota.
(1936)AnuarioGeneraldeEstad stica,Bogota.
Departamento Nacional de Planeacio n (1977) `La econom a colombiana, 19501975',
RevistadePlaneacionyDesarrollo,OctoberDecember.
(1962) Colombia: Plan general de desarrollo economico y social, Bogota: Imprenta del
BancodelaRepu blica.
D az-Alejandro, C.F. (1976) Foreign Trade Regimes and Economic Development: Colombia,
NewYork:NationalBureauofEconomicResearch.
Echavarr a,J.J.,Caballero,C.andLondono,J.L.(1983)Èlprocesocolombianodeindus-
trializacio n:algunasideassobreunviejodebate',CoyunturaEconomica,September.
EconomicCommissionforLatinAmericaandtheCaribbean(1957)Analisisyproyecciones
deldesarrolloeconomico:EldesarrolloeconomicodeColombia,Vol.III,Mexico:ECLAC.
(1965)EltransporteenAmericaLatina,Santiago:ECLAC
Fajardo,D.(1983)Haciendas,campesinosypol ticaagrariaenColombia,19201980,Bogota:
Fundacio nFriedrichNaumannEditorialOvejaNegra.
Garay,L.J.,Mart nez,M.,Giordanelli,J.,Villaveces,R.andCastro,Y.(1974)Analisisdela
estructuradecontroldeimportacionesenColombia,ResearchReport,Fedesarrollo,August.
Gaviria, J.F., Go mez, F.J. and Lo pez, H. (1971) Contribucion al estudio del desempleo en
Colombia, Bogota: CIEDANE. Centro de Investigaciones Econo micas and Departa-
mentoAdministrativoNacionaldeEstad stica.
Go mez, H. (1976) `Los grupos industriales y el desarrollo colombiano: conjunturas e
interpretaciones',CoyunturaEconomica,December.
Grisales, C. (1983) Èl desarrollo de la industria automotriz colombiana en el contexto
internacional',MAthesis,UniversidaddelosAndes.
Hutchenson, T.L. (1973) Ìncentives for industrialization in Colombia', doctoral thesis,
MichiganUniversity.
Junguito, R. and Pizano, D. (1997) Institucion e instrumentos de la pol tica cafetera en
Colombia,Bogota:FondoCulturalCafeteroFedesarrollo.
(1993) El comercio exterior y la pol tica internacional del cafe, Bogota: Fondo Cultural
CafeteroFedesarrollo.
(1991)ProducciondecafeenColombia,Bogota:FondoCulturalCafeteroFedesarrollo.
Kalmanovitz, S. (1985) Econom a y nacion: una breve historia de Colombia, Bogota: Siglo
XXI,UniversidadNacionalCINEP.
(1978)DesarrollodelaagriculturaenColombia,Bogota:EditorialLaCarreta.
Londono,J.L.(1995)Distribuciondelingresoydesarrolloeconomico:ColombiaenelsigloXX,
Bogota:TercerMundoBancodelaRepu blicaFedesarrollo.
(1989) Àgricultura y transformacio n estructural: una comparacio n internacional',
RevistadePlaneacionyDesarrollo,JulyDecember.
Lorente,L.(Salazar,A.andGallo,A.(1985)DistribuciondelapropiedadruralenColombia,
19601984,Bogota:MinisteriodeAgriculturaCEGA.
Machado,A.(1988)Elcafe:delaaparcer aalcapitalismo,2ndedn,Bogota:TercerMundo.
(1986)Pol ticasagrariasenColombia,19001980,Bogota:CID,UniversidadNacional.
(1984)`Lareformaagraria:unamiradaretrospectiva',Econom aColombiana,August
September.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Maddison A. (1989) The World Economy in the 20th Century, Paris: Organization for
CooperationandDevelopment(OECD).
Mart nez,A.(1986)Laestructura arancelariaylasestrategiasdeindustrializacionenColom-
bia,19501982,Bogota:UniversidadNacional.
MinisteriodeTransporte,(1995)Eltransporteencifras19701994,Bogota.
Misas,G.(1975)Contribucionalestudiodelgradodeconcentracionenlaindustriacolombiana,
Bogota:TiempoPresente.
Misio n de Empleo (1986) Èl problema laboral colombiano: diagno stico, perspectivas y
pol ticas',Econom aColombiana,no.10,AugustSeptember.
Misio ndeEstudiosdelSectorAgropecuario(1990)EldesarrolloagropecuarioenColombia,
Bogota:MinisteriodeAgriculturaDepartamentoNacionaldePlaneacio n.
Misio ndeFinanzasIntergubernamentales(1981)FinanzasintergubernamentalesenColom-
bia,Bogota:DepartamentoNacionaldePlaneacio n.
Misio n Econom a y Humanismo (1958) Estudio sobre las condiciones del desarrollo de
Colombia,Bogota:PresidenciadelaRepu blica,Comite NacionaldePlaneacio n.
Ocampo,J.A.(1992)`ReformadelEstadoydesarrolloecono micoysocialenColombia',
AnalisisPol tico,no.17,Bogota,SeptemberDecember.
(1990) `The Transition from Primary Exports to Industrial Development in Colom-
bia', in M. Blomstrom and P. Meller (eds), Diverging Paths: Comparing a Century
ofScandinavianandLatinAmericanEconomicDevelopment,WashingtonDC:Inter-Amer-
icanDevelopmentBank.
(1989a) `La consolidacio n de la industria cafetera, 19301958', in Nueva historia de
Colombia,Vol.V,ch.10,Bogota:Planeta.
(1989b) `Ciclo cafetero y comportamiento macroecono mico en Colombia, 1940
1987',CoyunturaEconomica,OctoberandDecember.
(1987) `Crisis mundial y cambio estructural (19291945)', in J.A. Ocampo (ed.),
HistoriaeconomicadeColombia,Bogota:SigloXXIFedesarrollo.
Ocampo, J.A. and Montenegro, S. (1984) Crisis mundial, proteccion e industrializacion:
ensayos de historia economica colombiana, Bogota: CEREC (Centro de Estudios de la
RealidadColombiano).
Ocampo, J.A., Bernal, J., Avella, M. and Errazuriz, M. (1987) `La consolidacio n del capi-
talismo moderno', in J.A. Ocampo (ed.), Historia economica de Colombia, Bogota: Siglo
XXIFedesarrollo.
Ordonez,A.andPolan a,D.(1992)`CambiosdeestaturaenColombiaduranteelpresente
siglo',CoyunturaSocial,no.6,June.
OspinaVasquez,L.(1955)IndustriayproteccionenColombia,18101930,Bogota:Editorial
SantaFe.
Palacios,M.(1983)ElcafeenColombia,18501970:unahistoriaeconomica,socialypol tica,
2ndedn,Mexico/Bogota:ElColegiodeMexico/ElAncoraEditores.
Perry,S.(1983)LacrisisagrariaenColombia,19501980,Bogota:ElAncoraEditores.
Pieschaco n, C. (1973) `La pol tica econo mica y el mercado bursatil', in El mercado de
capitalesenColombia,Bogota:BancodelaRepu blica.
Posada, C.E. and Rhenals,R. (1988) `La dinamica econo mica colombiana: el caso de los
crecimientosindustrialyglobalentre1950y1985',PerfildeCoyunturaEconomica,no.3,
March.
Saenz, E. (1992) La ofensiva empresarial: industriales, pol ticos y violencia en los anos 40 en
Colombia,Bogota:TercerMundoEdicionesUniandes.
n n y determinantes de la pro- Sanchez, F., Rodr guez, J.I. and Nuez, J. (1996) Èvolucio
ductividadenColombia:unanalisisglobalysectorial,19501994',inR.Chica(ed.),El
crecimientodelaproduccionenColombia:resultadosdelestudionacionalsobredeterminantes
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
del crecimiento de la productividad, Bogota: Departamento Nacional de Planeacio n
ColcienciasFonade.
Sandoval,D.(1983)`Pol ticaecono micayfinanciamientodelaindustriamanufacturera
colombiana, 194582', mimeo, Bogota: Corporacio n Centro Regional de Poblacio n,
August.
Syrquin, M. (1987) `Crecimiento econo mico y cambio estructural en Colombia: una
comparacio ninternacional',CoyunturaEconomica,December.
Urrutia,M.(ed.)(1990)40anosdedesarrollo:suimpactosocial,Bogota:BancoPopular.
(1969)HistoriadelsindicalismoenColombia,Bogota:UniversidaddelosAndes.
Urrutia, M. and Berry, A. (1984) La distribucion del ingreso en Colombia, Medell n: La
Carreta.
Wiesner,E.(1978)`Devaluacio nymecanismodeajusteenColombia',inE.Wiesner(ed.),
Pol ticaeconomicaexternadeColombia,Bogota:Asociacio nBancariadeColombia.
World Bank (1950) Bases de un programa de fomento para Colombia, Bogota: Banco de la
Repu blica.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
10
Import-SubstitutingIndustrialization
inArgentina,194080:Its
AchievementsandShortcomings
JorgeKatzandBernardoKosacoff*
AhistoricalaccountofArgentina'slong-termindustrial
developmentprocessduringtheperiod194080
Themacroandmicroeconomicenvironmentofthe1930sand1940s
Argentinadidnotstartitsimport-substitutingindustrialization(ISI)processin
thepostwaryears,norwereeffortsinthisdirectioninitiallyundertakenunder
theinfluenceofconventionalimport-substitutionpoliciesandtariffprotection.
StudiescarriedoutbyVillanueva(1972),Schvarzer(1996)andotherArgentine
scholarsshowthattheindustrialexpansionoftheperiod193044(Table10.1)
wasbynomeansamajordeparturefrompreviouslyestablishedtrends.Inthis
respectVillanuevawrites:Ìndustrializationtakesoffwellbeforethe1930s...In
1935asmuchas78percentofindustrialoutputwasproducedbyfirmswhich
werealreadyinoperationbefore1930'.Moreoverbetween1924and1930indus-
trial investment and imports of machinery and equipment grew quite signifi-
cantly,particularlysoinsectorssuchaschemicals,pharmaceuticals,foodstuffs
andelectricalequipment.Thenewplantscomingon-streaminsuchsectorsadd
uptoaratherlargenumberofmeatpackinghouses,sugarmills,tanningfirms
andsoon,alreadyactiveinArgentinasincetheearlydaysofthecentury.
Furthermore,largedomesticfirmslikeBungeyBorn,SiamDiTellaorAlpar-
gatas were already investing in Brazil before 1930, showing early signs of
domesticindustrialcompetenceandanexplicitintentionofmovingoutwards
intoneighbouringmarkets.
1
It is important to understand, however, that such early steps towards local
manufacturingproductionweretakenwithinthecontextofanopeneconomy
* BothauthorsbelongtotheEconomicCommissionforLatinAmericaandtheCaribbean.Theideas
presentedhereareourexclusiveresponsibility.
282
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table10.1 Argentina:theexpansionof
manufacturingindustry,193049
Production (1970100) %GDP Annual
growthrate
1930/34 16.5 21.4 1.1
1935/39 21.5 23.7 5.4
1940/44 25.6 24.5 3.6
1945/49 36.6 27.6 6.8
Source: BancoCentraldelaRepu blicaArgentina(1976).
which did not have an explicit industrial policy supporting such efforts, or a
centralbankmanaginganactivemonetarypolicyorexchangecontrolswhich
were first introduced in Argentina in September 1931. Neither were there
industrial incentives, such as import permits, which began to be used only
latein1933.Itisonlyafterthe1933devaluationofthelocalcurrency,andthe
introductionin1932oftariffprotection,thatwecanrightlytalkaboutexplicit
import-substituting industrialization policies of an orthodox nature. Invest-
ment in industry in the earlier days of the century was mostly induced by
marketforces,thusreflecting`worldprices'and`natural'comparativeadvant-
ages,thatis,theactualopportunitycostoftheresourcestherebyemployed.
Itwasonlyaftermultilateralismbrokedownintheearly1930sandArgentina
moved into bilateral relations with the UK after signing the RocaRunciman
Treaty granting special treatment to British imports, that an active policy
regime developed, with a vast array of new institutions being created in
support of the industrialization process. These were years of high social and
economicturbulenceintheaftermathofthe1930militarycoupwhichledthe
way into a period of deeply-rooted nationalism under a pro-German military
establishment.
Inadditiontotheabove,however,weshouldbecarefulnottolookatwhat
wasgoingoninArgentinainisolationfromwhatwasgoingonintheworldat
that particular point in time. The strong disarray that prevailed in world
markets and the increasing worldwide degree of protectionism that followed
the collapse of the Economic and Monetary World Conference of 1933 con-
tributed to the development of the ideological climate conditioning Argen-
tina'sresponse,aswellasthatofmanyothercountries.Manynewinstitutions
were created locally as a response to the breakdown of the gold-standard
regime; initially they were seen as transitory measures but were later
retained as permanent features of the domestic institutional landscape. That
was true also in countries such as Spain, Brazil or Korea, confirming the
fact that Argentina was part of a more general worldwide trend. Ideas pre-
viously advanced in the international arena both by Keynes and Beveridge
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
284 ISIinArgentina,194080
concerningtheroleofthestateasanengineforgrowthandasaguarantorand
providerofpublicwelfaretookonlocalsignificanceinthehighlyidiosyncratic
interpretation of generals Farrel, Perlinger and, later on, of Coronel Pero n
himself,whowasatthatpointreturningfromatrainingperiodinMussolini's
Italy.
These were some of the outstanding features of the local and international
environment in which import-substituting industrialization began to be dis-
cussed and implemented in Argentina. The fragility of the local political and
institutionalfabric,aswellasofthecountry'sproductionstructure,shouldbe
noted as we proceed to evaluate the impact of the new policy regime in the
years that followed. It is this fragility that conditioned the way in which the
initial steps in import substitution were taken and we have to take that into
account if we want to provide arealistic description of what happened in the
1950sand1960sintermsofindustrialdevelopment.
Twomajoractorsintheindustrializationprocessrapidlyincreasedtheirrole
inthe1940sand1950s.Ontheonehand,largestateenterprisesexpandedand
increasedinnumber,involvedinproductionforthedefensesectorincluding
intermediateinputssuchasironandsteelorpetrochemicalsandresponsible
aswellforservicessuchasenergy,telecommunicationsandtransport.Onthe
otherhand,smallandmedium-sizedfamilyenterprisesproliferatedmanyof
themcreatedbyformerimmigrantsandundertooktheerectionofsmall-scale
manufacturingfacilitiesfortheproductionofhouseholddurables,pharmaceuti-
cals,andsimplepiecesofequipmentsuchaselectricmotors,pumps,industrial
boilersoragriculturalmachinery.
A new industrial policy regime now gradually emerged, centred around the
operation ofthe Bancode Credito Industrial which wasset upin1944 with the
explicit purpose of providing subsidized funding to hundreds of small and
mediumsizedenterprisesenteringthemarket.Longwaitinglistsforconsumer
durables,largerepairandmaintenanceeffortsonelderlyvehiclesandwornout
pieces of machinery, together with high tariff protection and low-cost funds
providedbytheBancoIndustrial,providedtheincentiveswhichinducedfirms
toappearandexpandintheearlydaysoftheISIprocess.
As far as the public sector is concerned, priority was given to the so-called
industries of `national interest' which were basically those related to the
defense sector. The Armed Forces through the Direccion de Fabricaciones Mil-
itaresplayedamajorrolenotjustbypointingoutwhatwerethoughttobethe
priorities for industry, but also undertaking as their explicit responsibility the
building-up of production capacity in iron and steel, railway equipment and
secondary petrochemicals. Private investment in such activities was discour-
aged,thusprovidingastrongcorporatistimprinttothewholeprocess.Suchan
institutionalclimatewastohavealong-lastingimpactformanyyearstocome,
as large public enterprises and state-run R&D laboratories YPF, YCF, Fabrica
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
MilitardeAvionesundertookacentralroleinthetrainingofhumanresources
aswellasinthedevelopmentofthelocaltechnologicalinfrastructure.
Whilesuchprocessesunfoldedinthepublicdomainoftheeconomy,inthe
privatesectoralargenumberofnewsmallandmedium-sizedenterprisescame
on-stream engaged in the production of consumer durables, garments, shoes,
pharmaceuticals and simple capital goods The country's high degree of isola-
tion from international sources of technology, the fragility of the preexisting
local industrial fabric, and the lack of qualified human resources explain the
highlyidiosyncraticnatureoftheproductionstructurethatdevelopedinthose
industriesintheearlyyearsoftheISIprocess.Localfirmsweremuchsmallerin
size than comparable facilities in more developed and mature societies. They
were also much more artisanal and immature as far as plant `lay-out' and
production organization were concerned. They normally started repairing
andcopyinglocally-availablemachinesorproductdesignswhichwereoneor
eventwodecadesbehindtheworld'stechnologicalfrontier.
Some of the outstanding features of the emerging production structure can
besummarizedasfollows:
. smallsizeofplant;
. highdegreeofverticalintegration;
. outmoded, secondhand or `homemade' machinery and equipment as well
asoldproductdesignsandorganizationaltechnologies;
. averywideproductionmixcoupledwithsmallindividualproductionruns
orbatches;
. in-house engineering and technological efforts primarily carried out with
thepurposeofcopyingand/oradaptingproductdesignsaswellasproduc-
tionandorganizationaltechnologiesbroughtfromabroad.
Therewasnoperceptionwhatsoeveramonglocalpolicy-makersthattheabove-
mentioned microeconomic features were significantly to affect the country's
growth possibilities in the years to come. Little was therefore done in those
early days of the ISIprocess tocorrect them, and many of these features were
toremainaslong-termliabilitiesofthelocalproductionstructure,someeven
until the present. As distinct from the East Asian case, competition in the
local market was absent and therefore the search for productivity improve-
ments was probably much less significant than in the case of the East Asian
economies.
Hightariffprotectionandthelackofindependentproducersforintermedi-
ate parts and components forced many firms to operate with a much higher
degree of vertical integration than otherwise advisable, supplying themselves
goodsandserviceswhichwouldhavebeenpurchasedunderàrm-length'con-
tractsfromspecializedcontractorsinmorematureindustrialsocieties.Thishad
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
astrongnegativeimpactuponbothstaticanddynamicefficiencyasitinduced
an excessive diversification of engineering activities within the plant and a
learningprocesswhichwasless-specializedthanitmighthavebeen.External-
itiesandsynergieswerethereforesmallerthantheymightotherwisehavebeen.
Plantswerenomorethan,say,10percentofthesizeofcomparableproduc-
tion facilities in more developed industrial countries. A small size of plant
togetherwithapoorfactorylay-outandanadhocproductionandorganization
technologyexplainthehighincidenceofdowntimeandlowinitialtotalfactor
productivity. Capital turnover was quite low and so was the initial compet-
itiveness of local production facilities, even if domestic wages were just a
fractionofwagesindevelopedcountries.
Thepreviouslydescribedcircumstancesinducedmanycompaniestostartin-
house technical and engineering departments engaged in the production of
incrementalunitsofengineeringandorganizationalknowledgeonthebasisof
which to upgrade product designs, production processes and work-organiza-
tion technologies. Such activities had a major cumulative impact upon pro-
ductivitygrowth,eventhoughtheywerenotparticularlydirectedatenhancing
the international competitiveness and export capabilities of local firms. For-
eign markets were not a source of attraction for local entrepreneurs at that
point in the industrialization process, nor were they made more attractive by
publicpolicies.
Within the highly particular environment we have so far been describing,
labourproductivitygrewataratherfastpace.Inaninward-orientedincentives
regime and without having to face the challenge of external competition,
manyplantsproducingtextiles,garments,shoes,chemicalandpharmaceutical
products,consumerdurablesandsimplecapitalgoodsmanagedtogrowquite
rapidly producing previously imported goods. In spite of their rather fragile
start-up many firms managed to proceed rather fast along their `learning
curve', steadily improving their technical and engineering capabilities. It is
quite understandable that the initial productivity gap was quite large but we
shouldalsonoticethatinmanycasestheinitialgapcouldbegraduallyreduced
asaresultofdomesticengineeringeffortsandtechnologicallearning.
The`goldenyears'oftheISImodel:the1950sand1960s
A number of new institutions appeared in the late 1940s fostering the expan-
sion of industry. Probably the most influential was that of import quotas and
permitsmanagedbyIAPIInstitutoArgentinoparalaPromociondelIntercambio
created in 1946. Foreign trade was nationalized and IAPI became part of a
global geopolitical strategy which went far beyond the objective of building
up local industrial capabilities. The Peronist government envisaged making
Argentina a major regional power with some voice in international affairs.
The production of arms and the attempt to develop domestic nuclear power
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
capabilities and aeronautics became part of a rather arrogant stand as far as
foreignpolicywasconcerned,
2
whichwastohavealong-lastingimprintupon
thenatureoftheindustrializationprocessand,indeed,uponArgentinesocial
lifeingeneral.
TheinitialyearsofthePeronistadministrationafterwinningthepresiden-
tial election of 1946 were a period of rapid economic growth, industrial
expansion and upwards mobility of the newly-emerging urban working class.
Table 10.2 shows a clear improvement in labour's share in GDP during those
years.
The early 1950s were characterized by the availability of foreign exchange
reserves US$1700 million of that time a rapid rate of investment, and
massive employment generation in the expanding new industrial areas of
Cordoba, Rosario, Mendoza and the outskirts of the city of Buenos Aires.
However, imports of intermediate inputs, petroleum and machinery for the
rapidly expanding industrial sector grew significantly faster than exports of
primary products, unveiling a central structural weakness of the ISI strategy:
industry was still too far behind international standards to be able to expand
intoforeignmarkets,letalonehaveaninterestinsodoing.Publicpoliciesdid
not put any pressure in this direction, as they did in many of the successful
southeastAsianNICswhichatthattimehadalowerlevelofindustrialmaturity
thanArgentina.CountriessuchasKoreaorTaiwanstartedwell-behindArgen-
tinaintermsoflocalindustrialandtechnologicalcapabilities,butmanagedto
administer in quite a different way the set of subsidies and incentives with
which their military governments fostered the pace of industrialization. The
financialandtechnologicalsupportoftheUSA,intheframeworkoftheCold
War, also helps us to understand the more rapid pace of the industrialization
processbothofKoreaandTaiwan.
By 1948 foreign exchange reserves were already exhausted and the country
began to show the initial signs of dollar shortages. Unfilled import permits
amountedinthatyeartonearlyUS$1500million,thatis,morethanoneyear's
Table10.2 Argentina:theparticipation
oflabourindomesticGDP
Shareoflabourin
domesticGDP(%)
193539 48.9
194044 46.5
194549 49.4
195054 51.7
Sources: ECLAC,BuenosAires,office;
ArgentineCentralBank.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
worth of imports. Import permits thus became a central component of
the country's industrial policy. The picture worsened even further with the
195152 drought which induced the Peronist government to drastically
change its overall growth strategy. Energy shortages were dramatic and the
country was forced to use corn as fuel. A major currency devaluation was
introduced,togetherwithatwo-yearwagefreezeandadrasticimprovementin
agriculturalprices
3
intendedtorecovertheirlaggingpositionvis-a-visindustrial
prices.Moreover,afteryearsofharshanti-imperialiststandsthePeronistgovern-
mentmanagedtogetaquickparliamentaryapprovalforanewlawattracting
foreigndirectinvestment,whichwasnowkindlyinvitedtocometoArgentina.
Theseactionsnodoubthadasignificantimpactupontherateandnatureof
theindustrializationprocess,asweshallnowsee.Considerfirsttheimpactof
policy reversals on the agricultural front. As mentioned before, the objective
here was that of redressing a longstanding conflictive relationship between
Peronism and rural landlords, as well as the policy objective of subsidizing
domestic wages through cheap staples. It was not, however, until 1956 that
is after the 1955 coup d'etat which ousted the Peronist administration and
forced Pero n himself into exile that policy along this front actually
started toachieve success following the creationof INTA(InstitutoNacional de
Tecnologia Agropecuaria)
4
and the approval of a large number of tax incentives
inducing rural producers to introduce new forms of mechanization and
organizational changes. These were to have a major impact upon yields per
acre as well as upon labour productivity. The Prebisch Report, written in the
mid-1950s, stressed the dramatic decay into which the primary sector of
Argentina had fallen during the previous decades
5
and strongly advised the
5
10
15
20
25
30
35
40
M
i
l
l
i
o
n

t
o
n
s

1910 1920 1930 1940 1950 1960 1970 1980
Figure10.1 TotalproductionofgraininArgentina,191084
Source: Authors'elaborationfromCEPALdata.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
post-Peronist administration to give priority to actions in this field. The
advicewasfollowedandtheresultsdidnottakelongtoappear,aswecansee
inFigure10.1.
A major industrial `block' emerged in the 1960s in association with the
revival of the primary sector, with a strong metalworking and chemical com-
ponentrelatedtotheproductionoftractors,agriculturalmachineryandequip-
ment, fertilizers and pesticides, whose demand increased quite substantially
(Table10.3).Thisprocesswasalsoinfluencedbythearrival,orrapidexpansion,
of MNCs like Fiat, J. Dhere, Monsanto or Dow Chemicals, which strongly
benefitedfromtheabove-mentionedchangeinagriculturalpolicy.
6
As already mentioned, the other major policy reversal occurred on the for-
eign investment front, where the militant anti-imperialist language of the
immediatepostwaryearsgavewayin1953toaForeignInvestmentLawoffer-
ingforeignfirmsimportpermitsandvariousfiscalandnon-fiscalincentivesin
exchange for the creation of domestic production capacity. Prior to this law,
government officials had already been conducting negotiations with MNCs
such as Squibb, Mercedes Benz, California Petroleum Co. and Kaiser which
very early between 1948 and 1952 decided to invest locally, receiving
major privileges for so doing. It is important to understand that Argentina's
long-termgrowthperspectivelookedatthattimesignificantlybrighterthanit
actuallyturnedouttobelateron,inducingalargenumberofforeigncompan-
ies to respond to the call. Many of them even entertained the idea of making
Argentina their focal point for the whole of Latin America.
7
As in the case of
agriculturalpolicy,however,thismajorreversalinthepolicyregimeinitiated
underPeronismonlyattainedsignificantresultsafterthemilitarytakeoverof
1955whichoustedPero nfrompower.Between1957and1965some200major
internationalfirmsdecidedtoerectmanufacturingplantsinArgentina.
Athirdmajoractorthemultinationalcorporationwasenteringthestage
onamassivescaleatthatpoint.Aswiththecaseoflargepublicfirms,orwith
small and medium-sized family enterprises, this new actor was also to have a
Table10.3 Argentina:theexpansionofmetalworkingand
chemicalproductioninthe1950sand1960s(%ofdomestic
industrialoutput)
Period Foodstuffs/ Textiles/ Chemicals Machinery/
beverages leather equipment
195054 24.6 23.4 13.0 17.0
195559 22.5 19.9 14.1 21.9
196064 20.4 15.2 15.9 28.5
196569 18.8 13.4 17.5 30.0
Source: ECLAC,BuenosAiresoffice;ArgentineCentralBank.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
long-lasting impact upon the local production fabric. Many aspects of the
prevailing industrial culture suffered the impact of this massive inflow of
foreign companies which strongly affected standards and norms, quality-
control techniques, patterns of subcontracting, and the organization of work
ontheshopfloor.Theyalsoaffectedmarketstructureandperformanceasmany
localfirmscouldnotfacetheircompetition,giventhattheyarrivedinArgen-
tina with product designs and process technologies which were well above
domesticstandards.
Central to this process was the rapid development of the metalworking
complex, led by the automobile and tractor industries. Particularly important
isthecaseofthevehicleindustry,thefirstlocalexperiencewith`continuous-
flow' Fordist production organization methods. Given the above it seems
interestingtoexplorethistopicfurther.
Consonant with prevailing state-led development ideas, the first domestic
ventureintotheproductionofvehicleswasthatofDINFIA(DireccionNacional
deFabricacionesMilitares)whichby1958hadproducedsome13000rastrojeros
andwasstartingwiththeproductionof2000Gracielaperannumasmallfour-
strokesedanbasedonGermantechnology.
Kaiser Argentina the newly-erected subsidiary of the declining US Kaiser
Corp. produced 26 000 units in 1958. In the early 1960s annual output was
well over 30 000 units, among them a large number of Bergantins, a locally-
designedvehicle(usedbytaxidrivers)whichmanagedtoputtogetherthebody
ofanItalianAlfaRomeoandthepowerplantofaJeepWillys!Theproduction
oftheBergantininvolvedmajorlocalR&Defforts,forcingthefirmtocreateits
ownproductdesignandprocessengineeringdepartments,aswellasthetrain-
ingandupgradingofdozensofdomesticauto-partssuppliers.
Many domestically-owned firms attempted at this time to enter the rapidly
expanding automobile market, competing with the recently erected subsidi-
ariesoflargeMNCs.Itisinterestingbrieflytodocumentheretheirfailure,asit
tellsusastoryaboutentrybarriers,economiesofscaleandimperfectfactorand
technologymarketshinderingthelong-termcompetitivestrengthofdomestic
companies. The case of DiTella is probably useful to illustrate the point. As a
natural follow up of its previous involvement in the production of consumer
durables,scootersandcapitalgoods,DiTelladecidedtoenterthecarmarketin
thelate1950s,producingaBritishMotorCorporationlicenseddesign.Forthe
production of some 12 000 cars per annum it employed nearly 3500 people,
withanaveragelabourproductivitywhichwasaboutone-thirdoftheworld's
best-practicefrontieratthatpointintime.Nothavingmuchexperienceinthe
fieldofautomobileproduction,thefirmranintomajordifficultiesinrelation
to production planning and organization, as well as in its access to working
capital. It also faced large diseconomies of scale as a result of its small size of
plant and the high degree of vertical integration of its overall operation. The
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
prevailingpolicyregimeinducedthedomesticfabricationofpartsandcompon-
ents instead of encouraging exports, as was the case in the southeast Asian
NICs.Thispolicyregimewasintendedtocapitalizeontheaccumulatedlearn-
ing experience of local auto-part producers attained during the previous two
decades.
While domestic demand expanded rapidly and DiTella was able to capture
the benefits of operating in a sellers'market it did quite well. The 1962 reces-
sion plus the entry of Ford and GM to the market, with clearly superior
organizational,financialandengineeringknow-how,markedtheendofDiTel-
la'spossibilities.ThefirmtriedtoenterintopartnershipwithKaiserArgentina,
by then clearly the market leader, but this also failed and Kaiser ended up
purchasing DiTella's facilities and closing them down in 1968, achieving a
strongmonopolisticstandinthedomesticcarmarket.
Summarizing: by the mid-1960s the major features of the local ISI model
werealreadyinplace.Largepublicly-ownedenterprisescontrolledbasicsectors
of the economy such as iron and steel and petroleum, and, more generally,
extendedtheirdomainintoenergygenerationanddistribution,telecommun-
ications, transport and shipping facilities. MNCs covered industries such as
vehicles, pharmaceuticals, petrochemicals, tobacco, tractors and agricultural
equipment and food processing. Finally, SMEs many of them family-owned
firms were prominently engaged in textiles, garments, furniture, and shoes
andleathergoods.
The èvolutionary dynamics' of this model need to be understood. Techno-
logical learning took place within the framework of a closed economy whose
policy regime did not induce firms to move into exports. Under such
circumstances exports only appeared as a counter-cyclical response to the
contractionofdomesticdemand,ashappenedafterPinedo's1962devaluation.
In spite of the above, however, the ISI process triggered off the creation and
diffusion of major economic institutions of crucial importance for modern
capitalist development, as well as a steady accumulation of engineering and
technicalknowledge,whichhastobeseenasamajorsourceof`socialcapital'
inthelocalenvironment.Fromthisperspectivewehavetoconcludethatthe
expansion of industry induced the gradual generation and diffusion of a
sophisticated ìndustrial culture' of long-lasting significance for the country's
growth path. Pari passu with the expansion of industrial production, many
individual firms managed to accumulate a large stock of firm-specific
technological skills and engineering know-how which allowed them to
gradually expand their productivity, closing the gap with the international
`state of the art'. Manufacturing exports did not expand significantly
until later, in the 1970s, when a specific set of incentives was used to induce
such expansion. To this third stage of the industrialization process we now
turn.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Theplateauofthe1970sandthefirststepsofArgentinaintotrade
liberalization
TogetherwithothercountriesinLatinAmericaUruguay,ChileandPeru(but
not Brazil, Ecuador or Mexico) Argentine industrial growth (Table 10.4)
slowed down in the mid-1970s and fell in absolute terms in the early 1980s.
Internalaswellasexternalreasonsexplainwhythiswasso,asweshallnowsee.
Consider first the set of internal circumstances that induced such an out-
come. The country's social climate deteriorated badly in the early 1970s with
the return of Pero n to public office and the expansion of paramilitary repres-
sion in 197475. The labour market became increasing tense and industrial
relations deteriorated badly at the shopfloor level. The degree of uncertainty
increased quite considerably, negatively affecting ànimal spirits' and invest-
ment.Inthiscontextthegovernmentintroducedawiderangeofsectoraland
regional fiscal subsidies with the intention of inducing a deepening of the ISI
process and the expansion of capital-intensive raw material processing indus-
tries, basically oriented to the local market. Special policy regimes were
designed,supportingtheexpansionofthepulpandpaperindustry,thepetro-
chemicalsector,theproductionofaluminum,steel,andsoforth.Atthemacro
levelthefiscalsituationdeterioratedrapidlyandtheeconomyenteredaperiod
ofturbulencewhichwastoexertamajornegativeimpactuponthebehaviour
ofeconomicagents,exacerbatingopportunismandrent-seekingattitudes.This
macroeconomic evolution constitutes evidence as to the structural limits of
growthstrategiesbaseduponincomeredistributioneffortsintendedtoexpand
theinternalmarket.
8
Ontheexternalside,varioussourcesofinstabilityalsoemerged.Ontheone
hand,thetermsoftradedeterioratedbadlyaftertheinternationaloilcrisis.On
the other, the increase of the international interest rate late in 1978 further
damagedthecountry'sexternalposition.Inadditiontotheaboveasignificant
factornegativelyaffectingthecountry'sexternalcompetitivenesswastheout-
wardsmovementoftheinternationaltechnologicalfrontierresultingfromthe
discovery and rapid diffusion of microprocessors. World demand for capital
good slowly started to shift towards numerically controlled machines. An
Table10.4 Argentina:averageannualgrowth
ratesofindustrialvalueadded
195074 197480 198090 199490
4.9 0:6 1:4 6.9
Source: Authors'elaborationonthebasisofECLAC's
PADIdatabase(ProgramadeAnal sisdelaDinamica
Industrial).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
expanding technological lag and the erosion of comparative advantage in
metalworking production explain why Argentina lost previously established
positions in regional markets for machine tools, agricultural equipment and
automobiles. Concomitantly with the contraction of such industries, the
recently approved sectoral subsidies for raw material processing induced a
major shiftof investmenttowards resource-basedprocessingindustries with a
lowerdomesticvalue-addedcontent.
Therateofinflationincreasedunderthepressureofarapidlygrowingfiscal
deficit 15 per cent of GDP by the mid-1970s and the economic landscape
became populated by numerous types of indexation mechanisms. Opportun-
ism, risk-aversion and the shortening of the planning horizon on the part of
privateeconomicagentsbecamepartofthelocaleconomicscenario.
TheMarch1976coupd'etatwhichoustedMariaEstelaMart nezdePero nfrom
governmentfoundthecountryinastateofhighsocialunrestanddeepmacro-
economicdisequilibrium.Onthepoliticalside,thedegreeofsocialrepression
reached levels previously unheard of in Argentina; over 30000 people disap-
peared under the impact of military repression. On the economic side, trade
liberalization,thederegulationofeconomicactivitiesand,in1978,aChicago-
inspiredneoliberalprogrammeofmacroeconomicadjustmentwereintroduced
by the new military authorities in the hope that the invisible hand of the
market could introduce new discipline into the daily functioning of the eco-
nomy.Themovementtowardstradeliberalizationandmarketderegulationhas
tobeconsideredanearlypredecessorofapolicyregimewhichwastobeapplied
in considerably more depth in the late 1980s and early 1990s. It is therefore
usefultolookinsomedetailattheearlyexperimentrunbyMartinezdeHoz.
ThepolicyregimewasintroducedinApril1976.Itsbasicphilosophywasthat
ofacompletefaithinmarketmechanismsandintradeliberalizationasaway
of furthering the rate of modernization and the degree of efficiency of the
productionstructure.
9
Asfarasindustrialpolicyisconcerned,twoquitediffer-
ent subperiods can be identified during the years in which Martinez de Hoz
acted as Minister for the Economy. The first one 197678 showed an
expansion of investment and, thereafter, of the demand for capital goods, as
well as forconsumer durables. This effect was triggered by a 40 percent wage
reduction in real terms and a major income redistribution favouring upper-
middle-class incomes. In spite of the fact that import duties were drastically
reducedonaveragetheycamedown40percentagepointsfrom90to50per
cent imports did not expand significantly during the first year of the stabil-
ization programme. On the other hand, local industry still managed to stay
fairlycompetitiveonthebasisofpreviouslyaccumulatedproductivitygains.A
highrateofexchangeandtheexpansionofconsumercreditasaresultofthe
liberalizationoffinancialmarketsalsoactedinsupportofarathershortreces-
sionafterthe1976 devaluation.
10
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
The second period started late in 1978 with a Chicago-inspired programme
basedonthemonetarymanagementofthebalanceofpayments.Theauthor-
ities expected local inflation to converge to international levels during the
period197678.Changesinthelocalexchangerateadjustedtosuchexpecta-
tions through the use of a preannounced programme of mini-devaluations
which were expected to be lower with time as the convergence was to take
place in the framework of an increasing degree of trade liberalization. This
scheme of passive monetary policy assumed that the economy had to go
throughatransitionperioddeterminedbythedifferentadjustmentcapabilities
ofinternationallytradedandnon-tradedgoodsandservices.Oncethistransi-
tion period was over the economy would have converged to a new set of
relative prices reflecting the `true' opportunity cost of the resources used. In
conjunction with trade liberalization the process was to help productivity
growth and modernization of the production structure, inducing the exit of
the less-efficient firms and industries in the economy and the entry of new,
moreefficientfirms.
Needlesstosay,theexpectedconvergencenevertookplace.Asfarastradable
goods are concerned the adjustment was very slow and imperfect, and it did
not take place at all in the case of non-tradable goods and services. The
observed changes in the rate of interest were highly dependent upon the
degree of macroeconomic uncertainty and also negatively affected by the
traditionally high cost of financial services in Argentina. Meanwhile, the rate
ofexchangewhichhadbeenfixedandpreannouncedattheexpectedequilib-
rium level of inflation was systematically below the current value of foreign
currencies.
The overvaluation of the peso and the reduction of tariff protection had a
dramatic impact upon the external trade account, fostering the massive
entrance of imported goods in 197980. Concomitantly, the unrestricted
inflow of external capital mostly short-term speculative flows induced by
thehighinterestrateandthederegulationoffinancialmarketscompensated
thecurrentaccountdeficitgeneratingarapidlyexpandingexternaldebt.
The above-mentioned dynamics clearly signalled an imminent devaluation
of the local currency. The country was at that point approaching a scheduled
change in the military cabinet running local affairs and this fuelled even
further the expectation of change. The risk factor remained high, putting a
very high premium upon the rate of interest. In such a macro context the
industrialsectorcontractedinabsolutetermsandlostshareindomesticGDP,
asTable10.5shows.
Various reasons explain why this happened. On the one hand, aggregate
demandcontractedandimportedsubstituteswereby197980alreadycrowd-
ing out local markets. The overvaluation of the exchange rate favoured such
substitution even future. On the other hand, high interest rates pushed firms
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table10.5 Argentina:industrialoutputduringthe
1970splateau(1970100)
Manufacturingoutput(atfactorcost)
Years Index %ofGFDP Annualrate(%)
.2 197579 121.6 36 1.2
198084 110.6 32.4 2.0
Source: ECLAC(1983).
Table10.6 Argentina:capitalflight(US$millions)
1979 1980 1981 1982 1983
2995 6400 6494 2563 1415
Sources: ECLAC(1986 and1987).
into abnormal levels of indebtedness, which in many cases went beyond the
actualvalueoftheirassets.
WiththechangeofcabinetinMarch1981,anewpolicyregimewasimple-
mented aiming at solving some of the more urgent short-term difficulties
faced by manufacturing industry. In spite of such actions and a series of
devaluations, industry remained stagnant, faced with a very high interest
rate. The high rate of indebtedness being the major problem entrepreneurs
were then facing, the government introduced a new medium-term financial
mechanism which made resources available to them at a low and regulated
interestrate.Inthecontextofanincreasingrateofinflation,thismechanism
ended up enforcing a massive transfer of resources to industry. Moreover, as
the state made hard currency available at an insured price it ended up by
assumingmostoftheexternaldebtpreviouslycontractedbyprivatecompan-
ies. Capital flight see Table 10.6 became the counterpart of a massive
transfer of external debt to the public sector. As we can see, this first
attempt to implement a neoliberal policy agenda in Argentina ended up in
failure and the economy went back to higher tariff protection and import
controls.
The1980s:therestructuringofmanufacturingindustry
In the course of the 1980s manufacturing production suffered a major struc-
tural transformation. The industrial sector became smaller in size, more con-
centratedandsignificantlymoreinvolvedintheproductionofresource-based
industrial commodities such as pulp and paper, steel, aluminum, petrochem-
icalsandvegetableoil.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
The restructuring towards raw material processing industries was induced
both by the set of industrial policies implemented in the early 1970s and by
trade liberalization efforts attempted from 1976 onwards. The country's pre-
viousspecializationpatterninconsumerdurablesandcapitalgoods,thatisin
metalworking production, was the result of inward-oriented industrialization
policieswhichheavilysubsidizedinvestmentinsuchfields.Ontheotherhand,
the sectoral and regional policies of the early 1970s induced a major shift
towardsthecountry'sstaticcomparativeadvantage,thatis,raw-materials-pro-
cessing industries and primary products. Table 10.7 describes this structural
transformation.
We should notice that the above process reflects both the erection of a
significant number of new, highly capital-intensive, plants during the 1980s,
andtheclosingdownofrelativelyoldproductionfacilitieswhoseperformance
wasclearlyinferiortothatofthenewplantsenteringtheeconomy.Table10.8
listssomeofthenewplantsbuiltintheearly1980s.
With the coming on-stream of the above-mentioned rawmaterials-proces-
singplantstheagestructureoftheavailablecapitalstockbecamesignificantly
youngerandlabourproductivityexpandedatafastpaceduringthelate1980s
and early 1990s, the result of the introduction of labour-saving technical
changes.
It is under such circumstances that the ISI process came to an end as
Argentina moved towards theliberalization of trade, thederegulation of mar-
ketsandtheprivatizationofpublicassetsintheearlyandmid-1990s.Numer-
ous aspects of market structure and performance have been changing in
recent years as a result of the above, but it is outside the scope of the present
papertoexaminetheminanydetail.Rather,ournextsectionconcentrateson
the examination of the policy instruments used by Argentina during the ISI
period.
Table10.7 Argentina:structureofindustrialvalueadded,1974and
1990(%)
1974 1990
Metalworking(exc.autombiles) 17.1 14.3
Transportequipment(ISIC384) 10.5 8.5
Sub-total 27.5 22.8
Foodstuffs,beveragesandtobacco 20.3 24.3
Ind.commodities(ISIC341,351,3546,3712) 16.1 22.4
Sub-total 36.5 46.7
`Traditional'(textiles,garments,leather) 36.0 30.5
Total 100 100
Source: ECLAC(1983).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table10.8 Argentina:rawmaterialprocessingplantserectedduringthefinalstagesof
theISIprocess
Industry Start-up Observations
Celluloseandpaper
AltoParana
PapelPrensa
PapeldeTucuman
1982 Newplant
1980 Newplant
1983 Newplant
Ironandsteel
Siderca 1976 1976:directreduction
1988:newplant
Acindar 1979 1979:Directreduction
1981:amalgamatedwithSantaRosaandGurmendi
1983:enlargelaminator
1989:plantrestructure
Somisa 1985 1961:firstsmelting
1974:secondsmelting
1985:newplant
Aluminium
Aluar 1978 Newplant
Uboldi 1983 NewplantPto.Madrin
Kicsa 1980 Expansionandmodernization
Petro leo
YPF 1988 ModernizationofEnsenadarefinery
Petro-chemicals
Gral.Mosconi 1974 Newplant,Ensenada
1990 OilRefiningComplex
RioTercero 1980 Newplant
Bah aBlanca 1981 Extendedplant
Polisur/Ipako 1981 Newplant
Copetro 1983 Newplant
Indupa(Bah aBlanca) 1986 Newplant
PASA 1986 Expandedplant
Petroqu micaCuyo 1988 Newplant
Source: Schvarzer(l993).
TheincentivesregimeunderlyingtheISIprocess
Duringthefourdecadescoveringtheperiod194080theexpansionofindustry
was encouraged by a complex set of instruments including tariff protection,
importpermits,fiscalincentivesfortheattractionofforeigncapital,subsidized
creditandexportpromotionincentives.Weshallnowexaminetherolethese
variousinstrumentshadinfosteringthepaceofindustrialization.
Theoveralllegalframework
TwobroadpiecesoflegislationLaw14.630of1944andLaw14.781of1958
providedtheoveralllegalframeworkofindustrialpoliciesduringthepresiden-
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
tial periods of Pero n and Frondisi, respectively. Although in both periods
policieswerehighlyselectiveintermsoftheindustrieschosenforpromotion,
the basic philosophy underlying their original design was significantly differ-
ent. Whereas the first set of policies was mainly the reflexion of the tight
militarycontrolthearmedforcesexertedonthelocaleconomyintheimmedi-
ate postwar years, and was therefore geared to the promotion of the so-called
`defenceindustries',thesecondreflectsanattemptonthepartofgovernment
tofillmissinggapsintheindustrialstructure,relatedtomanufacturingsectors
suchasoil,pulpandpaperorvehicles,wherethecountryhadaheavyexternal
deficit on the current account due to massive imports. As such, Law 14.781
constituted an effort to push the local industrial structure into a `capital-
deepening'stageinwhichforeigndirectinvestmentplayedamajorrole.
Itisimportanttonoticethatalreadyin1953thePeronistadministrationfelt
the need for a law attracting foreign private capital. In October that year it
passedLaw14.222whichwaslateroncomplementedbyCentralBankcirculars
nos 2324 and 2881, from 1956 and 1957 respectively, further specifying the
various subsidies the authorities were ready to grant to foreign companies
investing in Argentina. When compared with the impact of the 1958 Law,
noneoftheseearlylegalinstrumentspassedbythePeronistauthoritiesexerted
a significant impact upon the economy, if we are to judge by the amount of
external resources they managed to bring to Argentina. Table 10.9 provides
informationinthisrespect,whileTable10.10showsthesectoraldistributionof
theexternalresourcesreceived(Katz,1969).
There are obvious differences in the response foreign firms gave to both
pieces of legislation. No matter how much the Peronist government tried
duringitslasttwoyearsinofficetodivestitselfofitspreviousimage,itnever
actually succeeded. In the late 1950s foreign direct investment increased sig-
nificantly, rising to close to 20 per cent of total manufacturing investment,
Table10.9 FlowofforeignprivatecapitaltoArgentine
manufacturingaccordingtothelegalregime(US$000s)
Years Law Circular Circular Law Total
14.222 2324 2881 14.780
1954 2174 2174
1955 10025 10025
1956 19986 19986
1957 14727 2804 17532
1958 4127 987 7835 12860
1959 289299 289299
1960 26568 26568
1961 83834 83834
Source: Altimiretal.(1966).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table10.10 Argentina:foreignprivate
capitalbyindustry,195461(US$000s)
Industry Foreignprivate
investment
Foodstuffs 8622
Textiles 7516
Woodandfurniture 1447
Paper 3266
Chemicals 15896
Petrochemicals 38169
Machineryandequipment 29044
Vehicles 109037
Source: Altimiretal.(1966).
having been about 3 per cent at the beginning of the decade. The sectoral
distribution of the incoming resources (Table 10.10) also provides interesting
evidence concerning the pattern of industrial expansion that developed
aroundforeigndirectinvestment.
Thus,Law14.780regulatingthereceptionofforeigncapital,andLaw14.781
on industrial promotion, jointly determined the incentives regime in which
the chemical, petrochemical and metalworking expansion of the 1960s and
early 1970s was to take place. Law 14.781 continued in operation until 1973,
the year in which a newly-elected Peronist government came to office once
again.Aclearpolicyreversalwasattemptedbythenewauthoritieswhotriedto
bringbacksmallandmedium-sizeddomesticenterprisestocentre-stage,giving
highprioritytobroadobjectivesofnationaltechnologicalindependence.The
experiment, however, did not last long and after the 1976 coup d'etat which
ousted Mar a Estela Mart nez de Pero n from the presidential palace a new
industrialpromotionlawwasissuedbythemilitaryauthorities.Althoughthe
1977legislationusedmorethelanguageofmicroefficiency,takinganexplicit
standinfavouroftradeliberalizationasfarastheproductionofcapitalgoods
was concerned, it is rather difficult to isolate its impact from that of earlier
legislation. In actual fact it was during the late 1970s that many new capital-
intensive raw-material-processing plantswere erectedand brought on-stream,
mostofthembenefitingfromtariffprotectionandfiscalincentivesgrantedby
oneormoreoftheabove-mentionedlegalinstruments.
Such legislation coexisted with the so-called `provincial promotional
regimes' among which those related to Tierra del Fuego, San Luis, La Rioja,
Catamarca and San Juan were the more prominent. It was the provincial
authorities that in those cases were responsible for the application of the
law, this having become over the years a major source of fiscal tension
between the central government and the local authorities. Various studies
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
carried out by Azpiazu (1988) and PRIDE (1987) throw light upon the fiscal
cost, both at the national and provincial level, of these regional promotional
schemes.
Accordingtoboththesestudies,between1973and1987some1200firms
accounting for 2800 investment programmes benefited from promotional
regimes,eitherlocalorfromfederalsources.Thelatter were,ingeneral,fewer
but much larger investment projects than the former, with an average of
US$125000 investment per person employed. Projects receivingregional sup-
port were, on average, five times smaller in terms of investment per person
employed.Itisimportanttomentionthattwo-thirdsofthepromotedinvest-
mentprogrammesbelongedtoraw-material-processingfirmsproducingchem-
ical and petrochemical products, steel and pulp and paper. This strongly
confirmsthemajorshiftArgentinaexperiencedinitspatternofsectoraldevel-
opment during the course of the 1980s and the important role both national
and regional industrial promotion schemes played in this respect.(Katz et al.,
1986).Atamoremacrolevelitseemsworthpointingoutthatduringtheir14
years of operation these promotional schemes induced investment projects
worthsomeUS$9.0billion,onlyabouthalfofwhichactuallycameintoopera-
tion;around90000newjobswerecreatedthroughtheseprojects.Evenifatan
aggregate level such figures do not look very impressive, at the regional level
their impact has not been negligible, both in terms of direct employment as
well as in terms of externalities and synergies in less-developed regions of
Argentina. A social cost/benefit evaluation of these programmes therefore
becomes difficult to carry out, even though they can be criticized on account
ofinadequateauditingthroughouttheirimplementation.
Externaltariffprotection
Tariff protection has historically been granted on the basis of three different
criteria:first,accordingtodomesticvalueadded,ordegreeoflocalelaboration;
second, according to economic classification as final consumer goods, inter-
mediate products or capital goods; and, third, according to the existence of
domesticsupplyofthegoodsinquestion.Followingsuchcriteria,highertariff
protectiontendedtobegrantedtoconsumptiongoods,lowerforintermediate
products and even lower for capital goods. On the other hand, products not
produced locally tended to be less protected than those where there was
domestic supply. The tariff structure was therefore far from neutral as it
provided stronger incentives for the domestic production of final
products and much lower incentives for the production of capital goods. In
spite of the above it is important to note (Lucangeli, 1989) that Argentina's
industrial structure of 1969 was considerably more protected than the one
of 1977 and the latter more than the one corresponding to 1985. Lucangeli
further reports that he has identified through the measurement of effective
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
rates of protection that there was a generalised reduction in such rates
as a reflection of major productivity gains throughout the production
structure. Clearly the process was led by the metalworking and chemical
industries which were the ones that grew faster in terms of total
productivity.
Inadditiontothislong-runreductionintariffprotection,studiescarriedout
by Nogues (1978) show that there was a large gap between the legal and the
real effective tariff. Such a high degree of `redundancy' in tariff protection
allowed the government to reduce nominal tariffs without much affecting
behaviour at the individual firm level. This actually happened in 1976 as we
shallnowsee.
Probably the first attempt ever to introduce some order into the somewhat
chaotictariffstructureprevailinginArgentinawasthatof1967,underKrieger
Vasena. In 1976 Martinez de Hoz following Vasena's logic went deeper into
reformingthetariffsystem.Wecanidentifytwodifferentstagesinthepolicies
followedbythegovernmentasfarastheexternalsectorisconcernedafterthe
military takeover of 1976. The first stage, called fase de apertura, covers the
period197681.FromApril1976tolate1978ratesofprotectionwerereduced
and quotas and other quantitative restrictions were eliminated. By the end of
1976 theaveragelegalrateofprotectionhadfallenfrom94percentto53per
cent. The average was once again reduced in 1979, this time to 26 per cent,
togetherwithasignificantreductionintheinterindustryvariance.Acalendar
of further reductions was then announced by the authorities, stating that the
expected average tariff protection level for 1981 was 15 per cent. The macro-
economic adjustment policy did not, however, attain price stability and the
governmenttriedtouseimportstodisciplinedomesticprices.Forthispurpose
it brought forward its schedule of tariff reductions. The combination of
exchangerateappreciationandtradeliberalizationresultedinamajorcurrent
account deficit in 1980, after four successive years of surplus. Massive capital
flightfollowed,aswasshowninTable10.6.
The second stage of the programme after the resignation of Mart nez de
Hoz (called fase de ajuste externo) covers the period 198288. External and
domesticcircumstanceshadchangedquitestronglyandthesearchforexternal
equilibriumwascarriedoutonthebasisofanentirelydifferentpolicypackage.
The country was now facing a drastic shortage of external funds after the
Mexicanmoratoriumof1982.Itwasalsosufferingalargedeficitonitscurrent
account. The policy option was to go backwards as far as trade liberalization
was concerned, increasing once again the rate of external protection, reestab-
lishing import quotas and exchange control, after a major currency devalua-
tion.Forallpracticalpurposestheeconomywasonceagainclosedtoexternal
competition,thoughatthispointnon-tariffbarrierswerefrequently`jumped'
through import permits granted by the very many national and provincial
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
industrialpromotionmechanismswhichallowedfirmstoimportcapitalgoods
and intermediate productson atariff-free basis.The movementtowards trade
liberalizationandmarketderegulationbeganonceagaininthelate1980sand
early 1990s, but a detail examination of the results must remain outside the
scopeofthischapter.
AsfarastheISIperiodisconcerned,wehavetoclosethepresentdiscussion
by stating thatalong-term trendof falling external ratesof protectionand of
rationalizationofthetariffstructurecanbenoticedwhenweexaminegovern-
ment actions on this front over the 40-year period covered by the present
enquiry. It does not seem correct to argue that there was a lack of perception
onthepartofthelocaleconomicauthoritiesregardingthelong-termrelation-
shipbetweentotalfactorproductivitygrowthandthedegreeofexternaltariff
protectionneededbytheeconomy.Onthecontrary,inthelate1970sandearly
1980swecanidentifyattemptsatpolicyinterventionsaimedatgettingridof
tariff redundancy after a long period of steady growth in total factor produc-
tivitythroughoutthe1960sandearly1970s.However,thesimultaneousincid-
enceofmacroeconomicuncertainty,exchangerateappreciationandalackof
external financing condemned the initial attempts at trade liberalization to a
dramaticfailure.
Foreigndirectinvestment
It is only towards the end of the 1950s and throughout the 1960s that local
subsidiariesoftransnationalcorporationsbegantoplayasignificantroleinthe
process of manufacturing growth in Argentina. In a short period of just five
years between 1958 and 1963 some 200 foreign companies opened local
productionfacilities.InspiteofthefactthatmanyofthembroughttoArgen-
tina secondhand machinery already depreciated in their home country, it is
also true that they brought with them technologies for the organization of
production,qualitycontroltechniquesandsubcontractingpracticespreviously
unheard of in the domestic environment. Their rapid expansion catering for
the domestic market explains why their share in manufacturing GDP moved
upwardsfromaroundone-fifthin1955toaboutone-thirdtowardstheendof
the 1960s. Studies carried out by Sourrouille et al. (1985) and by Azpiazu and
Kosacoff(1985)showthatdomesticsubsidiariesofMNCsarecharacterizedby
larger plants than their locally-owned competitors, as well as more modern
process-engineering and organizational technologies. They therefore have
higher labour productivity than domestic firms, and have access to external
andinternalsourcesoffinancealmostentirelyinaccessibletolocalfirms.Many
ofthesefirmsopeneddomesticproductdesignandprocess-engineeringdepart-
ments with the specific purpose of generating incremental units of techno-
logical know-how aimed at improving their local productivity performance
andenhancingthequalityoftheirproducts,andsoon.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
A rapidly expanding degree of social unrest, a growing perception that
domesticdemandwasreachingaplateauafteritssignificantexpansioninthe
1960s, and increasing signs of macroeconomic instability and turbulence
account for the fact that foreign direct manufacturing investment slowed
down in the late 1970s, with many foreign companies such as Squibb Inc.,
General Motors, Olivetti, Ely Lilly and Citroen actually leaving the country
duringthe1980s.
Theregulatoryenvironmentandtheincentivesregimebecamefavourableto
foreignfirmsonceagaininmorerecentyearswhenthecountrymovedbackto
tradeliberalizationandthederegulationandprivatizationofproductionactiv-
ities during the 1990s. It is, however, outside the scope of this chapter to
examineindetailsuchrecenttrends.
Thefinancingofindustrialexpansion
ThehistoryofArgentineindustrializationrevealsmajordifficultiesinthelinks
between manufacturing firms and the sources of finance for their long-term
development.Twomajorissuesemergehereaspermanentfeaturesofthelocal
landscape:ontheonehand,thelackofalong-termstrategyspecifyingastable
set of criteria channeling domestic and foreign savings into manufacturing
investment, and, on the other, the almost complete absence of medium- and
long-termcapitalmarkets.
From a global perspective we can identify two quite different stages in the
financing of manufacturing activities, and the break-point between the two
stagescanbelocatedinthelate1970swhenthemilitarygovernmenttriedto
implement a drastic macroeconomic stabilization programme based on the
monetary approach to the balance of payments. Let us briefly examine both
periods.Thefirstbeganinthe1940swiththecreationoftheBancolndustrialin
1944andthefinancialreformof1946.Thelatterinvolvedthenationalization
ofbankdepositsandtheestablishmentofcreditlinesforvariousmanufactur-
ingactivities.ObviouslyArgentinahadareasonablystrongdomesticfinancial
systempriortothis,butitwasoneinwhichtherewasnolong-termcredit,so
investment programmes frequently had to be financed through short-term
borrowing.
VariousArgentinescholarshaveexaminedthemajorfeaturesofthefinancial
model underlying the process of industrialization in Argentina during this
period. Among them, Altimir et al. (1966), Itzcovich and Feldman (1969),
Brodersohn (1972), Schvarzer (1982), and Guadagni (1972) looked at the net
financialpositionofmanufacturingenterprisesandarrivedatasetofrelevant
conclusionswhichwenowsummarize.
On the one hand these studies show that manufacturing firms invested
more than they saved,becoming therefore net debtors vis-a-vis therest of the
economy.Giventheprevailingnegativerateofinterestthisclearlyrepresented
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
a net subsidy in their favour. On the other hand, the studies also show that
financinghasbeenbasicallyofashort-termnaturemoreusefulforsupporting
access to working capital than for investment in machinery and equipment
and that long-term resources have been almost absent from local capital
markets. The recurrent renewal of short-term borrowing thus replaced the
lackoflong-termfunds,butinvolvedahigherdegreeofuncertaintyandlarger
coststhaniflong-termborrowinghadbeenavailable.
Theabove-mentioned studieshavealsoshownthat credit fromsubcontrac-
tors and suppliers were additional important sources of funds for industry,
particularly so for small and medium-sized enterprises. Large enterprises and
conglomerates had easier access to banks, and this in circumstances in which
the rate of interest was negative and well-below the rate of inflation. Yet
another feature of the local capital market was the almost complete absence
ofamarketforcommercialpaperandbonds.Thefactthatmanyenterprises
even large ones were still family-run and therefore tightly controlled by a
smallnumberofagentsmadethemlesspronetosearchforfundsthroughthe
local stock exchange. Reinvestment of profits constituted the major source of
capitalization, and inflation made companies rather sceptical of building up
reservesandprovisionsforfutureinvestment.Self-financingthereforeconstitu-
tedthemajorsourceofresourcesfortheexpansionofproductivecapacity.
Another source of finance whose role expanded in the 1960s was indebted-
nesswiththepublicsector.Firmsgrewaccustomedtoleavingunfulfilledtheir
social security and income-tax obligations. Given the high rate of inflation,
thatmeantyetanotherformofresourcetransferwithinthesociety.
Asfrom1977thesecondstagebegan.Thefinancialmodelunderlyingman-
ufacturing production and growth suffered significant changes. The liberal-
ization of the interest rate, the elimination of rediscount lines from the
Central Bank specifically aimed at financing manufacturing production, the
openingofcapitalmarketstoexternaloperators,andtheroleplayedbysectoral
andprovincialindustrialpromotionschemesassourcesoffinanceforindustry
constitutesomeofthemoreoutstandingfeaturesoftheemergingnewregime.
Therulesofthegamechangedquitedramaticallyafterthefinancialreformof
1977, and particularly so after the government adopted a monetary approach
tothebalanceofpaymentslatein1978.Completemobilityoffinancialassets
inacontextinwhichthelocalrateofinterestwasabovetheinternationalrate
induced a systematic inflow of resources. At the beginning of the 1980s the
industrial sector had accumulated a major stock of debt with the financial
sector, in many cases greater than the actual value of the physical assets
ownedby thefirms, thussignallingthatthedefaultofthelatter wasgoingto
takethefinancialstructurewithit(Feldman,1983).
The domestic interest rate was well above the current profit rate of manu-
facturingenterprises,forcingmanyofthemtooperatemoreasfinancialentities
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
speculating on short-term financial assets rather than worrying themselves
withquestionsofproductionorganizationoroftechnologicaldevelopment.
The impact of the above upon the industrial structure was large and dra-
matic.Ontheonehandinvestmentfellquiteconsiderably,notevencovering
the ordinary depreciation costs of equipment and machinery. Maintenance
and replacements schedules were very imperfectly kept and the operating
capacity of industry deteriorated badly. On the other hand, a high positive
interestrateforcedfirmstodrasticallychangetheorganizationofworkandthe
handlingofinventoriesattheshopfloorlevel.
Thefailureofthetradeliberalizationandfinancialderegulationpolicypack-
ageendedwithamassivecapitalflightearlyin1981,withmanufacturingfirms
heavilyindebtedwiththefinancialsector,andthestatenowbeingforcedtopick
upthewreckageofthefinancialstructure.Bymid-1982theeconomywasback
in a new spell of import-substitution industrialization once tariffs had been
increasedagainandtheprivatedebthadbeentransferredtothepublicsector.
Thepromotionofmanufacturingexports
One of the more notorious features of the Argentine process of industrializa-
tion has been its inward-orientation. Up to the mid-1960s non-traditional
manufacturing exports (that is, excluding exports from meatpacking houses,
sugar and flour mills and of vegetable oil producers), accounted for less than
U$S100 million annually, that is less than 5 per cent of total exports. This
structural feature was clearly identified by Rau l Prebisch in his report on the
Argentine economywritten inthemid-1950s (ECLAC, 1958b). Export incent-
ives began to be used by Argentina in 1962, and about a decade later, in the
early 1970s, manufacturing exports with an industrial origin had grown to
US$1000 million per annum and represented close to one-quarter of total
exports.Variousstudies(Monti,1978;Felix,1964;AblinandKatz,1977)have
triedtoaccountforthesourcesofsuchexpansion.Thelatterstudyinparticular
assignedspecialexplanatorypowertotheaccumulationofdomestictechnolo-
gical capabilities and skills on the part of local industries, particularly in the
period1960throughtothemid-1970safterthearrivalinArgentinaofalarge
number of multinational corporations which brought with them somewhat
superiorproductionandorganizationaltechnologies.Productivitygrowthwas
quite significant throughout this period, allowing many firms local and
foreign to attain rapid progress, eventually achieving export capabilities
after partially closing the gap with the world's technological frontier. Ablin
andKatz'sstudy,however,alsorecognisedthefactthatotherimportantinsti-
tutional factors explaining the expansion of manufacturing exports are the
various financial and tax incentives used by the government to induce such
activities, as well as bilateral agreements and treaties signed by Argentina
within the framework of LAFTA particularly with Cuba and the Andean
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
countries. Also, intracompany arrangements allowing local subsidiaries to
exportwithintheglobalframeworkoftheirrespectivecorporationsconstituted
animportantexplanationoftheexpansionofindustrialexports.
Export promotion was originally confined to tax exemptions and financial
incentives,suchasdrawbacksandexportcredits.Duringthe1970s,however,a
muchwidersetofinstrumentswasemployedincorporatingotherobjectivesas
well,suchas,forexample,thatofinducingtheexpansionofcertainregionsor
of using particular docks and shipping facilities. Each one of these incentives
operatedseparately;inconjunction theyconstitutedaconfusingweboflaws,
administrative resolutions and so on, difficult to understand and evaluate for
cost-efficiency.
However, the role of export incentives was highly qualified by the macro
performanceofArgentina.Manyoftheseinstrumentswerefrequentlyofonly
nominal value given the financial restrictions under which the public sector
becameaccustomedtooperate.
Between 1973 and 1984 manufacturing production once again became
strongly inward-looking, and only 6 per cent of total industrial output was
exported during these years. More than 85 per cent of total manufacturing
productionoriginatedinsectorsinwhichlessthan10percentoftotaloutput
was exported; and if we look at the individual firm level we notice that only
about 150 companies exported more than 25 per cent of their output. The
structure of industrial exports also changed quite considerably. Within the
resource-processing industries, vegetable oil exports expanded rapidly while
exports from the meatpacking sector contracted quite significantly. As far as
manufacturing industry was concerned, industries related to metalworking
production saw their participation in exports fall, while chemicals and petro-
chemicals, aluminum, steel, pulp and paper and fishmeal expanded their
exportcapabilities.Paripassuwiththerestructuringofmanufacturingactivities
towardsnatural-resource-processingindustries,thestructureofexportsgradu-
allyshiftedinthesamedirection.Bycontrast,exportscontractedquitesignific-
antly in leather goods, consumer durables, printing, agricultural machinery
and machine tools, all sectors in which domestic production capacity was
reducedduringthecourseofthe1980s.
TheISIstrategyinretrospect
Thepurposeofthischapterhasbeen,firstly,todescribetheperformanceofthe
Argentine economy (and of its manufacturing sector in particular) during the
period covered by the import-substitution industrialization process, that is
fromtheearly1940stothelate1980s.Secondly,wehaveattemptedtoidentify
the particular policies and institutions that emerged during these 50 years of
Argentinehistory.Thirdly,wehavetriedtoevaluatetheconsequencesofISIin
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
termsofthecountry'sgrowthrecord,oftheobservedpatternofdistributionof
the resulting benefits, and of the development of domestic technological and
institutionalcompetencesusefulforfutureeconomicexpansion.Theprevious
sections having examined the first two sets of topics, we now turn to a final
evaluationoftheresultsattained.
Industrializationinaninward-lookingenvironment
A balanced account of the pros and cons of ISI efforts in Argentina is rather
difficulttoachieve,althoughitisquiteobviousthatthecountryhasnotdone
aswellasotherNICsduringthepostwarperiod.Thisisnotjustincomparison
withKoreaorTaiwan,butalsowithrespecttoBrazilorMexicowhicharemore
comparablecases.Further,ithasalsonotdonewellevenwithrespecttoitsown
growth potential. Stopgo cycles, dramatic episodes of hyperinflation, capital
flight and unemployment of skilled human resources all constitute dramatic
proofofunfulfilledexpectations.
An inward orientation probably constitutes part of the explanation of why
the long-term performance of Argentine manufacturing was so clearly disap-
pointing.Notbeingsubjecttotheacidtestofexternal`contestability'probably
madedomesticentrepreneurslesseagertosearchfornewproductdesignsand
newproductionprocessesonthebasisofwhichtocompetebothdomestically
andinternationally.Highlyconcentratedsellers'marketsinthedomesticenvir-
onmentclearlypushedinthesamedirection.Inspiteoftheabove,however,it
does not seem correct to argue that the ISI process in Argentina constitutes a
clearexampleoffailure.Letusseewhy.
Diditreallygowrong?
During the 40 years under examination many Argentine firms and industries
managed to accumulate a vast stock of technological, organizational and
managerialcapabilitiescrucialforfutureexpansion.Alargeurbanskilledmid-
dleclassemergedassociatedwiththedevelopmentofindustryandasophisti-
cated industrial culture diffused throughout most corners of the local society
pari pasu with the process of industrialization. Working habits, standards and
normsandavaststockofengineeringroutineswereassimilatedbyallwalksof
society adding up to a massive `social capital' of invaluable significance for
futuredevelopment.
Furthermore, the 15 years covering the period 196074 appear as a particu-
larly successful moment in which manufacturing output, labour productivity
andmanufacturingexportsallexpandedatafastpace,asFigures10.2and10.3
show. The particularly good performance of the industrial sector throughout
thisperiodexplains whyitsrelativeshareofdomestic GDPincreasedfrom24
percentin1964to28percentin1974.Manufacturingexportsofanincreasing
degree of technological complexity also expanded very rapidly over these
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
240
220
200
180
160
140
120
100
1960 1961 1962 1963 1964 19651966 1967 1968 1969 19701971 1972 19731974
80
Product
Employment
Labour productivity

I
n
d
e
x

1
9
6
0

=

1
0
0
Years
Figure 10.2 Argentina: performance of the industrial sector, 196474
Sources: Banco Central de la Repu blica, Producto bruto industrial, Argentina. Employment
data from Beccaria (1989).
1964 1974
0
500
1000
1500
2000
3000
3500
4000
2500
U
S

d
o
l
l
a
r
s

m
i
l
l
i
o
n
s
Total
Industrial
Manufactures
Figure 10.3 Argentina: total exports, industrial andmanufacturingexports, 1964and1974
Source: ECLAC (1986).
years, showing that industry was gradually searching for new forms of inter-
nationalization. Furthermore, as Table 10.11 shows, exports of `pure' techno-
logy under the form of manufacturing facilities sold by local entrepreneurs on
a `turnkey' basis in other Latin American countries also provide an indication of
a growing degree of technological maturity on the part of local entrepreneurs.
However, the long-term growth performance of industry deteriorated badly
as the economy entered a period of high social unrest and macro turbulence in
the second part of the 1970s, and particularly so throughout the 1980s when
308 ISI in Argentina, 194080
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table10.11 Exportsof`pure'technologybyArgentinefirmsunderthe
formofturnkeyproductionfacilitiessoldinLatinAmerica
Technology- No.of Valueofsales Average Percentage
purchasing contracts (US$) contractvalue oftotal
country exports
Cuba 8 50075612 6259452 14.70
Bolivia 7 63940512 9134358 18.77
Uruguay 4 1547102 386775 0.45
Paraguay 3 52278671 17426223 15.33
Chile 3 3230625 1076875 0.95
Ecuador 2 2677157 1338578 0.79
Venezuela 2 252500 126250 0.07
Peru 1 120000000 35.22
Brazil 1 200000 0.06
Mexico 1 90000 0.03
Honduras 1 450000 0.13
Ivorycoast 1 46000000 13.50
Totals 134 340742179 100.0
Source: AblinandKatz(1977).
the country was almost completely cut off from international sources of
finance. It is difficult to blame the ISI process for the dramatic contraction
of GDP that occurred in the 1980s. A high degree of social and economic
uncertainty prevailed in every other aspect of Argentine life, inducing manu-
facturing firms (as well as ordinary citizens) into speculative and rent-seeking
activities. But it would take a great stretch of the imagination to argue that
suchanoutcomewastheconsequenceofimport-substitutingindustrialization
policies.
Looked at from this perspective, and considering the long-term conse-
quences of ISI in the case of Argentina, we come to a somewhat different
conclusion from that advanced by mainstream neo-classical economists. The
latterputmoreweightonstaticinefficienciesinresourceallocationandtendto
disregard dynamic aspects associated with the long-term accumulation of
domestic technological and institutional capabilities. Our own description of
thefactsputsusclosertoSchumpeterthantoPareto, eventhoughwedonot
denythatthelong-termbenefitscouldprobablyhavebeengreaterhadArgen-
tina not insisted for so long on an inward orientation, or rejected for so long
theblessingsofitsrichnaturalresourcebase.
Notes
1 Schvarzer (1996) has recently published a rich account of the early days of the
Argentineindustrializationprocess.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
2 The Atomic Energy Commission was created in 1951. During the war years the
Argentinemilitaryforcesentertainedtheideaofdevelopingtechnologicalself-suffi-
ciencyintheatomicenergyfield.Itwasonlyinthe1960sthatthecountrydecidedto
buildupnuclearpowerstationsforelectricitygeneration.Thefirstsuchunitwasput
intooperationin1973andthesecondin1980.
3 Afternearlyfourdecadesofrapidexpansionwhichbroughttheareaundercultiva-
tiontonearly20millionacresin193840andtotalgrainoutputto20milliontons
theprimarysectorenteredalongperiodofdecayandstagnationthatlastednearly30
yearsandconstitutesone of the major `blackspots' in Argentineeconomichistory.
Thelong-lastingconfrontationbetweenPeronismandtheSociedadRuralArgentina
constitutesacentralchapterofthenationalsocialhistorywhichhelpsexplainwhy
Argentina did so much worse than either Australia or Canada in terms of mechan-
ization,technologyacquisitionandproductivitygrowthintheprimarysector.Inthe
early 1950s the area under cultivation had fallen to less than 15 million acres and
totalagriculturaloutputwasroughlyequivalenttothe191920level.Itwasonlyin
thelate1960sthattheagriculturalsectorofArgentinareturnedtoits193940output
levels, with 20 million acres under cultivation and production at about 23
million tons. In the prewar years Argentina's annual exports of wheat reached
some 3.3 million tons and represented about 20 per cent of total world
exports. Canada with 4.8 million tons accounted for 28 per cent of the total, Aus-
tralia with 2.8 million tons absorbed 16 per cent and the USA another 7 per cent.
From1950to1954,withamuchhigherworlddemandforwheatArgentineexports
had fallen to 2.2 million tons which constituted only 9 per cent of total world
exports, whereas Canada had doubled its exports and the US exports expanded
from 1.3 to 8.9 million tons. In regard to maize, in the prewar period Argentina
exported 6.3 million tons, that is 64 per cent of world trade. From 1950 to 1954
exports fell to just over one million tons per annum. Thirty years were lost in the
fruitless confrontation between rural landlords and Peronist officials ! (Katz and
Bercovich,1993).
4 This is only (!) forty years after Australia created the Commonwealth Scientific
and Industrial Research Organization (CSIRO) in 1916 with the purpose of
improving the performance and modernity of its primary sector. This is a
difference in institutional behaviour which should be kept in mind when we try
to account for the rather disappointing long-term performance of the Argentine
economy.
5 ECLAC(1959b).
6 SeeMallonandSourrouille(1975).
7 In response to import prohibition in the field of penicillin salts which could
have resulted in the establishment of local competitors in the antibiotic markets,
SquibbCo.decided,forexample,tobuildupafairlymodernresearchanddevelop-
ment laboratory where highly trained Argentine biologists carried out research
activities which were relatively close to the world's knowledge frontier of the
time. A future Argentine Nobel Prize winner in chemistry Dr F. Leloir was at
some point part of the staff of this laboratory. There is no doubt that Squibb
imagined at that point a much more promising future for its venture in Argentina
thanactuallydeveloped.Thisexamplecanbe replicatedinotherareasofmanufac-
turingproduction.
8 SeeAdolfoCanitrot(1983).
9 SeeCanitrot(1983),Schvarzer(1983),Sourrouille(1985)andRodr guez(1979).
10 SeeBerlinsky(1977),Nogues(1978).
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
References
Ablin,E.and Katz,J. (1977) `Tecnolog a y exportacionesindustriales:unanalisismicro-
econo mico',DesarrolloEconomico vol.17,no. 65, BuenosAires: Instituto de Desarrollo
EconomicoySocial.
Altimir, O. et al. (1966) `Los instrumentos de promocio n industrial en la posguerra',
DesarrolloEconomico,vols.6 and7,BuenosAires:InstitutodeDesarrolloEconomicoy
Social.
Azpiazu, D. (1988) La promocion a la inversion industrial en la Argentina. Efectos sobre la
estructuraindustrial,194787,workingpaperno.27,BuenosAires;ECLAC.
Azpiazu,D.,andKosacoff,B.(1985)LasempresastransnacionalesenlaArgentina,Working
paperno.16,BuenosAires:ECLAC.
Azpiazu, D. (Bisang, R. and Kosacoff, B. (1987a) Exportacion de manufacturas y desarrollo
industrial. Dos estudios sobre el caso argentino, 197384, working paper no. 22, Buenos
Aires:ECLAC.
(1987b)ExportacionesIndustriales:Perfilycomportamientodelasempresasexportadoras
de manufacturas, Serie Estudios no. 6 Instituto Nacional de Estadisticas y Censos:
BuenosAires.
BancoCentralde laRepu blica de Argentina(1976) CuentosNacionales, Series Histo ricas,
BuenosAires.
Beccaria, L. (1989) Industrializacion, mercado de trabajo y distribucion del ingreso, Buenos
Aires:ECLAC\LC\BUE\R.146
Berlinsky, J. (1985) `La proteccio n efectiva de actividades seleccionadas de la industria
manufacturera argentina', Working paper, Instituto Torcuato Di Tella no.119, Buenos
Aires,December.
(1977)Proteccio narancelariadeactividadesseleccionadas delaindustriamanufac-
tureraargentina,BuenosAires:MinistryofFinance.
Braun,O.andLeonard,J.(1968)ÀModelofEconomicStagnation.ACaseStudyofthe
Argentine Economy', The Economic Journal, no. 312 reprinted in Desarrollo Economico,
vol.20,no.80,BuenosAires:InstitutodeDesarrolloEconomicoySocial.
Brodersohn, M. (1972) Financiamiento de empresas privadas y mercados de capital, Buenos
Aires:InstitutoLatinoamericanodelaIntegracion.
Canitrot, A. (1983) `La experiencia populista de redistribucio n de ingresos', Desarrollo
Economico,vol.15,no.59,BuenosAires:InstitutodeDesarrolloEconomicoySocial.
Damill,M.andFanelli,J.M.(1987)Evoluciondelasestructurasfinancierasylasdecisionesde
inversion de cartera de las empresas en el per odo 197985, Buenos Aires: Banco de la
RepublicaArgentina,CentrodeestudiosdeestadoySociedad.
D az Alejandro, C. (1970) Essays on the Economic History of the Argentine Republic, New
Haven,Conn.:YaleUniversityPress.
Dorfman,A.(1983)CincuentaAnosdeIndustrializacionenlaArgentina,19301980,Buenos
Aires:EdicionesSolar.
ECLAC(1987)`Notasobrelaevolucio delaeconom aargentinaen1986',LC\BUE\L105,
BuenosAires.
(1986)`Notasobrelaevolucio n de laeconom aargentinaen1985', LC\BUE\L101,
BuenosAires.
(1983) Indicadores macroeconomicos de la Argentina, serie trimestral perio dica from
1983,BuenosAires.
(1959a) Ànalisis y proyecciones del desarrolloecono mico: no. V. El desarrollo eco-
no mico de la Argentina', Las Inversiones Extranjeras en la Argentina, n.u. 59, II G3,
Mexico:UnitedNations.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
(1959b)ÀnalisisyProyeccionesdelDesarrolloEcono mico:V.ElDesarrolloEcono m-
ico de la Argentina', Los Sectores de la Produccion, N.U. 59, II G3, Mexico: United
Nations.
Feldman, E. (1983) `La crisis financiera argentina, 198082: Algunos comentarios',
Desarrollo Economico, no. 91, Buenos Aires: Instituto de Desarrollo Economico y
Social.
Felix, D. (1964) Ìndustrializacio n de importaciones y exportaciones industriales en
Argentina',WorkingPaper,BuenosAires:InstitutoDiTella,CentrodeInvestigaciones
EconomicasdelInstitutodiTella.
Fernandez, R. (1983) `La crisis financiera argentina: 198082', Desarrollo Economico, no.
89,AprilJune.
Gatto, F. (1995) `Las PYMES argentinas y su insercio n exportadora', in B. Kosacoff (ed.),
HaciaunaNuevaEstrategiaExportadora,UniversidadNacionaldeQuilmes.
Guadagni, A. (1972) Las tasas de interes negativas y la distorsion de la inversion real en la
Argentina,BuenosAires:InstitutoLatinoamericanodelaIntegracion.
Itzcovich,S.andFeldman,E.(1969) Unsistemade transacciones financieras parala Argen-
tina,BuenosAires:InstitutoDiTella.
Katz,J.(1986)Ùnaempresaargentinaproductorademaquinas-herramienta',inJ.Katzet
al., Desarrollo y crisis de la capacidad tecnologica latinoamericana, Buenos Aires: BID\E-
CLAC\CIID\PNUD.
(1969) Production Functions, Foreign Investment and Growth: A Study Based on the
ArgentineManufacturingSector,Amsterdam:NorthHolland.
(1968) `Verdoorn Effects, Returns to Scale and the Elasticity of Factor Sustitution',
OxfordEconomicPapers,July.
Katz, J. et al. (1986) Desarrollo y crisis de la capacidad tecnologica latinoamericana, Buenos
Aires:BID\ECLAC\CIID\PNUD.
KatzJ.andKosacoff,B.(1989)Èlprocesodeindustrializacio nenlaArgentina:evolucio n,
retrocesoyprospectiva',BuenosAires:CentroEditordeAmericaLatina.
(1984) `Multinationals from Argentina', in S. Lall (ed.),The New Multinationals: The
SpreadofThirdWorldEnterprises,Chichester:Wiley.
Katz, J. and Bercovich,N. (1993) `NationalSystems of InnovationSupporting Technical
Advance in Industry: The Case of Argentina', in R. Nelson (ed.), National Innovation
Systems:AComparativeAnalysis,Oxford:OxfordUniversityPress.
Kosacoff, B. (1984) El proceso de industrializacion en la Argentina, working paper no. 13,
BuenosAires:ECLAC.
n,crecimientoydesempen Lucangeli,J.(1989)`Proteccio oindustrial',Bolet nInformativo
Techint,BuenosAires.
Llach, J. J. (1984) Èl plan Pinedo de 1940. Su significado histo rico y los or genes de la
econom a pol tica del peronismo', Desarrollo Economico, vol. 23, JanuaryMarch, Bue-
nosAires:InstitutodeDesarrolloEconomicoySocial(IDES).
Mallon, R. and Sourrouille, J.V. (1975) La pol tica economica en una sociedad conflictiva,
BuenosAires:AmorrortuEditores.
Monti,A.(1978)`Lasexportacionesmanufacturerasargentinas',inPol ticasdePromocion a
lasExportaciones,part4,Santiago:ECLAC.
Nogues, J. (1978) `Proteccio n nominal y efectiva: impacto de las reformas arancelarias
durante19767',EnsayosEconomicosno.8,BuenosAires:BancoCentraldelaRepublica
Argentina.
O'Connell, A. (1984) `La Argentina en la depresio n: los problemas de una econom a
abierta', Desarrollo Economico, vol. 23, no. 62, Buenos Aires: Instituto de Desarrollo
EconomicoySocial.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
PRIDE, joint programme between CFI and ECLAC, Buenos Aires, on aspects of reindus-
trializationanditsregionaldeterminants(15studies).
Rappaport, M. (1976 `La pol tica britanica en la argentina a comienzos de la decada de
1940', Economico, vol. 16, no. 62, Buenos Aires: Instituto de Desarrollo Economico y
Social.
Rodriguez,C.(1979)Elplanargentinodeestabilaziciondel20dediciembre,Workingpaper
no.5,BuenosAires:CentrodeEstudiosMacroecono micasdeArgentina(CEMA).
Schvarzer,J.(1996)Laindustriaquesupimosconseguir,BuenosAires:EditoriaPlaneta.
(1995)`Grandesgruposecono micosenlaArgentina.Formasdepropiedadylo gicas
de expansio n',in P. Bustos (ed.),Mas alla de la estabilidad, Buenos Aires:Fundacio n F.
Ebert.
(1993)Èxpansio n,maduracio nyperspectivasdelasramasbasicasdeprocesosenla
industriaargentina.Unamiradaexpostdesdelaeconom apolitica'DesarrolloEconom-
ico,October.
(1983) `Cambios en el liderazgo industrial argentino en el per odo de Mart nez de
Hoz, Desarrollo Economico, vol. 23, no. 91, Buenos Aires: Instituto de Desarrollo Eco-
nomicoySocial.
(1982)ÈlBancoNacionaldeDesarrolloyeldesarrollotecnolo gicoenlaArgentina',
BuenosAires:CentrodeInvestigacionesSocialesSobreelestadoylaAdministracion.
Sourrouille, J.V. et al. (1985) Transnacionalizacion y pol tica economica en la Argentina,
BuenosAires:CentroEditordeAmericaLatina.
Thorp, R. and Eshag, E. (1965) Èconomic and Social Consequences of Orthodox Eco-
nomic Policies in Argentina in the Post-war Years', Bulletin of the Oxford University
InstituteofEconomicsandStatistics,vol.27,no.1.
Villanueva, J. (1972) Èl origen de la industrializacio n argentina', Desarrollo Economico,
vol.12,no.47,BuenosAires:InstitutodeDesarrolloEconomicoySocial.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
11
ImportSubstitution,Economic
IntegrationandtheDevelopmentof
CentralAmerica,195080
IsaacCohen*
Introduction
Ithasbeendifficultforobserverstorecognizethatthethreedecadesfrom1950
to 1980 were among the most intense and profound periods of change in
Central America's history.
1
We can reflect on how to interpret this difficulty
with the help of Albert Hirschman. Observing the same difficulty with recog-
nizing change in Latin America, Hirschman identified various reasons which
mightexplainit(Hirschman1987,pp.736).
First, based on Hegel's òwl of Minerva', Hirschman holds that this is a
consequence of à fairly general human trait', by which acknowledgement of
`the positive features of a period' takes place ònly when we enter into a
subsequentone,whosetroubleswenowwishtounderlinebypaintingastrong
contrast with what camebefore'. Finally, when`celebration' comes, itdoes so
only because it coincides `with the lamentation about its passing'. Second,
HirschmanfindsthatthishumantraitisintensifiedinLatinAmerica,because
of a `reluctance to celebrate or even acknowledge progress while it unfolds
before our eyes'. This Latin American trait, which Hirschman elsewhere bap-
tisedas`fracasomania',
2
leadstoàconspiracyofsilenceaboutthegoodnews'.
Initsturn,thisisaggravatedbya`tacticallyinspiredtendencytounderlinethe
negativesoastopresentthecontinentasalong-victimizedclaimantuponthe
internationaleconomicandfinancialsystem'.
All the reasons identified by Hirschman as leading to areluctance to recog-
nize the positive, are present in Central America, but perhaps in an extreme
* FormerDirector,WashingtonOffice,UnitedNationsEconomicCommissionforLatinAmericaand
the Caribbean (ECLAC). The opinions expressed here do not necessarily coincide with those of the
institution.
314
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 315
degree.AsintherestofLatinAmerica,orelsewhere,changewascontradictory,
non-linear and unbalanced. The difference in Central America was that the
economicdownturncamewithanintensificationofColdWarconflicts,result-
ing in one of the most convulsive periods of the region's history. Thus it is
peculiar to Central America that this most intense period of economic and
socialchangeofitscontemporaryhistoryendedsodramatically.However,the
upheavalmayhaveapositiveconsequence.Afterpayingaheavyprice,Central
Americansocietiesarenow,itistobehoped,morecapableofaccommodating
and absorbing change in a more positive and less convulsive manner. On the
other hand, if history offers any guidance, the danger is always there that
changewillcometothisregiononlyataheavyprice.
This chapter describes the changes that took place in Central America
between 1950 and 1980, how it was that this most intense period of change
cameaboutandhowitendedinamostconvulsiveperiodoftheregion'shistory.
Finally, inappropriately perhaps for a work on economic history, it explores
someperspectivesthathaveemergedasalegacyoftheseconvulsiveyears.
3
Theboom
Perhapsthemostimpressivechangewasdemographic.Thepopulationofthefive
countrieswentfromeightmillionpersonsin1950tomorethan20millionin
1980. This quantitative change came with a profound transformation in the
structureofthepopulation,sincetwo-thirdswasruralin1950andmorethan
halfwasurbanby1980.However,thesefiguresdonotrevealthewholepicture,
becausein1950two-thirdsoftheruralpopulationwasscatteredthroughoutthe
region, with about 90 per cent living outside the limits of any administrative
centre.Theonlyurbanconcentrationsthatcouldbesocalledin1950werethe
capitalcities,GuatemalaCitybeingthelargestwithlessthan300000inhabitants.
Aslate as 1967, theWorld Bank(1967, p. 2) found that this`highdegree of
dispersionofthepopulation'contributedtotheìndustrialunderdevelopment
which these countries have demonstrated'. What the World Bank called `the
diffusionofthemarket'inCentralAmericawascharacterizedby`therelatively
small size of the major urban centres as well as the limited numbers of those
withpopulationsinexcessof100000'.
4
By1980,thepercentageofthepopulationlivinginthefivecapitalcitieshad
doubled to 20 per cent of the total population. Such unprecedented levels of
urbanization generated profound changes in the self-awareness of these soci-
eties. Thus, realities which had been hidden behind the lethargy of the rural
environmentnowbecameevidentandunavoidablewithinthemorecomplex,
urbanizedsettings.
Economic growth was another major change. Between 1950 and 1980, the
five Central American economies experienced intense growth in their gross
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
316 TheDevelopmentofCentralAmerica,195080
Table11.1 CentralAmerica:totalGDPandpercapitarealgrowthrates
GDPgrowthrate(%,basedon1970prices) GDPper
capitainUS$
195078 195055 195560 196065 196570 197078 1950 1978 Percentage
change
195078
Total 5.3 4.7 4.6 6.0 5.1 5.4 242 428 76.9
CostaRica 6.7 8.3 6.0 6.5 6.1 6.1 322 758 135.4
ElSalvador 5.1 4.6 4.7 6.8 5.2 5.2 203 347 70.9
Guatemala 5.1 2.2 5.3 5.2 6.1 6.1 255 451 76.9
Honduras 4.2 2.5 4.6 5.2 4.4 4.4 234 297 26.9
Nicaragua 5.6 8.3 2.3 10.2 4.0 4.0 223 409 83.4
Source: Author'scompilationbasedonECLACdata.
domestic product (GDP), at an average annual rate of more than 5 per
cent. Per capita income almost doubled, in 1970 prices, from US$242 in
1950 to US$428 in 1978 (Table 11.1). This is equivalent to one of the most
sustained and intense periods of economic growth ever experienced by
these economies. Furthermore, the composition of the region's product
underwent deep transformations. Agriculture, in 1950, was hiring about
two-thirds of the economically active population and contributing about
half of the region's product. In 1980, agriculture was still hiring about 50
per cent of the labour force, but contributing only one-fourth of the region's
GDP.
By contrast, industry emerged as a significant sector of these economies.
Fromcontributingabout10percentoftheregion'sdomesticproductin1950
andhiringaboutthesameproportionofthelabourforce,theindustrialsector
in1980contributedaboutone-fourthoftheregion'sproductandhiredabout
one-fifth of the workforce (Table 11.2). These rates of economic growth were
accompanied by the construction of physical infrastructure, finally over-
coming the lack of physical integration that existed within each country and
throughout the Isthmus. The region's mountainous topography, copious
rainy seasons and the concentration of the population along the Pacific
coastal strip had until then been formidable obstacles against the physical
integration of the individual economies and of the region as a whole. There-
fore, changes in physical infrastructure were spectacular, coming from a very
low and narrow base. The few means of transportation available primarily
served the export of basic commodities. By the end of 1951, there existed
5179kmofrailroads,ofwhich2900km,or56 percent,wereownedandused
exclusively by the bananacompanies that operatedmainly along the region's
Atlanticcoast.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 317
Table11.2 CentralAmerica:GDPcomposition(%,basedonconstant-1970prices)
Primary Secondary Tertiary
1950 1960 1978 1950 1960 1978 1950 1960 1978
Total 37.9 31.9 26.9 14.6 16.3 24.1 47.5 51.8 49.0
CostaRica 31.5 25.8 19.7 15.6 17.6 27.8 52.8 56.6 52.5
ElSalvador 41.9 35.9 26.2 15.8 18.1 25.9 42.3 46.0 47.9
Guatemala 35.5 32.8 28.9 15.0 14.8 20.6 49.5 52.4 50.5
Honduras 46.2 33.5 31.0 12.6 16.9 23.4 41.1 49.6 45.6
Nicaragua 37.4 29.6 28.0 12.1 15.8 27.1 50.5 54.6 44.9
Source: ECLACdata.
Bycontrast,roadtransportwithinandamongthecountrieswasalmostnon-
existent,withfewroadsbeingall-weather.Theamountofkilometresofpaved
roads tripled from less than 3000km in 1950 to more than 9000km in 1980.
Furthermore, whilst in 1950 no link existed between these countries by road,
because only one-third of the 2000km war-inspired Pan American Highway
had been completed, by 1960 the completion of this highway linked the five
CentralAmericancountriesbyroadforthefirsttime.Nevertheless,in1967the
WorldBankfoundthatinspiteofrecentdevelopmentssuchasthecompletion
of the Pan American Highway and of highways along the Pacific coast of
Guatemala and El Salvador, `the transport system for Central America con-
tinues to reflect a foreign trade orientation with most of the main arteries of
traffictunnellingtowardtheports(WorldBank1967,p.3).
Other components of the region's infrastructure experienced spectacular
increases. In 1980, the number of telephones per 1000 inhabitants doubled,
from4.6 in1950to11.6 in1980.Powergenerationincreasedfromalmost400
GWto5700GWandportcapacityincreasedfrom3754tonsin1950toalmost
11000tonsin1980.
Moreimportantwasthataccesstosocialservicesbythepopulationincreased
considerably,provingthatsomebenefitsofgrowthreachedthepoorestsectors
despite the repeatedly denounced deficiencies in the distribution of these
benefits. Therefore, life expectancy increased from 49 years in 1960 to 60
years in 1975, literacy increased from 38 per cent in 1950 to 57 per cent in
1975, and the percentage of the population with access to drinkable water
doubledfrom22percentin1960to46percentin1975.
However,therewereconsiderabledifferencesamongthecountrieswhichare
notreflectedintheseregionalaverages.Particularlyinsocialindicators,Costa
Ricashinesasanexceptioninallregionalaverages.Forinstance,by1975Costa
Rica's life expectancy was 70 years, literacy had reached 90 per cent and the
populationwithaccesstodrinkablewaterwasalmost80percent.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
TheCostaRicanexception
Costa Rica has always been an exception that contrasts with the evolution of
theotherCentralAmericaneconomies,particularlywiththoseofthenorthern
part of the Isthmus. However, this contrasting outcome remains one of the
leaststudiedenigmasofCentralAmericandevelopment.
5
Sincecolonialtimes,
due to the centralization of the Spanish administrative authorities in Guate-
mala, Costa Rica was the southernmost and the least integrated of the pro-
vinceswhichcomposedtheAudienciadeSantiagodelosCaballerosdeGuatemala.
Costa Rica was closer to Panama, from where originated some of the first
entradas carried out by the conquistadors into Central America. By contrast,
those whoconquered andsettled Guatemalawere drawnfrom theforcesthat
conqueredMexicounderHernanCortes.
Thislaxcommunicationwiththecolonial administrativecentrewasaccen-
tuatedbythelowlevelsofpopulationdensityandconsequentsparseeconomic
activity which characterized the province throughout the colonial period.
Costa Rica was a contrasting exception, at that time as well, to the relatively
high levels of population density and economic activity that prevailed in the
northernpartoftheIsthmus,particularlyinGuatemalaandElSalvador.
6
At the time of independence, in 1821, the total population of Costa Rica
amounted to 50 000 persons, concentrated in the central valley around the
provincial capital of Cartago. Thus in colonial times Costa Rica was the least
populated, the most isolated and the poorest of the provinces of Central
America. For these reasons, throughout the colonial period Costa Rica's eco-
nomic history is characterized by numerous attempts and failures to develop
productiveactivities.Forinstance,atthebeginningoftheeighteenthcentury
an unsuccessful attempt was made in the Matina region of Costa Rica to
increasetheproductionofcacao,usingslavelabour.
7
Thus,anothercharacter-
istic which distinguished Costa Rica from the rest of Central America was the
inability,throughoutthecolonialperiod,todevelopakeyexportproduct.
According to V ctor Cespedes and Claudio Gonzalez-Vega, `the scarcity of
labour, the abundance of land, and the limited opportunities for trade led to
theprevalenceoffamilyunits,toself-sufficiency,toindependence,andtothe
dispersion of economic decision making, despite mercantilistic regulations
(Cespedes and Gonzalez-Vega, 1993, p. 60). However, when a key export pro-
ductwasfinallydeveloped,inthefirsthalfofthenineteenthcentury,theseare
amongthefactorsthatmakeadecisivecontributiontothecontrastingoutcome
generated by the emergence of coffee as a key export product. In Guatemala
and El Salvador, with relatively abundant supplies of land and labour, the
exploitation of coffee led to the domination of latifundia and authoritarian
meanstosecuretheavailabilityoflabour.Bycontrast,inCostaRicatheintro-
ductionofcoffee,withabundantlandandscarcelabour,ledtoitsexploitation
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 319
insmallplotsoflandandtotheemergenceofsmallfarmers,whichcoexisted
with some large exploitations where processing activities were carried out to
preparetheproductforexport.
However,whatiscontrastinginCostaRica,again,isthatthecoexistenceof
both forms of production led to the emergence of a negotiating culture
between the large enterprises and the small farmers. On this culture rests a
widerdistributionofthebenefits,amoreegalitariansocialstructureandamore
democratic political system which resulted from the exploitation of coffee.
Later,afteracenturyofcoffeeexports,acoalitionledbytheurbanizedmiddle
class, allied to small farmers and some small entrepreneurs and industrialists,
inaugurated the modern democracy and the welfare state that characterizes
contemporary Costa Rica. Such an outcome makes Costa Rica, once again,
standapartfromtherestofCentralAmerica.
Theexternalsector'sdecisiverole
As has been the case throughout the economic history of these small, open
economies, the main source of dynamism came from the world demand for
oneortwocommoditiesproducedforexport.Thiswasnodifferentfromother
periods of the region's history, with one or, at the most, two key export
products such as coffee and bananas dominating and decisively influencing
thelevelsofeconomicactivity.Whatwasnew,however,inthe1950swasthat
allthetraditionalexportswereexperiencingasimultaneoussustainedboomin
world demand. To be sure, the cyclical upswings and downturns which char-
acterizethedemandforthesetraditionalexportswerepresentthroughoutthe
threedecades,butthesewerebriefandnotprofound.
Inaddition,theCentralAmericancountriesbenefitedgreatlyfromopportun-
ities available to export new commodities, such as sugar, cotton and meat.
The new products joined the more traditional exports in an unprecedented
diversificationwhichledtoaspectacularincreaseinexportstotherestofthe
world,fromUS$250millionin1950toUS$3.3billionin1977.Thisamountsto
anaverageannualrateofgrowthof10percent,ortoamultiplicationoftheir
valueby13.
Bothincreasesinvolumeandinvalue,atdifferenttimes,wereresponsiblefor
thesustainedexpansionofexports.Forinstance,thefirsttwodecades,between
1950and1970,wereofsustainedexpansioninthevolumeofexports.During
the1970s,onaccountofasteepincreaseinthepriceofcoffee,theexportboom
was mainly fuelled by price increases. However, the increase in imports
exceededbyfartheexportboom,thuscontributingtothesustainedexpansion
of these economies and generating a deficit in the trade account. Imports
increased from US$300 million in 1950 to US$5.5 billion in 1977 and experi-
enced a significant transformation in their structure. There was a sustained
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
increase in imports of consumer goods, but also a marked trend towards
imports of raw materials, intermediate products and capital goods. In 1950,
consumer goods represented 45 per cent of total imports, with intermediate
goodsamountingto24percentandcapitalgoods17percent.In1977,imports
of consumer goods had fallen to 23 per cent while intermediate products
increasedto34percentandcapitalgoodsto23percent.
This diversification of the external sector was strengthened by an incipient
industrialization supported by the process of economic integration which
began in the 1950s. Starting from a very narrow base, the manufacturing
structure was dominated by the production of traditional consumer goods
such as food processing, beverages and tobacco, textiles, shoes and clothing,
which according to the World Bank accounted for 75 per cent of output. Of
decisive importance in the expansion of manufacturing activities was the
increaseindomesticdemandcausedbyboththeexportboomandbyregional
tradeliberalization.In1967,theWorldBankestimatedthatbetween1962and
1965 almost two-thirds of the increase in manufacturing activity was due to
increasesindomesticdemand.However,theWorldBankalsoestimatedthatà
littleover20percentoftheincrement'intotalmanufacturingoutput,inthe
sameperiod,wascausedbytheexpansionofintraregionaltrade(Cespedesand
Gonzalez-Vega1993,p.8).
Regionalmarketexpansion ledto anincrease inintraregional trade, mostly
inmanufactures,fromlessthan3percentoftotaltradein1950tomorethan
one-fourth by 1970. The perception was that à new autonomous source of
dynamism'wasemerging(ECLAC1992,p.16).Thegeographicdistributionof
Central America's total trade was also more balanced. In 1950, the USA sup-
plied more than two-thirds of imports and purchased the same amount of
exports. By 1980, total trade with the USA hadgone down to one-third while
intraregional trade had increased from 3 per cent to one-fifth. Also, in 1980,
JapanandtheEuropeanUnioneachaccountedfor15percentoftotalCentral
American trade, with the rest going mainly to Latin America and the Carib-
bean.
Contrary to some predictions, despite the increase in industrial protection-
ismwhichaccompaniedtheexpansionoftheregionalmarket,theseeconom-
ies remained more open than ever to international trade. Thus, the average
import coefficient went from 16 percent of GDP to more than 33 percent in
1977, which placed the Central American economies among the ranks of the
mostopenofLatinAmerica.
Finally, the persistent deficit in the current account was financed with
increasedamountsofexternalresources,towhichthesecountriesgainedaccess
at unprecedented levels. The ratio of net external finance to exports of goods
andservicesdoubled,from6.3percentin1955to12.8percentin1975.Public
externaldebtincreasedfromUS$93millionin1960toUS$3.2billionin1978,
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 321
while direct foreign investment flows went from US$400 million in 1960 to
US$1.1billionin1977.
Economicintegration:ECLAC'sroleandUSparticipation
Theprocessofregionalintegrationwaslaunchedintheearly1950swithstrong
encouragementandsupportfromECLAC.Activeinvolvementintheprocessof
CentralAmericaneconomicintegrationservedasatestinggroundforsomeof
the proposals made afterwards by ECLAC, in the early 1960s, on Latin Amer-
icanintegration.Therefore,itisofhistoricalinteresttorecallherebrieflywhat
werethemainelementsofECLAC'soriginalproposaloneconomicintegration
forCentralAmerica.
8
ThemainjustificationfortheeconomicintegrationofCentralAmericawas
the smallness of the individual, national markets. According to ECLAC, `the
smallness of the market is an obstacle to economic development, and in the
Central American republics it constitutes perhaps the major impediment to
their progress'. Instead of the immediate creation of an economic union,
ECLAC proposed to the Central American governments the gradual adoption
of `limited integration accompanied by a policy of commercial and industrial
reciprocity'(ECLAC,1952).
ECLAC stated that `within the factual domain, and responding to practical
reasons, it is not possible to proceed immediately to the complete economic
unionoftheCentralAmericanrepublics,evenwhenthisconstitutesahistoric
aspirationofallofthem'.Theìmmediategoal',ECLACstated,`shouldnotbe
too ambitious, butit shouldbe audacious' (ECLAC, 1952). Therefore, `limited
integration'meantapolicythat`leadstotheoptimallocationofsomeimport-
anteconomicactivities...withthepurposeofestablishingproductiveunitsof
adequatesize,thatutilisetheregion'srawmaterialsandareabletosupplythe
CentralAmericanmarketsasawhole,atthelowestcostsofproduction'(ibid.).
Economic integration was notan end in itself, it was an instrument for the
industrialization of these small economies. Because `the problem of inward
looking growth in Central America is fundamentally reduced to the scarce
magnitude of the markets, in which large capital investments are not always
justified' (ibid.). The conclusion was that `the industrial effort of each coun-
try...cannot be as productive as it would be if there was a wider division of
labour among the five republics, that avoids investment duplication and that
provides,throughwidermarkets,alargerbaseofsupporttotheindustriesand
activitiesofeachcountry'(ibid.).
Finally,integrationhadtobebasedonreciprocity,whichhadtwocompon-
ents. First, èach country agrees that in the others will be located certain
industries of adequate magnitude, because in exchange the other countries
accept that in the first country will be located other identified industries
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
basedonthesameprinciples'(ibid.).Theotheraspectofreciprocityhadtodo
withcommercialandtariffpolicyanddependedontheguarantee,givenbyall
the countries, that only the products of the industries agreed on in a general
plan would enjoy the benefits of free trade within the region. Òtherwise',
ECLAC concluded, `the maximum advantages from optimal location of the
activitieswouldnotbeobtained'(ibid.).
With these modest objectives, by 1958 after almost eight years of intense
activities, some results were accomplished in the fields of road transport and
intraregional trade.
9
These accomplishments took the form of several treaties
signed by the five Central American governments regulating road transport
andconstruction;liberalizingintraregionaltradeforalimitedlistofproducts;
and creating a regional procedure for the allocation of industrial investment
knownastheCentralAmericanRegimeofIntegrationIndustries.Underthese
instruments, themajor incentivefor attractinginvestmentinto theindustrial
sector was the exclusive ten-year privilege of supplying the expanded market
grantedtotheproductsoftheindustriesselectedtobenefitfromtheRegime.It
was also agreed that those industries authorized under the Regime would be
locatedequitablyamongalltheparticipatingcountries.
Generalizedfreetradewasnotthepurposeofthesetreaties.Thegoalofsuch
`limitedandgradual'integrationconsistedoftheestablishmentofpreferences
forsomemanufacturesasanincentiveforindustrialization.Therefore,in1959,
to complement free trade and integration industries, the Central American
governments signed another treaty setting up a common external tariff for a
limitedlistofproducts,particularlythosethatwereexpectedtoenjoyfreetrade
because they would be produced by the first regional industries. But by 1959
even these limited objectives were facing difficulties in getting approval,
because the governments of Costa Rica and Honduras were delaying the rati-
ficationofthecorrespondinginstruments.Theintegrationprocess,aftereight
years,wascharacterizedbytheresistanceofthemostscepticalparticipantsand
bytheimpatienceofthemosteagersupporters.Itwaspreciselyatthisjuncture
that the Central American process experienced an audacious transformation,
movingintoahigherlevelofintegration.
The government of El Salvador was the most active supporter of regional
integration, as a means to alleviate some of the consequences of what was
perceived as the overpopulation of its relatively small territory. In March
1959,PresidentLemusofElSalvadorvisitedPresidentEisenhowerinWashing-
ton. The communique issued at the end of the visit expressed the determina-
tion of El Salvador and the USA to support `the establishment of an
economicallysoundsystemfortheintegrationoftheeconomiesoftheCentral
American Republics'. Just a month before, the meaning of a `sound system of
integration' had beenclarifiedby Under-Secretaryof theUS TreasuryDouglas
Dillon at a meeting of the Committee of Twenty-One of the Organization of
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 323
American States (OAS). To obtain US support, Under-Secretary Dillon said
regionaltradeagreementsshouldaimatliberalizingalltradeamongtheparti-
cipants,thereforelimitedpreferentialagreementswouldnotbesupported.
The adaptation of the Central American integration process to these condi-
tionstooktheformofasignaturebyElSalvador,GuatemalaandHondurasof
whatwasknownastheTripartiteTreaty,acceleratingtheformationofacom-
pletefreetradeareaandacommontariffamongthesubscribers.Later,because
of the insistence of the Nicaraguan government, ECLAC was asked to draft a
projectknownastheGeneralTreatyofEconomicIntegration,signedinMan-
aguainDecember1960exceptbyCostaRica,creatingwhatbecametheCentral
AmericanCommonMarket(CACM).TheGeneralTreatyliberalizedacrossthe
board almost all intraregional trade and gradually set up a common external
tariff over a period of five years. Costa Rica adhered to the General Treaty in
1963 after an intense debate and after the government of President Mario
Echandi was succeeded by the opposition, led by President Francisco Orlich
fromthePartidoLiberacio nNacional.
The results of this audacious decision were seen in the spectacular increase
experiencedbyintraregionaltrade,whichwentfromrepresenting6 percentof
thefivemembers'totalimportsin1960,tomorethan20percentin1969.By
contrast, only two plants were established under the Regime of Integration
Industries,atyreplantinGuatemalaandacausticsodaplantinNicaragua.
In1967,therecentlycreatedUnitedNationsConferenceonTradeandDevel-
opment(UNCTAD),underthedirectionofRaulPrebisch,saidthattheCentral
American Common Market (CACM) was `the most clearly successful integra-
tion effort that has been undertaken by independent developing countries
whichpreviouslyconstitutedseparatenationalmarkets'(UNCTAD1967,p.15).
Some of the indicators of success cited by UNCTAD were the abolition of
tradebarriersforallbutafewproductsoriginatingintheregion;theagreement
on uniform duties towards third countries on 98 per cent of tariff items; a
framework for recognizing ìntegration industries' and an agreement for the
harmonizationoffiscalincentives;thecreationofcommoninstitutions,such
as the Central American Bank, decisively supported by the USA, the Central
American Clearing House and an industrial research institute. These accom-
plishments, UNCTAD concluded, had pushed intraregional trade from 3 per
centoftotaltradein1950to15percentin1965,withmorethantwo-thirdsof
suchtradebeinginmanufacturedproducts.
Theshortcomings
Despite the profound changes described, many shortcomings remained.
ECLAC'sMexico Office,in1980,attemptedtodraw abalanceoftheperform-
ance of these economies during the three previous decades. Their conclusion
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
was expressed in terms of six `contradictions or dilemmas', as follows: (a)
dynamicgrowthwithunequaldistributionofbenefits;(b)wideninganddivers-
ificationoftheexternalsectorwithincreasedexternalvulnerability;(c)initial
expansion of the process of economic integration and signals of exhaustion;
(d) agriculture as the image of biased development; (e) dynamic but still in-
sufficient accumulation and investment; and (f) increased social complexity
withinefficientparticipationmechanisms(ECLAC1992,pp.23).
ECLACrecognizedthattherehadbeenintensegrowth,butthebenefitshad
been distributed unevenly, even inequitably. The numbers of those living in
poverty demonstrated that the benefits of growth had not reached large seg-
ments of the population. The poorest 20 per cent of the population only
received between 3 and 4 per cent of national income, while the wealthiest
20 per cent received almost 60 per cent of national income (Table 11.3).
Furthermore, while the middle income groups experienced the fastest rate of
growth in their income, leading to the emergence of middle classes, this also
widenedthedistancebetweentheextremes.
Traditional exports sustained the impressive rates of growth of imports,
contributingtoagreaterdegreeof opennessoftheseeconomies. Butthisalso
meant that they had become more vulnerable to external shocks for several
reasons.First,exportdiversificationwasstillincipientandtraditionalexports
coffee and bananas remained dominant. Second, domestic agriculture was
unabletosupplytheexpandingdomesticdemandforfoodleadingtoincreased
imports. Finally, some countries gained access to international lending,
enabling them to finance for a short while theircurrent account deficits thus
makingthemevenmorevulnerabletoexternalshocks.Bytheendofthe1970s,
noneofthesefactorshadmademuchdifference,sinceuptothatpointexport
shortfalls were brief and the impact of external shocks was shallow. In these
conditions, the Central American economies weathered the first oil shock of
1973 through price increases in their traditional exports such as coffee, com-
plementedwithaccesstoexternalfinancing.
Table11.3 CentralAmerica:incomedistributionbyhouseholds
Percentagesofincome
CostaRica ElSalvador Guatemala Honduras
1961 1971 1961 196567 1969 1970 196878
20%poorest
30%underthemedian
30%abovethemedian
20%richest
5%highestincome
Source: ECLACdata.
6.0 5.4 5.5 3.2 3.7 4.9 2.3
10.4 15.4 10.5 12.0 14.9 12.5 8.5
23.6 28.5 22.6 26.7 30.6 23.8 21.4
60.0 50.6 63.3 58.1 50.8 58.8 67.8
35.0 22.8 32.9 28.5 20.6 35.0 38.4
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 325
The other source of dynamism of these economies, besides the traditional
export sector, was the process of import substitution sustained through the
expansion of the regional market. Most of the evaluations carried out at the
timecametotheconclusionthattheratesofgrowthoftheseeconomieswould
have been lower without the process of economic integration (World Bank,
1971; SIECA, 1972; and ECLAC, 1974). However, around 1980, ECLAC found
evident `signals of exhaustion' in the Central American integration process,
particularly the instability of the free trade regime which was discouraging
regional investments. This was evident in a slowdown in the rates of growth
ofintraregionaltrade,fromalmost30percentayearthroughoutthe1960sto
15percentinthe1970s.
The relative participation of intraregional trade in total exports was falling,
fromalmost25percentinthe1960stolessthan20percentinthe1970s(see
Table 11.4). ECLAC even found that a process of import substitution of intra-
regional imports was underway, with the relative participation of industrial
production in intraregional trade falling from 25 percent in 1968 to 11.5 per
cent in 1977. Furthermore, a common complaint was that the integration
process had not gone beyond its original objectives, leaving aside key areas
suchasagriculture,energyandseveralsocialsectorssuchaseducationorhealth.
In 1969, the integration process survived an armed confrontation between
twoparticipatinggovernments,knownasthe`soccerwar'betweenElSalvador
andHonduras.However,thesequelalsoshowedthattheseparationofeconom-
icsandpolitics,whichcontributeddecisivelytotheearlysuccessoftheintegra-
tion process, was breaking down. Political differences also began to interfere
with the functioning of the free trade regime. As verified by a World Bank
missionin1971,`thevolumeofintra-regionaltrade,whichhadincreasedata
spectacularpaceinthe1960s,sufferedanabsolutedeclineofabout4percentin
1969asaresultoftheHonduras/Salvadorconflict'.Inaddition,`themomentum
ofindustrialproductionwasalsointerrupted'(WorldBank1971,p.45).
Briefly,bytheendofthe1970sthebeliefthattheprocessofimportsubstitu-
tionandeconomicintegrationhadbecomeanàutonomous'sourceofdynam-
ism was constantly being questioned, as much on grounds of difficulties
inherent to the process as because of external factors. `Viewing the industrial
landscapein1970',thesameWorldBankmission`wasimpressedbythetrans-
formationthathadtakenplaceinthecompositionofindustrialproductionin
CentralAmerica'.Thistransformationwascharacterizedasfollows:`thetradi-
tional lines of manufacture food, beverages, tobacco, clothing, footwear
retained their importance in the overall structure but superimposed on the
scenewereawidearrayofnewactivitiesandnewproducts'.
However, the World Bank noticed that it was èasy to be misled by all the
novelty and the statistics. A balanced view has to recognize that Central
Americanindustrialisationtodateisskin-deep'.Thereasonwasthattherewas
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Table11.4 CentralAmerica:theshareofintraregionaltradeintotaltrade(%)
1968 1975 1980 1981 1982 1983 1984 1985 1986 1987
ExportstoCACM/totalexports
CostaRica 21.3 21.7 27.0 23.6 19.2 22.7 19.2 15.5 9.3 7.7
ElSalvador 40.3 27.6 41.1 422.1 42.7 35.2 25.4 15.7 12.1 19.4
Guatemala 31.2 27.0 27.4 32.0 29.5 27.6 26.1 20.7 17.7 21.8
Honduras 17.3 9.1 10.3 9.2 7.9 9.3 6.8 3.0 2.6 2.9
Nicaragua 15.2 24.7 17.3 14.2 12.8 7.8 9.6 8.8 6.6 7.5
TotalCACM 23.5 23.3 25.4 24.5 22.4 21.6 18.9 13.9 10.4 11.9
ImportsfromCACM/totalexports
CostaRica *** 16.5 14.4 12.6 12.6 12.2 10.5 8.4 *** ***
ElSalvador *** 22.9 32.8 29.2 27.6 24.2 19.1 19.6 17.2 18.3
Guatemala *** 14.1 10.0 11.5 15.5 19.9 12.9 7.7 *** ***
Honduras *** 12.8 10.3 12.5 12.5 12.7 12.2 8.6 *** ***
Nicaragua *** 21.8 33.9 21.1 15.1 15.3 9.0 5.3 5.0 4.6
TotalCACM 21.3 17.6 18.5 16.7 16.9 17.1 13.2 9.9 *** ***
Sources: IBRD,reportno.2325b-CAfor1968figures;SIECA,SeriesEstadisticasSeleccionadasde
Centroamerica,June1987forotherfigures.
`scarcely any production of capital goods or the heavy category of intermedi-
ates'. In effect, the World Bank concluded, `there have come into existence a
largenumberofplants,manyoflessthanoptimumsize,whichhavebittenoff
a tiny chunk of value added at the finishing end of the processing operation'
(WorldBank1971,p.9).
After three decades of import substitution, agriculture remained the key
sector, the main employer, source of dynamism and generator of foreign
exchange. The World Bank estimated that in 1969, total employment in the
industrialsectorwasaround535000,oralmost12percentofthelabourforce
(WorldBank1971,p.8).Also,by1985,theratioofmanufacturingtoGDPhad
increasedonlyto17.5percent.Themainreason,identifiedbytheWorldBank
in1990,wasthedeclineinintra-CentralAmericantrade,whichhadstimulated
industrial growth during the previous two decades. The manufacturing sector
was`highlydependentontheregionalmarket'.Forinstance,between1970and
1980,`manufacturingexportsincreasedfromUS$234milliontoUS$1.2billion.
Morethanhalfofthisincreaseisaccountedforbyintra-regionaltrade'.Exclud-
ingprocessedfoods, in 1980almost 19percentof manufacturing production
was exported to the Central American market, while less than 4 per cent was
exportedelsewhere(WorldBank1990,pp.910).
Itwasnotablethatindustrializationnevercollidedfrontallywithagriculture,
never even challenged its supremacy. Industry had been privileged and pro-
tected although not necessarily at the expense of export agriculture.
10
Both
import substitution and export agriculture coexisted without contradicting
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 327
each other, in a style of development characterized as àdditive'. This kind of
development was described as the `progressive juxtaposition, since the end of
the Second World War, of new layers of economic activity upon the basic,
traditional socioeconomic structure'. This addition of new layers contributed
todiversifyingthese economies,buthadnot`decisivelytransformed preexist-
ingconditions'(CohenandRosenthal1983,p.18).
Ifindustrializationwasnotpromotedattheexpenseofexportagriculture,it
was,however,attheexpenseoftheproducersoffoodforthedomesticmarket
andthesmallcottageindustriesthatexistedinthecitiesandthecountryside.
Themaininstrumentwasabiasinthedomestictermsoftradeagainstfoodand
staples,keepinglowthepricesoftheseproductseventhroughimportsoffood,
if necessary. A policy of fixed exchange rates, pegged to the US dollar, con-
tributed to price stability and low inflation. In addition, exchange rate over-
valuation strengthened the anti-export bias which characterized the policy of
import substitution and kept low prices for imported inputs, needed both by
traditionalexportsanddomesticindustries.Thepriorityformonetarystability
contributed, with the openness of these economies, to decades of remarkable
price stability, reflected in single-digit rates of inflation. Domestic prices were
thus determined by international price movements and monetary policy was
responsivetothebalanceofpayments.
Thiskindofdevelopmentwasalsoàdditive'becausenewactivitiesemerged
attheexpenseofthosewhohadlesscapacitytoasserttheirdemands,particu-
larlythosesectorsthatwerelessorganized.Bycontrast,themainbeneficiaries
werethosethatwerebetterorganised,suchastheurbanizedmiddlesectors,the
newindustrialistsandthetraditionalexporters.
Another indicator of àdditive development' was that from the ranks of the
traditional exporters and those involved in importexport trading activities
had emerged new industrial entrepreneurs. Looking at the role of financial
intermediaries in the industrial development of Central America in 1972, the
World Bank found that with the exception of Costa Rica where commercial
bankswerenationalized,intheothercountries`bankswereownedbyarather
narrowgroupoffamilies'.Inmanycases,theWorldBankalsofoundthatthese
were the `same interests which dominated industry, large agricultural estates
and commerce'. In these conditions, `barriers to entry were high. The small
new investor lacking connections and/or collateral found it difficult to raise
fundsfromtheseinstitutions'(WorldBank,1971,p.27).
Also prominent in the ranks of the new industrial entrepreneurs were the
transnational corporations that, before the erection of protectionist barriers,
were the main exporters of consumer goods to the region. Among the main
sourcesofindustrialfinanceweretheseforeigninvestors,followedbybilateral
and multilateral official lending agencies including the Central American
Bank for Economic Integration (CABEI) and the US Agency for International
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Development (AID). Estimates by the World Bank revealed that 27 per cent
of industrial finance between 196269 came from these external sources.
`Heavy reliance on foreign funds' was seen as a `distinguishing feature of
industrial finance in Central America', although this was also seen as an
overall characteristic of these economies, since `total capital formation in the
region was financed from abroad to the extent of 31 per cent' (World Bank,
1971,p.26).
Most of these foreign resources were direct investments, mainly from the
USA.TheWorldBankfoundthat,until1960,US-basedmultinationalcorpora-
tions èstablished 47 subsidiaries for manufacturing operations' in Central
America. `However, in the subsequent seven years 79 more subsidiaries were
born'.Bycontrastwithitspastorientationtowardstheprimaryexportsector,
the World Bank estimated that àbout 64 per cent of the total increase in US
foreigndirectinvestmentinCentralAmericaduring196369wasintheindus-
trial sector' (World Bank, 1971, p. 32). Some of these circumstances were
leading to an industrial structure dominated by monopolies. The World Bank
expressedconcernthatànexaminationof57productlines,whichaccountfor
the bulk of regional trade in manufactured goods', showed that `trading con-
ditions in the Common Market did not approach perfect competition by any
means'(WorldBank,1971,p.13).
Finally,theseintensivechangesdidnotleadtotheemergenceofnewactors
capableofcounterbalancingtheinfluenceoftraditionallydominantgroups.In
thoseinstanceswherethesegroupswerechallenged,itwasthroughopenand
even armed rebellions which originated among dissatisfied segments of the
middleclass.Reformismornon-violentattemptsatchallengingthedominant
groups were confined to very narrow limits, mostly within the constrained
spacelefttopublicpolicy-making.
Still, within these very narrow limits the public sector emerged as the
promoter of development. Relative participation of public expenditures in
GDP stagnated at around 11 per cent between 1955 and 1970, increasing
slightly to around 16 per cent in 1978. Once again the exception was Costa
Rica, where social indicators were pushed far above the regional average
throughpublicexpendituregoingfrom8.4percentofGDPin1950to23per
centin1978.
Because of these constraints in the availability of public resources, most
reform programmes initiated by the governments were either very limited or
decisively supported with external funding. These external resources were
grantedeither throughofficial bilateralandmultilateralchannels, orthrough
commerciallendingwhichemergedasanalternativeafterthefirstoilshockof
1973. These and other shortcomings became more salient because of the pro-
found depression unleashed by the second oil shock of 1979, which ravaged
theregionwithunprecedentedintensity.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 329
Thebust
The first oil shock of 1973 was weathered with minor disruptions by the
Central American economies, through a combination of price increases in
traditional exports and access to commercial lending. However, this was a
mixed blessing. When the second oil shock came in 1979, the three decades
ofsustainedexpansionandtheresponse of1973exercisedapowerfulgripon
decision-makers.Nobodyanticipatedinallitsmagnitudetheprofoundcorrec-
tion that was about to set in. Throughout Latin America and the Caribbean,
decision-makers became easy victims of the past, or of `path dependence',
11
attemptingdesperatelytoresortinvaintotraditionalremedies.
To complicate matters even more, it was Central America's infinite misfor-
tune that the depression of the 1980s, better known in Latin America as the
`lost decade', coincided with an intensification of Cold War confrontations.
Ignited by the overthrow of the dynastic dictatorship of Anastasio Somoza in
Nicaragua, these confrontations between the superpowers exacerbated other
domesticconflicts,inElSalvadorandGuatemala.
Itwasthiscombination,oftheeconomicdownturnofthe1980swithoneof
the last confrontations of the Cold War, which made the Central American
depression different from and nastier than that of the rest of Latin America
andtheCaribbean.Also,asifclaimingitsdue,externalvulnerabilityreturned
forcefully, this time in the guise of dependence on imported energy. Since
all the Central American countries were net importers of crude oil, more
than 70 per cent of power generation was from thermal plants. The value of
oil imports rose from US$60 million in 1970 to US$600 million in 1979
(ECLAC,1981).Inaddition,after1973,theavailabilityofcommerciallending
hadgivenaccessmainlyinCostaRicaandNicaraguatounprecedentedlevelsof
indebtedness. With the steep increase in interest rates that followed, debt-
servicing soared, demanding larger proportions of scarce foreign exchange
and burdening the external accounts, particularly those of the most indebted
countries.
Clarence Zuvekas has pointed out that Costa Rica can claim the dubious
honour of inaugurating the debt crisis in Latin America. The government of
Costa Rica was the first that unilaterally suspended payments to commercial
banks,inJuly1981,exactly13monthsbeforeMexicomadethesamedecision.
However,Zuvekasconcludes,`theabsolutesizeofCostaRica'sexternaldebtat
the time US$2.7 billion was too small to have a major impact on interna-
tionalfinancialmarkets'(Zuvekas,1996,p.4).
Consequently,inflationalsobecameunleashedinaregionthathadenjoyed
price stability and fixed exchange rates for more than half a century. For
instance, the average increase in the consumer price index amounted to 13
per cent between 1950 and 1970. Therefore, when in some countries the
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
consumerpriceindexreachedmorethan30percentyearly,perplexitywasthe
initialreactionamongpolicy-makers.
Even worse would follow, because another Central American country
achieved another dubious honour by registering some of the highest rates of
hyperinflation in the hemisphere. In 1988, the consumer price index of
Nicaragua soared to more than 14 000 per cent a year, went down to 4700
per cent in 1989 and soared again to more than 12000 per cent in 1990.
In addition, as if nature's ire was set against the Isthmus, throughout the
1970s an unprecedented succession of natural disasters devastated the
region. Four years of drought between 1970 and 1977 coincided with earth-
quakes in Nicaragua in 1972 and in Guatemala in 1976, while Hurricane Fif
razedtheregionin1974,hittinghardestinHonduras(ECLAC,1973,1974,and
1976).
Central America has always been a disaster-prone region, both natural and
manmade.However,describingtheeventsofthe1970s,ECLACconcludedthat
since records of disasters were kept, `not so many had been observed and of
such gravity in such a briefperiod of time'. On top of the loss of lives, recon-
struction required large amounts of resources and exacerbated existing social
tensions(ECLAC,1992,p.81).
Yet, if the economic depression had been the only negative circumstance,
to challenge these fragile economies during the 1980s, Central America
would not have been much different from the rest of Latin America and the
Caribbean. What made the difference was the simultaneous outbreak of a
major confrontation between the Soviet Union and the USA. As with the
misfortune of natural disasters, one of the last confrontations of the Cold
WarwasplayedoutintheIsthmus.Thelossoflivesandtheeconomicdamage
caused by the internal conflicts, which intensified after the overthrow of
Somoza in Nicaragua, were even worse than those caused by nature in the
1970s(ECLAC,1979).Forinstance,massivemigrationstookplaceinunpreced-
ented numbers. It is estimated that almost 10 per cent or two million
Central Americans migrated outside the region, mostly to the USA and
Mexico. In El Salvador alone, it is estimated that during the 1980s one-fifth
of the population, or one million persons, migrated outside the country,
about 600000 of them to the USA. In addition, political insecurity and
inflation spurred capital flight, deepening the scarcity of foreign
exchange, which became the dominant challenge to short-term stabilization
policies.
Summing up, Central America went from three decades of intense and
profound economic growth and social change to a decade of depression and
politicalconfrontationwhichthreatenedtounravelseveral,ifnotmost,ofthe
positiveaccomplishmentsofthepreviousdecades.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 331
Somelessons
Today,withthebenefitofhindsight,thereiswideconsensusabouttheoriginsof
whatinthe1980sbecametheCentralAmericancrisis.Mostobserversagreethat
afterthreedecadesofsustainedexport-ledgrowthandregionalintegration,the
events of the 1980s showed that the Central American economies were still
vulnerable to external circumstances. Not much could be done by Central
Americanpolicy-makersaboutfallingcommodityprices,steepincreasesinoil
prices,mountinginterestratesandonerousdebtburdens.
However, more than the increased vulnerability endemic to the region, the
reactions and responses to the turnaround in external circumstances, with
somedegreeofexceptionforCostaRica,alsorevealpronouncedinflexibilities
inthedomesticpolicy-makingprocess.Tobesure,someofthesereactionsand
responsescanbeexplainedasmanifestationsof`pathdependence'andofthe
normally lengthy and complex process of adaptation to different circum-
stances. This perhaps explains the degree of perplexity and the delayed
responsesfrompolicy-makersinthefaceofchangedcircumstances.
The question remains why three decades of sustained growth and change
did not lead to the emergence of more diversified and resilient economies,
more capable of responding and adapting to the frequent external shocks
which afflict them. After all, this was the declared purpose of some of the
policiesadoptedduringthethreedecadesofintensegrowth.However,import
substitution and the process of economic integration never became an alter-
native to the traditional exports that remained the main sourceof dynamism
for these economies. Bulmer-Thomas (1987, p. 279) argues correctly that
import substitution never really challenged the existing traditional export
activities.
12
One main lesson may be drawn from the three decades of change and the
subsequent period of instability and depression. Even such intense and pro-
found changes were not enough to move these economies to more sustained
developmentortolessvulnerabilitythroughmoreadaptability.Oneexplana-
tion for this resistance to change may be found in what has been called the
`limiteddimensionsofeconomicpolicyspace'(CohenandRosenthal,1983,p.
18).Throughoutthedecadesexamined,themarginforpolicychangewasquite
narrow. Those interested in promoting non-violent reforms never had the
capacity to fully carry through their projects. With the exception of Costa
Rica,the policyspaceforreform wasnarrowlysqueezedbetween the capacity
ofthetraditionalsectorstopreservetheirsupremacyandthoseattemptingto
overthrowthembyforce.
To complicate matters even more, the formidable presence of the USA did
notalwayscontributetoreform,particularlywhenthesecurityoftheIsthmus
wasatstake.Forcedtochoosebycircumstancesbetweensecurityandeconomic
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
andsocialreforms,theUSAoftensidedwiththetraditionalsectors,narrowing
even more the limited space for reforms. Consequently, there was a constant
deflectingofpolicyinitiativesforreformtowardsless-demandingorless-pain-
ful alternatives. To cite but a few examples, external finance was preferred to
taxreforms,whilethecolonizationofpubliclandswasfavouredoveragrarian
reform.Eventhewideningofthemarketthrougheconomicintegrationcould
beseenasaneasieralternativetoincreasingthelevelsofdomesticdemand.
Someoftheseconclusionsshouldbetakenintoaccountwhenlookingatthe
future of the Isthmus. With the end of the Cold War and the beginning of
economic reactivation in the 1990s, again, these economies are engaged in
anothercycle ofreforms. Thistime round, thereformsaretakingplaceunder
civilian regimes in the five countries, which contributes to an unprecedented
degreeofhomogeneity.Thistime,theobjectiveistoopentheseeconomiesto
internationalcompetition,tomakethemmoreresilientanddiversified.Unfor-
tunately, present rates of growth have not yet reached the levels of the three
decades examined. At around 3 per cent, with inflation remaining moderate,
presentaverageratesofgrowtharestillinsufficienttocontributedecisivelyto
overcomingpovertylevels.Inaddition,manifestationsofformidableresistance
tosomeofthesereformshavealreadyappeared,particularlyagainsttaxreform
andthereformofthepublicsector.EvenCostaRica,whichconsistentlyshowed
remarkable flexibility and adaptation during the last four decades, is showing
unprecedentedsignsofresistancetochange(MongeandLizano,1997).
In a context where there are no security threats such as those that existed
duringtheColdWar,thehopeisthatthistimechangeinCentralAmericawill
notbe resistedaseffectivelyandthatthespaceforreform willbewider.If so,
perhaps this time change will not end in another decade of depression and
violentconfrontation.
Notes
1 Central America is used here as it has been since colonial times; it includes Costa
Rica,ElSalvador,Guatemala,HondurasandNicaragua.Atthattime,Belizewasnot
yetindependentandthetermhasneverincludedPanama.
2 Hirschman translates fracasomania as `failure complex' and defines it as `the insist-
enceonhavingexperiencedyetanotherfailure'(Hirschman,1968,pp.889).
3 A note on the sources used in this study. Official documents from the World
BankandECLACareusedhereasprimarysources.Bothinstitutionshadadifferent
positionabouttheprocessofindustrializationandeconomicintegrationofCentral
America. While the World Bank never supported the Central American regional
process, or regional projects, it evaluated its performance repeatedly through-
out the period, at least three times. ECLAC, by contrast, has been consistently
supportiveofregionalismintheIsthmus.Unlessstatedotherwise,thedatapresented
in this paper was mainly drawn from these sources, above all from an evaluation
of the three decades at ECLAC's Mexico office (1992). The last version is quoted
here.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
IsaacCohen 333
4 ToillustratethispointtheBanksaid,in1967,`GuatemalaCitywithapopulationof
approximately570000[was]thelargesturbancentrefollowedbythecapitalcitiesof
theothercountriesSanJose (CostaRica)324000;Tegucigalpa(Honduras)170000.
Otherthanthesecapitalsthenextlargestcitiesareunder100000withSanPedroSula
(Honduras)approximately85000andSantaAna(ElSalvador)75000'(WorldBank,
1967,p.2).
5 ThemostcompletecomparativeattempttoexplainthisenigmaisbyYashar(1992).
6 Murdo MacLeod illustrates the magnitude of the demographic depression experi-
enced by the Isthmus as a result of the conquest with the following figures: in
Nicaragua, out of 600000 tributary Indians counted in 1525, only 4716 remained
in 1686;in Honduras, outof 400000 tributary Indians countedin 1524, only 4864
remained in 1590; and in Costa Rica, out of 20 000 tributary Indians counted in
1563,only400remainedin1681(MacLeod,1973,pp.53,59,332).
7 A description of this episode can be found in the chapter `Costa Rican Cacao' of
MacLeod(1973,pp.33040).
8 TheproposalwascontainedinadocumentdraftedatECLAC'sMexicooffice,created
inOctober1951underthedirectionoftheMexicaneconomistVictorUrquidi.The
documentwasapprovedandpresentedbyECLAC'sExecutiveSecretaryRaulPrebisch
totheMinistersofEconomyofCentralAmerica,atthefirstmeetingoftheCentral
AmericanCommitteefor EconomicCooperation heldin Tegucigalpa, Honduras, in
August1952(ECLAC,1952).
9 These events and those that follow, which led to the creation of what was later
known as the Central American Common Market (CACM) are described in Cohen
(1971,pp.3948)andCohen(1972,pp.1340).
10 Thisconfirms Hirschman's assertionthat,bycontrastwiththeearlyindustrializers,
in the `late-late-comers' of Latin America, industrialization `has ordinarily brought
even less immediate and fundamental sociopolitical changes'. In his terms, `some-
how the existing structures in these countries seem to be better at absorbing and
accommodating the new industries and their promoters, technicians, and labour
forcethanwasthecaseinthosesocietieswhereindustryfirstraiseditshead'(Hirsch-
man,1971,p.331).
11 This term is used here only to emphasise the weight of the past, although it also
illustrates the more precise way in which Paul Krugman uses the term to describe
how `the outcome of market competition depends crucially on historical accident'
(Krugman,1994,p.223).
12 Bulmer-Thomas has said that in Central America a `virtuous circle' between the
export-led economy and the non-export economy has not been achieved. He
foundthat`thenon-exporteconomyremainsintenselyvulnerabletoadverseexter-
nalconditionsandisfarfromachievingself-sustainedgrowth'.
References
Bulmer-Thomas, V. (1987) Political Economy of Central America Since 1920, Cambridge:
CambridgeUniversityPress.
Cespedes,V.andGonzalez-Vega,C.(1993)`CostaRica',inS.Rottenberg(ed.),ThePolitical
Economy of Poverty, Equity, and Growth: Costa Rica and Uruguay, New York: Oxford
UniversityPressfortheWorldBank,pp.1183.
Cohen,I.(1972)RegionalIntegrationinCentralAmerica,Lexington:D.C.Heath&Co.,pp.
1340.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
(1971) `The United States as a Regional Power in Central America', in Annals of
InternationalStudies,Lexington:D.C.Heath&Co.,pp.3948.
Cohen,I.andRosenthal,G.(1983)`TheDimensionsofEconomicPolicySpaceinCentral
America',inR.FagenandO.Pellicer(eds),TheFutureofCentralAmerica:PolicyChoices
fortheU.S.andMexico,Stanford:StanfordUniversityPress,pp.1534.
ECLAC(1973)InformesobrelosdanosyrepercusionesdelterremotodelaCiudaddeManagua
enlaeconom anicaraguense(E/CN.12/AC.64/2/Rev.1).
ECLAC (1974) Informe sobre los danos y repercusiones del Huracan Fif en la econom a
hondurena(E/CEPAL/AC.67/Rev.1).
ECLAC (1976) Danos causados per el terremoto de Guatemala y sus repercusiones sobre el
desarrolloeconomicoysocialdelpa s(CEPAL/MEX/76/Guat.1).
ECLAC(1992)Centroamerica:evolucioneconomicadesdelaposguerrahastafinesdelosanos
setenta(LC/MEX/L.211),pp.23,16,81.
(1981) El impacto del incremento del precio de los hidrocarburos sobre las econom as del
IstmoCentroamericano(ECLAC/MEX/1036/Rev.1).
(1979)Nicaragua:Economicrepercussionsofrecentpoliticalevents(ECLAC/G.1091).
(1974)Laindustrializacionylosesquemasregionalesdecooperacioneconomica:elcasodel
MercadoComunCentroamericano(ECLAC/Mex/74/18).
(1952)IntegracionyreciprocidadeconomicaenCentroamerica,(E/CN.12/AC.17/3).
Hirschman, A. (1987) `The Political Economy of Latin American Development', Latin
AmericanResearch Review,vol.xxii,pp.736.
(1971)ABiasforHope,NewHaven,Conn.:YaleUniversityPress,pp.328341.
(1968) `The Political Economy of Import-Substituting Industrialization in Latin
America',TheQuarterlyJournalofEconomics,vol.82,pp.232.
Krugman,P.(1994)PeddlingProsperity,NewYork:Norton.
MacLeod,M.(1973)Spanish CentralAmerica:ASocioeconomicHistory,15201720,Berkeley,
California:UniversityofCaliforniaPress.
Monge R. and Lizano F. (1997) Apertura economica e industrializacion en Costa Rica, San
Jose:AcademiadeCentroamerica.
SIECA (1972) El desarrollo integrado de Centroamerica en la presente decada, Anexo 3: Pro-
gramadedesarrolloindustrialintegrado(SIECA/72VII-6/36).
UNCTAD (1967) Trade Expansion and Economic Integration among Developing Countries,
(TD/B/85/Rev.1)UNsalesno.67.II.D.20.
WorldBank(1990)TradeLiberalizationandEconomicIntegrationinCentralAmerica(report
no.7625CAM),pp.910.
(1971)ReportoftheIndustrialFinanceMissiontoCentralAmerica:TheCommonMarket
anditsFuture,Vol.1(mainreport),pp.17.
(1967)EconomicDevelopmentandProspectsofCentralAmerica,Vol.IV:Industry(report
no.WH-170a).
Yashar, D. (1992) `Demanding Democracy: Reform and Reaction in Costa Rica and
Guatemala,1870s1950s',PhDdissertation,Berkeley,California:UniversityofCalifor-
nia.
Zuvekas,C. (1996) ChangingPatternsin CentralAmerica's Exports,19801994, Lewisburg,
17thAnnualConferenceoftheMid-AtlanticCouncilofLatinAmericanStudies.
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Index
Ablin,E. 305,309t
Abreu,M.deP. 171t
acceleratedindustrialization 3,4,98
AgencyforInternationalDevelopment
(AID) 3278
agriculture 20,71,107,166
Argentina 2889,288f,310n3
Brazil 1623,162t,171
CentralAmerica 316,324,3267,
332
Chile 13941,150n18,1501n20
Colombia 248,249,257,260,261,
264t,26771,2723,276
Mexico 17980,195,202n28
Venezuela 209,211,221,230n13
Ahumada,J. 140,143
ALALC(LatinAmericanFreeTrade
Association)laterLatinAmerican
IntegrationAssociation 12,134
Aleman,M. 186
Alessandri,A. 125
Alessandri,J. 133,145,150n19
AllianceforProgress 6,12,106,107,126,
244
Alter,G. 112n1
Altimir,O.etal. 298t,299t,303
AndeanPact 12,15,134,218,262
Arellano,J.P. 132
Argentina
agriculture 2889,288f,310n3
atomicenergy 2867,310n2
BancodeCreditoIndustrial 284,303
capitalflight 295,295t
debt 295,304
DiTella 2901
economicpolicies 292,2934,295,304
exchangerates 294
exportpromotion 3056
exports 24t,291,305,306,3078,308f,
309t
IAPI(InstitutoArgentinoparalaPromocion
delIntercambio) 286
imports 2878,293,294
import-substitutingindustrialization
3069;1930sand1940s 2826,283t;
1950sand1960s 28691,289t;
1970s 2925;1980s 2956;
finance 3035
legalframework 297300
protection 3002
incomedistribution 293
industrialgrowth 292,292t,3079,
308f
inflation 293,295
INTA(InstitutoNacionaldeTecnologia
Agropecuaria) 288
investment 288,293,2989,298t,
299t,300,3023,305
KaiserArgentina 290,291
labourproductivity 286,302,307,
308f
labourshareinGDP 287t,295t
manufacturing 3078,308f
multinationalcorporations 289,290,
291,302,305,306
Peronism 2867,288,310n3
protection 155,157,283,285,294,
3002
provincialpromotionalregimes
299300
rawmaterialprocessingindustries 292,
293,296,297t,300,306
repression 293
RocaRuncimanTreaty 283
sectoralgrowth 296,296t,302
small/mediumenterprises 284,2856,
291,304
socialdevelopment 307
stateenterprises 2845,291
technology 284-5,290,291,292,
293,302,308,309t
termsoftrade 292
tradeliberalization 2934,3012,
303
transnationalcorporations 289,3023,
310n7
vehicleindustry 2901
Aristiguieta,R. 231n26
Artiagoit a,P. 149n2
Asia 4950,167
Astorga,P. 21f
Azpiazu,D. 300,302
335
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
336 Index
balanceofpaymentscrises 27
Balassa,B. 112
Balogh,T. 143
Baptista,A. 212f,221t,223f,224f,225f,
233n58
Bell,P.L. 208,229n7,229n9,232n38,
234n64
Berry,A. 255
Betancourt,R. 213,216,232n35
Brazil
agriculture 1623,162t,171
BNDE(NationalEconomicDevelopment
Bank) 165
coffeeeconomy 154,155,15760,
159t,166,169t,172
democracy 170
domesticmarket 155
economicgrowth 156,164
economicliberalism 157
economicpolicies 1645,168,16970
exchangerates 15,159,160,161,
1645
exports 24t,167,1689,169t
finance 165
GDPgrowth 161,162t,1701
importsubstitution 1545,172;before
1930 15660;1930toearly
1960s 1608;early1960stolate
1970s 16871;government
intervention 164
imports 156,157,161,165,1701
industrialgrowth 162t,1678
internationalfinancial
institutions 166,173n13
labourproductivity 163,163t, 164
lifeexpectancy 171t
PlanodeMetas 165,168
protection 1645
sectoralgrowth 161,162,162t, 169t
tariffs 155,1567,15860,1645,169,
172
VoltaRedonda 164
BrettonWoodsagreement 4,99100,
101,109
Brodersohn,M. 303
Bruton,H.J. 79
Bueno,G. 194
Bulmer-Thomas,V. 331,333n12
CABEI(CentralAmericanBankfor
EconomicIntegration) 327
CACMseeCentralAmerica:Common
Market
Calvodoctrine 5
capitalaccumulation 13,1312,2523,
256
capitaloutputratios 85,86t
Cardenas,E. 180,181t,182,185t,188f,
188t,196t
Cardoso,F.H. 82
CentralAmerica
195080 31415,3312,332n1
additivedevelopment 327
agriculture 316,324,3267,332
BankforEconomicIntegration
(CABEI) 327
CommonMarket(CACM) 12,16,25,
323,328
conflicts 325,329,330
debt 329
depression 329,3301
economicgrowth 31516,316t,324,
330,332
emigration 330
exchangerates 327
exports 319,324,331
finance 3201,324,3278,332
GeneralTreatyofEconomic
Integration 323
importsubstitution 325,3267,331
imports 31920,323,324
incomedistribution 324,324t
industry 316,317t,325,326
inflation 32930
infrastructure 316,317,322
institutions 323
intraregionaltrade 320,322,323,325,
326,326t
investment 321,322,324,327,328
manufacturing 320,326
multinationalcorporations 328
naturaldisasters 330
oil 329
`pathdependence' 329,331
politics 325,329
populationgrowth 315
protectionism 320,322,323
publicsector 328,332
RegimeofIntegrationIndustries 322,
323
regionalintegration 3212,324,325,
332
socialdevelopment 317,332
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
trade 320
transnationalcorporations 327
TripartiteTreaty 323
urbanization 315,333n4
CEPALseeECLA(UnitedNations
EconomicCommissionforLatin
America)
Cespedes,V. 318
change 314
Chapultepecconference 5
Chenery,H. 89,79,104,110
Chile
agriculture 13941,150n18,
1501n20
Alessandriadministration 1334,
1456,150n19
capitalaccumulation 1312
CentralBank 125
ChristianDemocratprogramme
1335,1467
CONDECOR 130
CONICYT 1289,129t
CORFO 15,53,1268,141,146,
149n8,14950n9,1501n20
`developmentfromwithin' 114,118,
1212,149n1
economicgrowth 11524,1467
economicpolicies 122,1325,1428,
1457,150n19
exchangerates 15,1301,1335,
1357,136t,1445
exports 24t,120,127,133,134,1445,
148,150n12,150n15
GDPgrowth 115t
importsubstitution 130,131
imports 131,133,134,1367,136t,
141
industrialgrowth 11824,137
industrialization 11415,1249,
1434
investment 116,116t,117
KleinSaksMission 131,143,144,
151n21
124t,149n5
manufacturing 120
monetaryinstability 117
nationaldevelopmentplan 126
neo-liberalexperiment 11819,120,
121,1223
politicaltransformation 1458
PopularFront 126
Index 337
productivityofcapital 116,116t,
11718,149n2,149n3,149n4
protection 131,135,136,137
researchanddevelopment 1289,
129t
sectoralgrowth 11920,119t
socialpolicies 1389,141,142,146,
150n17
stateinstitutes 121,128,129,149n6
stateroleinindustrialization:
19201930s 1246;productive
sector 1268,150n10,150n11;
technologicalcapacity 1289
technology 118,121,125,1289,129t
tradebalance 1223,122f,123f
tradepoliciesforindustrialization
12930;192955 1301;
195662 1313;196570 1335;
197382 1357
UnidadPopular 127,147
urbanization 138
coffee
Brazil 154,155,15760,159t,166,
169t,172
Colombia 242,249,258,2636,264t,
267,270,272
CostaRica 31819
ElSalvador 318
Guatemala 318
Inter-AmericanCoffeeAgreement 266
InternationalCoffeeAgreement 266
Venezuela 206,20910,229n2
Cohen,I. 327,331
ColdWar 1012,106,329,330
Colombia 23940
agrarianreform 2723
AgrarianReformInstitute
(INCORA) 270,273
agriculture 248,249,257,260,261,
264t,26771,276
Andeantrade 262
capitalaccumulation 2523,256
coffee 242,249,258,2636,264t,267,
270,272
commercialbanks 270
constitutionalreforms 246
debt 242,244
depression 242
economicgrowth 2402,2445,
278n2
25763
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
338 Index
Colombia(contd.)
exchangerates 15,245,2578,258f,
261,262,270
exportpromotion 15,33n24
exports 24t,242,243t,253,258,261,
263,269
finance 2545,255t,270
FondoFinancieroAgropecuario 273
FrenteNacional 256,2712,273,278n9
importsubstitution 250,253,258,263
imports 242,271
incomedistribution 256,2767,277t
industrialprices 260,262f
inflation 245
infrastructure 2478,249t,254
InstituteofAgriculturalTrade(Instituto
deMercadeoAgropecuario,
IDEMA) 270,271
institutions 254,255,270
InstitutodeFomentoIndustrial
(IFI) 2556
internalmarketintegration 2478,
249,254
internalmigration 249
investment 244,244f
ìnward-lookingdevelopment' 249,263
LaViolencia 240,265,271
labourmovement(unions) 256
labourproductivity 252,253t
mixedmodel 263
NationalCoffeeFund(FondoNacionaldel
Cafe) 266
NationalFederationofCoffeeGrowers
(FederacionNacionalde
Cafeteros) 2656,270
NationalPeasants'Association
(AsociacionNacionaldeUsuarios
Campesinos,ANUC) 272,273
populationgrowth 240,2734,277
poverty 277
protectionism 260,261t,2701
publicenterprises 247,2556
regionaldiversity 2478,249
RepublicaLiberal 256,257
ruralviolence 2713
sectoralgrowth 240,241t,242,2502,
251t,253t,256,260,261
socialdevelopment 247,2736,274t,
275t,277
stateintervention 2457,26970
structuralchanges 2402,24950,257
technology 267,268,270
termsoftrade 258f
tradepolicyindicators 259t,270
urbanization 240
VallejoPlan 261
Conde,C. 301
Contreras,L. 210,212
CORFO(CorporationofDevelopmentand
Production)seeChile:CORFO
CostaRica 332
coffee 31819
colonialperiod 318
debt 329
economicgrowth 316t
exports 24t
GDPcomposition 317t
Integration 323
intraregionaltrade 326t
socialdevelopment 317,319,328
seealsoCentralAmerica
Curacao 2089,229n9
Currie,L. 247
David,P. 51,54n4
Dean,W. 155
debt 28,29,30
Argentina 295,304
CentralAmerica 329
Colombia 242,244
CostaRica 329
Mexico 329
Venezuela 219,2323n49
defencesector 42
depression
CentralAmerica 329,3301
Colombia 242
Mexico 17880,195,199,202n29
Venezuela 20910
devaluation 26
developmentbanks 19,53,111
Brazil 165
Mexico 17980,181,190,202n19
seealsoChile:CORFO;Inter-American
DevelopmentBank(IDB);WorldBank
D az-Alejandro,C.F. 30,202n27
Dillon,D. 105,3223
ECLA(UnitedNationsEconomic
CommissionforLatinAmerica) 6,
32n6
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
economictheory 5860,91n5;
critiqueofimport-substitution
model 6973;import
substitutionasmodelof
accumulation 608
externalconstraint 80t;
PrebischReport 288,305;
state-ledindustrial
accumulation 8490;
technologicalprogressandindustrial
strategy 736
seealsoECLAC(UNEconomic
CommissionforLatinAmericaand
theCaribbean)
ECLAC(UNEconomicCommissionfor
LatinAmericaandtheCaribbean)
16,36
CentralAmerica 320,3212,3234,
325,330
role 1213
WorldBankviewof 105,110
seealsoECLA(UnitedNationsEconomic
CommissionforLatinAmerica)
economicdevelopment 11,1034
economicgrowth
Brazil 156,164
CentralAmerica 31516,316t,324,
330,332
Chile 11524,1467
Colombia 2402,2445,278n2
CostaRica 316t
ElSalvador 316t
foreignsectorvariables 1556
Guatemala 316t
Honduras 316t
LatinAmerica 1618,17f;convergence
patterns 18,18f
Mexico 182,189,192,196,197
Nicaragua 316t
world 68,7t
economicliberalization 8,157
economicthought 3640
Edwards,S. 83
ElSalvador
coffee 318
conflicts 325,329
economicgrowth 316t
emigration 330
exports 24t
GDPcomposition 317t
regionalintegration 322
Index 339
employment 201,689
endogenousgrowththeory 7880
Escobar,G. 233n57
exchangerates 256,33n24
Argentina 294
Brazil 15,159,160,161,1645
CentralAmerica 327
Chile 15,1301,1324,1357,136t,
1445
Colombia 15,245,2578,258f,261,
262,270
Mexico 179,184,186,187,188t,191,
197,198,199
Venezuela 219,220,2223,223f,227
exportpromotion 3,10,99,11112,
1667
Argentina 3056
Colombia 15,33n24
exports 223,23f, 24t, 25
Argentina 291,305,306,3078,308f,
309f
Brazil 167,1689,169t
CentralAmerica 319,324,331
Chile 120,127,132,134,1445,148,
150n12,150n15
Colombia 242,243t,253,258,261,
263,269
Mexico 24t,1824,185
externalfinancing 27,27f
Fajnzylber,F. 82
Feldman,E. 303
Felix,D. 143
Ffrench-Davis,R. 115t,116t
fiscaldeficit 87,89,94n47
fiscalstructure 87,88t,89,89t,94n45
Fishlow,A. 155,163t
Fitzgerald,E.V.K. 21f, 88t
Flanders,M.J. 70
Furtado,C. 745,89,89t,94n45,155,
234n58
Gallegos,R. 214
Garay,L.J.etal. 261t
Garc a,D. 149n8
Gardner,R. 1012
Garner,R. 103,112n1
GeneralAgreementonTariffsandTrade
(GATT) 5,100
GeneralizedSystemofPreferences 8
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
340 Index
goldstandard 512,283
Go mez,J.V. 209,210,229n10,230n11,
230n18
Gonzalez-Vega,C. 318
government
institutions 52
interventioninindustrialization 823
nationalisticmilitary 42,43,53
Ground,R.L. 80t
Guadagni,A. 303
Guatemala 315
coffee 318
colonialperiod 318
conflicts 329
economicgrowth 316t
exports 24t
GDPcomposition 317t
industry 323
Haddad,C. 162t
HavanaCharter 32n4
heavyindustries 42
HeckscherOhlinSamuelson(HOS)
model 70,76,77
Herrera,L. 217,219
Herrera,O. 272
Hirschman,A.O. 812,229n1,314,
332n2,333n10
Hobsbawm,E.J. 149n7
Hofman,A.A. 86t
Honduras
conflicts 325
economicgrowth 316t
exports 24t
GDPcomposition 317t
Hutchenson,T.L. 260,261t
IBECseeInternationalBasicEconomic
Corporation
IBRD(InternationalBankfor
ReconstructionandDevelopment)see
WorldBank
IDAseeInternationalDevelopment
Association
IDBseeInter-AmericanDevelopmentBank
IFIsseeinternationalfinancialinstitutions
IMFseeInternationalMonetaryFund
imports
Argentina 2878,293,294
Brazil 156,157,161,165,1701
Chile 132,133,134,1367,136t,141
Colombia 242,271
Mexico 182,183,186,188f,193,195,
198
Venezuela 212,217
import-substitutingindustrialization
(ISI) 23,9,1011
Argentina 282309
benefits 401
bias 229n1
Brazil 15472
centre/peripherydevelopment 745
Chile 130,131
Colombia 250,253,258,263
critiqueoftheory 6973;neglectof
comparativeadvantage 701;
sectoralimbalance 712
fiscalweakness 90
genesisofstrategy 589,734
influenceofIDB 100,105,1089
influenceofIFIs 99,100
influenceofIMF 100,110
influenceofWorldBank 100,1045,
109,110,111
institutionaldevelopment 523
institutionalframework 434
Mexico 180,181t,182,183,185,
1867,1929,202n15,202n27
microeconomics 4550
modelofaccumulation:constraintson
growth 636;employment 689;
externalconstraint 668,80t,845;
open-economymodel 613;
origins 601
newtradetheories 768
periodization 845
aspolicyregime 989
politicaleconomy 31,804
state-ledindustrial
accumulation 8490
technologicalprogressandindustrial
strategy 736
totalfactorproductivity(TFP)
growth 7880,85
transitorypolicies 423
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
Venezuela 208,216,217,21819,
231n26
seealsoindustrialgrowth
incomedistribution 212
Argentina 293
CentralAmerica 324,324t
Colombia 256,2767,277t
Mexico 17980,184,185,193,197
industrialgrowth 99
Argentina 292,292t,3079,308f
Brazil 162t,1678
CentralAmerica 316,317t,325,326
Mexico 180,181t,184,185t,188t,189,
198
industrialplants 436
industrializationseeimport-substituting
industrialization(ISI);industrial
growth;state-ledindustrialization
inflation 267,678
Argentina 293,295
CentralAmerica 32930
Colombia 245
Mexico 191,197,198
Nicaragua 330
infrastructure
Colombia 2478,249t,254
Mexico 184
Venezuela 207,215,225f,226,228,
229n5
institutionaldevelopment 19,503
institutions 501,54n4
Inter-AmericanCoffeeAgreement 266
Inter-AmericanConferences 5
Inter-AmericanDevelopmentBank
(IDB) 6,98
establishment 1057
exportfinanceprogramme 109
influenceonISI 100,105,1089
lendingpolicy 106,1079
procurementprocedures 109
Inter-AmericanEconomicandSocial
Council 6,105
InternationalBankforReconstruction
andDevelopment(IBRD)seeWorld
Bank
InternationalBasicEconomicCorporation
(IBEC) 21314,231n28
InternationalCoffeeAgreement 266
InternationalDevelopmentAssociation
(IDA) 103,106,107
InternationalFinanceCorporation 105
Index 341
Internationalfinancialinstitutions(IFIs)
Brazil 166,173n13
developmentpolicy 102
influenceonISI 99,100
UScontrol 102,1067
InternationalMonetaryFund(IMF) 4,98
inColdWar 1012
influenceonISI 100,110
protectionism 6,11011
purpose 100
InternationalTradeOrganisation(ITO) 5,
32n4
investment
Argentina 288,293,2989,298t,299t,
300,3023,305
CentralAmerica 321,322,324,327,328
Chile 116,116t,117
Colombia 244,244f
disincentive 867
Mexico 1801,184,189,190,192,195,
196t,198,202n25
private 87,89,90,94n45
public 87
rates 85,86t,93n43
Venezuela 217,21819
ISIseeimport-substituting
industrialization
ITOseeInternationalTradeOrganisation
Itzcovich,S, 303
James,H. 100
Jimenez,P. 215
Jorgensen,S.L. 26
Kaldor,N. 132,140,143
Kalmanovitz,S. 269
Katz,J. 305,309t
Konig,W. 100
Kornblith,M. 225f
Kosacoff,B. 302
Krugman,P. 333n11
labourforce 201
labourorganization 48
labourproductivity 1819
Argentina 286,302,307,308f
Brazil 163,163t, 164
Chile 1201,122,123,124t,149n5
Colombia 252,253t
Mexico 1812,184,189,190,192
Venezuela 214t,215,2234,224f,
231n30,233n56
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
342 Index
labourunions 53,189,256,272,273
LatinAmericanFreeTradeAssociation
(ALALC)laterLatinAmerican
IntegrationAssociation 12,134
Leloir,F. 310n7
Leniz,S. 119t,124t
Lewis,A. 104,150n18
Little,I. 59,712
LlerasRestrepo,C. 270,272,273
Londero,E. 778
Londono,J.L. 276
Lucangeli,J. 3001
Lynn,J. 108
macroeconomicmanagement 289,30
Maddison,A. 7t
Maingo n,T. 225f
Mamalakis,M. 145
manufacturing
Argentina 3078,308f
Chile 120
Colombia 241t,2502,251t,253t
Venezuela 20810,212f,21315,214t,
21617,219,2202,221t,228
seealsoindustrialgrowth
Marcel,M. 119,119t
Marfan,M. 149n2
MarshallPlan 5,1012,166
MartinezdeHoz 293,301
McBeth,B.S. 2301n22
McNamara,R. 104,110
Medina,General 212
Meller,P. 119,119t
Mendoza,E. 216
Mexico
agriculture 17980,195,202n28
BancodeMexico 178,179,182,186
BancoNacionaldeComercioExterior 181
BancoNacionaldeCreditoEjidal 17980
CEIMSA/CONASUPO 189
debt 195,199,202n29,329
depression 17880
economicgrowth 182,189,192,196,
197
199,2001
exchangerates 179,184,186,187,
188t,191,197,198,199
exportexpansion 24t,1824,185,198
government 179,180,182,186,196,
199
importsubstitution 180,181t,182,
183,185,1867,1929,202n15,
202n27
imports 182,183,186,188f,193,195,
198
incomedistribution 17980,184,185,
193,197
industrialcompetition 111
industrialgrowth 180,181t,184,185t,
188t,189,198
industrialization 1768;1930s
17882;194045 1825;
194662 18590;196381 1919,
199f
inflation 191,197,198
infrastructure 184
investment 1801,184,189,190,192,
195,196t,198,202n25
190,192
labourunions 189
maquiladoraprogramme 15
NacionalFinanciera 181,190,202n19
oil 197,198
protection 1867,190,1919,201n13
sectoralgrowth 1845,185t,191,200t
state-ownedenterprises 190,191f,195,
198
termsoftrade 183,187,201n14
tradedeficits 186,192,195
transnationalcorporations 181,187,
1889
USinfluence 179,181,183,1856
migration
Colombia 249
ElSalvador 330
Venezuela 216,225f,2267,234n63,
234n64
MisionEconom ayHumanismo 248
Mitchell,B.R. 225f
`mixedmodel' 3,1011
monetarypolicy 51
Montalva,F. 140
Argentina 289,290,291,302,305,306
CentralAmerica 328
Colombia 255
seealsotransnationalcorporations
Munoz,O. 115t,116t
nationalism 42,43,53
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
nationalization 523
neo-classicaleconomics 3840,478,49,
50,545n5
Nicaragua
ColdWar 329,330
economicgrowth 316t
exports 24t
GDPcomposition 317t
Integration 323
industry 323
inflation 330
Nixon,R.M. 106
Nogues,J. 301
Ocampo,J.A. 258f,259t, 264t,274t,277t
oil
CentralAmerica 329
Mexico 197,198
Venezuela seeVenezuela:oil
OrganizationofAmericanStates
(OAS) 3223
Packenham,R. 101
Paldam,M. 26
Pedregal,G.del 149n8
periodization 845
Pero n,J. 55n9,284,292,298
Peru 15,24t
Pinto,A. 1412,143
politicaleconomy 16,301,812,823
CentralAmerica 315
Colombia 240,2734,277
CostaRica 318
poverty 212,107,277
Prebisch,R. 8,12,613,70,81,105,323
PrebischReport 288,305
PrebischSingerthesis 70
productivitygrowth 7880,85
protectionism 56,7,8,10,245,51,83,
167
anti-protectionism 100
Argentina 155,157,283,285,294,
3002
Brazil 155,1567,15860,1645,169,
172
Chile 131,135,136,137
Colombia 260,261t,2701
Index 343
ECLAC 12,13
IMF 6,11011
Mexico 1867,190,1919,201n13
Venezuela 211,216,219,231n32
WorldBank 56
publicsectorbias 42
Ram rez,P. 125
regionalintegration 25,61
CentralAmerica 3212,324,325,332
Rist,L. 112n1
Rockefeller,N. 213,232n35
Rodr guez,O. 845
Rosenthal,G. 327,331
Rozas,M. 119t,124t
Saez,R. 149n8
Schumpeter,J. 37
Schvarzer,J. 297t,303
Scitovsky,T. 59,712
Scott,M. 59,712
sectoralgrowth
Argentina 296,296t,302
Brazil 161,162,162t, 169t
Chile 11920,119t
Colombia 240,241t,242,2502,251t,
253t,256,260,261
Mexico 1845,185t,191,200t
Venezuela 2214,221t
securityemergency 106,107
Simian,E. 149n8
Argentina 307
Brazil 171,171t
Chile 1389,141,142,146,150n17
Colombia 247,2736,274t,275t,277
CostaRica 317,319,328
livingstandards 21f
Venezuela 211,225f,226,228,
234n62
sociallending 106,107
SocialProgressTrustFund 6,106,107
Solimano,A. 80,83
Sol s,L. 1934
Solow,R. 389
Sourrouille,J.V.etal. 302
state-ledindustrialization 34,14
alternatives 1316
benefits 1622
Chile 1249
classification 89
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
344 Index
state-ledindustrialization(contd.)
costs 229
demiseofthemodel 2931
externalcontext 48
internalcontext 812
Mexico 1825,1979
roleofECLAC 1213
stages 911
Venezuela 20719
structuralism 367,40,59,73
Sunkel,O. 143
tariffsseeprotectionism
Taylor,A.M. 91n5,934n43
technology
Argentina 2845,291,292,293,302,
308,309t
Chile 118,121,125,1289,129t
Colombia 267,268,270
developmentoflocalcapabilities
4250
industrialplants 456
andindustrialstrategy 736
in-houseproductivity 46,48
innovation 13,19,37,39
institutionalframework 434
andlevelsofprotection 167
microeconomicbehaviour 4650,
47f
newtradetheories 768
researchanddevelopment 523,
1289,129t,2845,290
Teitel,S. 778
totalfactorproductivity(TFP)
growth 7880,85
trade
Argentina 2924,3012,303
CentralAmerica 320
Chile 1223,122f,123f,12937
Colombia 258f,259f,270
liberalization 100
Mexico 183,186,187,192,195,
201n14
theories 768
world 78,22
transnationalcorporations 44,21516
Argentina 289,3023,310n7
CentralAmerica 327
Mexico 181,187,1889
seealsomultinationalcorporations
UNConferenceonTradeand
Development(UNCTAD) 323
UNEconomicCommissionforLatin
America seeECLA
UNEconomicCommissionforLatin
AmericaandtheCaribbean see
ECLAC
UnitedStatesofAmerica
AID 3278
bilateralaid 102,166,173n12
andCentralAmerica 322,328,330,
3312
ColdWar 1012,106,330
influenceonMexico 179,181,183,
1856
andinternationalfinancialinstitutions
(IFIs) 102,1067
post-waragenda 46
production 4
tradeagreements 211,230n21,
231n32
andVenezuela 211,229n9,230n21,
2301n22,231n32
urbanization 12
CentralAmerica 315,333n4
Chile 138
Colombia 240
Mexico 191,202n21
Urrutia,M. 275t
Vasena,K. 301
Velazquez,F. 189
Venezuela 2056,228
agriculture 209,211,221,230n13
AndeanPact 218,262
BancoCentraldeVenezuela 211
BancoIndustrialdeVenezuela 211
coffee 206,20910,229n2
communications 207,215,226,229n5
CorporacionVenezolanadeFomento
(CVF) 213,215,216,21819,
232n35
CorporacionVenezolanadeGuagana
(CVG) 217,232n41
CorporacionVenezolanadePetroleo
(CVP) 217
debtcrisis 219,2323n49
depression 20910
`Dutchdisease' 2201
economicpolicies 210,211,215,217,
21819,230n17
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6
education 211,225f,226,234n62
exchangerates 219,220,2223,223f,
227
exportdiversification 21718
exports 24t
government 210,212,213,214,
215,216,217,229n10,
230n11,230n18,2301n22,
232n44
GroupCorimon 216
IBEC(InternationalBasicEconomic
Corporation) 21314,231n28
immigration 216,225f,2267,
234n63,234n64
importsubstitution 208,216,217,
21819,231n26
imports 212,217
industrialization 207;
192035 20710;193647 21014;
194883 21419
infrastructure 207,215,2256,225f,
228,229n5
institutions 211,217,218
investment 217,21819
ironandsteel 216,218,232n35
labourproductivity 214t,215,2234,
224f,231n30,233n56
manufacturing 20810,212f,21315,
214t,21617,219,228;and
oil 2203;structural
changes 2202,221t
oil 2069,21112,229n8,229n9,
230n11,232n40;exports 21112;
HydrocarbonsLaw 215,2312n33;
andindustry 2202;
nationalization 218;refining 215,
Index 345
229n8,229n9,230n11,231n29;
rent 225f,2278
protection 211,216,219,231n32
sectoralgrowth 2214,221t
socialdevelopment 211,228(seealso
Venezuela:education)
stateenterprises 216,218
transnationalcorporations 21516,
231n32
UStradeagreement 211,229n9,
230n21,2301n22,231n32
VenezuelaTrustFund 109
Villanueva,J. 282
Villareal,R. 65,200t
Vollmer,A. 216
WorldBank(IBRD) 4,98,100,247
balance-of-payments
problems 10910,112
CentralAmerica 315,317,320,325,
326,327,328,333n4
inColdWar 1012
economicresearch 103
exportpromotion 111
influenceonISI 100,1045,109,110,
111
inLatinAmerica 101,103,1045
lendingpolicy 1024,106,10910,112
procurementprocedures 105
projectappraisal 105
protectionism 56
purpose 100
relationswithECLAC 105,110
Zuvekas,C. 329
C
o
p
y
r
i
g
h
t

m
a
t
e
r
i
a
l

f
r
o
m

w
w
w
.
p
a
l
g
r
a
v
e
c
o
n
n
e
c
t
.
c
o
m

-

l
i
c
e
n
s
e
d

t
o

Q
4

P
u
s
h

-

P
a
l
g
r
a
v
e
C
o
n
n
e
c
t

-

2
0
1
0
-
1
2
-
0
6

An Economic History of Twentiet - Enrique Cardenas

Enviado por

Dados do documento

Descrição original:

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

An Economic History of Twentiet - Enrique Cardenas

Enviado por

Direitos autorais:

Formatos disponíveis

An Economic History of Twentieth-Century Latin America

) required to reduce unemployment.

Você também pode gostar