European Economic Review 50 (2006) 12971315

Ination targeting regimes

Alina Carare

, Mark R. Stone
International Monetary Fund, 700 19th St. NW, Washington DC 20431, USA
Received 19 December 2003; accepted 22 February 2005
Available online 5 May 2005
Abstract
This paper divides the monetary frameworks of countries that use an ination target to dene
their monetary framework into three different regimes: (i) full-edged ination targeting, (ii)
implicit price stability anchor, and (iii) ination targeting lite. The regimes are differentiated by
the clarity and credibility of the commitment to the ination target. The revealed preference for
different regimes is related empirically to cross-country structural differences in economic and
nancial development. Policy implications of moving from one regime to another are drawn,
especially for emerging market countries aiming at full-edged ination targeting.
r 2005 Elsevier B.V. All rights reserved.
JEL classication: E52; E58; E61
Keywords: Monetary policy framework; Ination targeting; Central bank
1. Introduction
An ination target is being used by an increasing number of countries to dene
their monetary framework. These countries choose to adopt a exible exchange rate
to limit their vulnerability to an exchange rate attack and to maintain an
independent monetary policy. At the same time, a monetary target is not practical
owing to instability in money demand. In 2001 some 42 medium and large country
central banks had some sort of a oating exchange rate arrangement, leaving their
degree of commitment to an ination target as the dening monetary objective.
ARTICLE IN PRESS
www.elsevier.com/locate/eer
0014-2921/$ - see front matter r 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.euroecorev.2005.02.003

Corresponding author. Tel.: +1 202 623 9749; +1 202 623 6532; fax: +1 202 623 6067.
E-mail addresses: acarare@imf.org (A. Carare), mstone@imf.org (M.R. Stone).
This paper classies these countries into three separate ination targeting regimes
to gain insights into the appropriate design of monetary policy conditional on a
countrys circumstances. The regimes are dened by the clarity and credibility of the
central banks commitment to the ination target. Clarity is gauged by the public
announcement of the ination target and by the institutional arrangements in
support of accountability to the target. Credibility is proxied by the actual ination
outturn and by market ratings of long-term local currency government debt. The
three regimes are marked by qualitatively differentand intuitivecombinations of
self-reported combinations of commitment and discretion.
Countries in the rst regime practice full-edged ination targeting (FFIT), which
is the best-known form of ination targeting. FFIT countries have a medium to high
level of credibility, clearly commit to their ination target, and institutionalize this
commitment in the form of a transparent monetary framework that fosters
accountability of the central bank to the target. New Zealand was the rst country
to adopt FFIT, and by 2001 some seven industrial and eleven emerging market
countries were practicing this regime.
Implicit price stability anchor (IPSA)
1
countries have so much credibility that they
can maintain low and stable ination without full transparency and accountability
with respect to an ination target. Their record of low and stable ination and high
degree of nancial stability affords them the exibility to pursue the objective of
output stabilization, as well as price stability. Five developed country central banks
are classied here as practicing IPSA, including the European Central Bank and the
Board of Governors of the Federal Reserve System.
Ination targeting lite (ITL) countries announce a broad ination objective but
owing to relatively low credibility are not able to maintain ination as the foremost
policy objective. Their relatively low credibility reects their vulnerability to large
economic shocks and nancial instability and a weak institutional framework. The
number of ITL countries is 19 and all are emerging market countries.
The ination targeting regimes can be viewed as reecting a revealed preference of
individual countries for different monetary frameworks conditional on their
economic structure. The empirical analysis of this paper suggests that there are
systematic differences in economic structure across the three regimes. In particular,
GDP per capita and the level of nancial development are highest for IPSA countries
and lowest for the ITL countries. The different combinations of credibility and
discretion across the three regimes seem to reect these structural differences.
The analysis also offers some practical guidance for countries considering a switch
from one regime to another. The main policy implications of this paper are for
emerging market countries moving from ITL to FFIT. Econometric analysis
suggests that these switches are facilitated by a deep and broad nancial sector,
which reduces systemic risks and potential policy conicts, provides for market-
based monetary policy implementation, and allows the government to raise the bulk
of its funding in nancial markets.
ARTICLE IN PRESS
1
In a previous draft of the paper we refer to this group as an eclectic ination targeting (EIT) regime,
Carare and Stone (2003).
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1298
This paper builds on the policy-oriented ination targeting literature. Detailed
discussions of pros and cons of FFIT and country experiences can be found in
Bernanke et al. (1999), Mishkin and Schmidt-Hebbel (2001), and Truman (2003).
FFIT for emerging market countries is the subject of Masson et al. (1997), Mishkin
(2000), Schaechter et al. (2000), Blejer et al. (2001), and Carare et al. (2002). Studies
of the determinants of the adoption of FFIT versus regimes for selected countries
include Gerlach (1999), Schaechter et al. (2000), and Mishkin and Schmidt-Hebbel
(2001). This paper appears to be the rst to dene and analyze three ination
targeting regimes.
The paper is organized as follows. The selection of the ination targeting countries
and their classication into three regimes is described in the next section. Self-
reported regime indicators are reported in Section 3. The structural underpinnings of
the different ination targeting regimes are examined empirically in Section 4.
Regime switches are discussed in Section 5, and Section 6 concludes.
2. Classication of ination targeting countries into three regimes
This section describes the selection of the IMF member countries that operate with
an ination target and their classication into three regimes. The selection and
classication is based on the design of the monetary regime in March 2001 and
involves three steps.
First, small and less-developed countries are eliminated because they face a
different set of monetary regime options. They have undeveloped nancial sectors
and concentrated production proles and so they tend to choose a xed exchange
rate or adopt the currency of their largest trading partner. Thus, the 89 countries
with GDP under US$4 billion and countries with per capita GDP less than $720 as
of 2000 were dropped from the 185 IMF member countries (data are from the World
Economic Outlook database of the IMF and are available from the authors). In
addition, Belarus, Paraguay, Trinidad and Tobago, and Tunisia were dropped owing
to lack of data, and Turkey because it was in the midst of a nancial crisis during
2001.
Second, the ination targeting countries were selected based on the presumption,
which is documented below, that countries with a oating exchange rate make a
commitment to an ination target. Therefore, countries with one of the ve xed
exchange rate arrangements in the International Financial Statistics (IFS) of the IMF
were dropped. This leaves 42 central banks with some sort of a oating exchange
rate, which encompass the arrangements of independently oating, managed oating
with no preannounced path for exchange rate, and exchange rates within crawling
bands (Table 1). Although those countries have ination targeting in common, they
cover a wide range of development levels and vulnerability to shocks.
2
These
ARTICLE IN PRESS
2
The countries included in our sample have all oating exchange rate regimes, as based on the March 31,
2001 Exchange Rate Arrangements and Anchors of Monetary Policy, published in IMFs International
Financial Statistics. This classication is based on de facto practices.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1299
ARTICLE IN PRESS
Table 1
Ination targeting countries, exchange regimes and clarity of commitment
a
Exchange rate regime Clarity of commitment
to ination target
Albania Independently oating Not clear
Algeria Managed oat Not clear
Australia Independently oating Clear
Brazil Independently oating Clear
Canada Independently oating Clear
Chile Independently oating Clear
Colombia Independently oating Clear
Croatia Managed oat Not clear
Czech Republic Managed oat Clear
Dominican Rep Managed oat Not clear
Guatemala Managed oat Not clear
Honduras Exchange rates within crawling bands Not clear
Hungary Exchange rates within crawling bands Clear
Iceland Independently oating Clear
Indonesia Independently oating Not clear
Israel Exchange rates within crawling bands Clear
Jamaica Managed oat Not clear
Japan Independently oating Not clear
Kazakhstan Managed oat Not clear
Korea Independently oating Clear
Mauritius Independently oating Not clear
Mexico Independently oating Clear
New Zealand Independently oating Clear
Norway Managed oat Clear
Peru Independently oating Not clear
Philippines Independently oating Not clear
Poland Independently oating Clear
Romania Managed oat Not clear
Russian Fed Managed oat Not clear
Singapore Managed oat Not clear
Slovak Republic Managed oat Not clear
Slovenia Managed oat Not clear
South Africa Independently oating Clear
Sri Lanka Managed oat Not clear
Sweden Independently oating Clear
Switzerland Independently oating Not clear
Thailand Independently oating Clear
United Kingdom Independently oating Clear
United States Independently oating Not clear
Uruguay Exchange rates within crawling bands Not clear
Venezuela Exchange rates within crawling bands Not clear
Sources: Central bank websites, IMF reports, Schaechter et al. (2000).
a
The source of this table for the exchange rate regime is Exchange Rate Arrangements and Anchors of
Monetary Policy, as of March 31 2001, International Financial Statistics, IMF, which does not contain
information about the exchange rate regime of the EMU.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1300
differences suggest that treating ination targeting countries as a homogenous group
would not be appropriate for understanding how to tailor the design and operation
of monetary policy to a countrys circumstances.
Ination targeting countries are further classied by the clarity and credibility of
their commitment to their ination target. Clarity is an indispensable facet of the
ination targeting monetary policy framework because adherence to an ination
objective can be observed only with long and variable lags owing to the time between
a change in the policy stance and its impact on ination. Clarity is less important
under an exchange rate target framework because the target is easily observed in real
time.
The clarity of an ination target can be viewed as consisting of two elements. The
rst element is the authorities public description and communication of the ination
target. The central bank descriptions of their ination targeting objective are from
either their own websites, or from IMF country reports, publication of which is
agreed to with the country governments (the sources are available from the authors).
The second element is the transparency of the institutional framework, which is
needed for the public and markets to hold the central bank accountable to its
ination target. Institutional transparency is gauged by the communication vehicles
employed by the central bank, including the release of ination reports and the
frequency and detail of these reports, the announcement of changes in the stance of
monetary policy via press release, reviews of ination performance and changes in
monetary policy, the publication of ination forecasting models, and the use of
media and other public presentations. These communication vehicles are documen-
ted in Schaechter et al. (2000), Schmidt-Hebbel and Tapia (2002), and Fracasso et al.
(2003), which assesses ination target reports in detail.
Examination of these two measures of clarity leads to the separation of the 42
ination targeting countries into groups that do and do not make a clear
commitment to the ination target (counting the 12 EMU members as a single
country) (Table 1). The clear commitment group consists of 18 countries that make
an explicit commitment to an ination target and implement a transparent
framework to ensure that the central bank is accountable for the target. They have
numerical ination targets expressed as a point target or as a range dened in terms
of end-year ination, average annual ination, or, increasingly in recent years, over
an indenite horizon (Roger and Stone, 2005).
The other 24 countries do not explicitly commit to an ination target and have
some other stated objectives as well. They announce some sort of ination objective
or intention to aim at general price stability, with 13 of the countries announcing a
numerical ination target. Many explicitly specify other objectives, such as the
nominal exchange rate, real exchange rate, and international reserves. The
multiplicity of targets and their imprecise denition renders their commitment to
an ination target much less clear compared to the rst group of ination targeting
countries. These countries are a strikingly more diverse group than that of the clear
commitment countries in terms of size and the level of development. Indeed, the wide
range of countries that choose a less explicit commitment to an ination target is a
puzzle which raises important questions regarding how a country should design its
ARTICLE IN PRESS
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1301
monetary policy framework conditional on its circumstances. Are the determinants
of the monetary policy framework for the US really the same as for Albania and Sri
Lanka?
The countries in this second non-explicit group of ination targeting countries are
further divided by the credibility of the commitment to low ination. The rationale is
that a country with a clear commitment to an ination objective may not be viewed
as a credible ination targeter if ination is high and the markets have a poor
perception of its central bank vis-a` -vis other ination targeting countries.
Credibility is measured here using two gauges. The rst credibility gauge is simply
the actual rate of ination. Low and positive ination is supportive of high and
stable long term growth (e.g. Sarel, 1996) and a monetary policy supportive of long
run growth can be viewed as more credible.
3
The time period over which ination is
reported here is relatively recentJanuary 1999May 2002because many of the
ination targeting countries have only recently changed their monetary regime and
because historical data is less useful for forward looking policy analysis. The data are
monthly changes in the seasonally adjusted CPI index reported at an annual rate.
The wide dispersion of ination indicates that credibility varies quite a lot across
countries with a less clear commitment to an ination target (rst three columns of
Table 2). Of course, ination over a relatively short time period will reect
exogenous shocks as well as credibility, and, to the extent that the vulnerability to
shocks differs across country groups, the comparisons are less informative in the
absence of a formal model. Still, the wide ination range suggests qualitative
differences within this group of countries.
The second gauge of credibility is the debt rating of long-term local currency
denominated government debt. The ratings address some of the pitfalls of using
ination as a measure of credibility in that they are forward-looking and directly
capture market perceptions of the degree of long-term market condence in the
stability of a currency, which ultimately is the responsibility of the central bank. At
the same time, the ratings reect factors beyond the scope of monetary policy,
especially the strength of the scal position. Although this is not under the direct
control of the central bank, it also bears on the credibility of a commitment to an
ination target.
Again, the dispersion of this gauge of credibility is quite wide across the 24
ination targeting countries that do not make a clear commitment to their ination
target. The Standard and Poors long-term local currency government debt ratings
for 2001 are used here (middle three columns of Table 2). The highest value debt
rating, AAA, is equivalent to one, in our second credibility indicator order.
Countries with no rated debt are assigned a value of twenty, and countries in
selective default, have a rank of 19. The United States, Singapore, and Switzerland
enjoy an AAA rating, whereas the debt of ve countries is not rated, and one is in
selective default.
ARTICLE IN PRESS
3
Market-based measure of central bank credibility would be preferable, but such measures are available
only for a few countries. In addition, comparisons of actual with targeted ination are precluded by the
absence of rm quantitative targets for many of the countries.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1302
Construction of an overall ranking of credibility separates into two distinct groups
the 24 countries that do not make a clear commitment to an ination target. The
overall ranking of credibility is formed by constructing a simple average of the
ination and rating rankings (last two columns of Table 2). A high credibility group
consists of Singapore, Switzerland, Japan, the United States and the EMU. The
largest gap between successive index values is that between the EMU (5) and
Croatia/Peru/Philippines (9). A relatively low credibility group consists of 19
countries with values between 9 and 20. The distinction between the low and high
credibility countries would seem to point to useful policy implications regarding
design of the monetary framework.
Examination of the clarity and credibility of the commitment to the ination
target leads to the separation of the ination targeting countries into three regimes
ARTICLE IN PRESS
Table 2
Indicators of credibility, selected ination targeting countries
Ination, Jan. 1999May 2002 S&P long-term domestic currency
Government debt rating, 2001
Country Average Rank Country Rating Rank Average rank
Japan 0.9 1 United States AAA 1 High credibility
Singapore 0.6 2 Switzerland AAA 1 Singapore 1.5
Switzerland 1.3 3 Singapore AAA 1 Switzerland 2.0
Algeria 2.1 4 EMU AAA- 4 Japan 3.0
Peru 2.4 5 Japan AA- 5 United States 4.0
EMU 2.6 6 Slovenia A 6 EMU 5.0
United States 2.6 7 Slovak Republic BBB- 7
Albania 3.7 8 Croatia BBB- 7 Low credibility
Philippines 4.8 9 Philippines BB+ 9 Croatia 9.0
Uruguay 4.9 10 Jamaica B+ 9 Peru 9.0
Croatia 4.9 11 Kazakhstan BB 11 Philippines 9.0
Mauritius 5.6 12 Guatemala BB 11 Uruguay 11.5
Guatemala 5.7 13 Uruguay BB- 13 Slovenia 11.5
Dominican Rep 6.9 14 Peru BB- 13 Algeria 12.0
Jamaica 7.2 15 Dominican Rep BB- 13 Guatemala 12.0
Indonesia 8.2 16 Russia B+ 16 Jamaica 12.0
Slovenia 8.4 17 Romania B+ 16 Slovak Rep. 12.5
Slovak Rep 9.3 18 Venezuela B 18 Dominican Rep. 13.5
Honduras 9.5 19 Indonesia SD 19 Albania 14.0
Sri Lanka 10.6 20 Sri Lanka No rate 20 Kazakhstan 16.0
Kazakhstan 11.6 21 Mauritius No rate 20 Mauritius 16.0
Venezuela 17.1 22 Honduras No rate 20 Indonesia 17.5
Russia 26.2 23 Algeria No rate 20 Honduras 19.5
Romania 40.4 24 Albania No rate 20 Russia 19.5
Romania 20.0
Sri Lanka 20.0
Venezuela 20.0
Sources: IFS, Bankscope.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1303
(Table 3). The rst group of ve central banks consists of those that are highly
credible and make a less clear commitment to an ination target. These countries
share highly successful ination records, but they are rather heterogeneous in
their monetary policy frameworks with regard to the denitions of price stability
and the operation of monetary policy. They seem to have extra scope for exibility
to attain objectives other than an ination target and thus they are called implicit
price stability anchor (IPSA) countries. All the IPSA practitioners are industrial
countries.
The second regime consists of the 18 countries that make an explicit commitment
to an ination target and implement a transparent framework to ensure that the
central bank is accountable for the target. These countries are called full-edged
ination targeters because they make a full commitment to their ination target.
Interestingly, the FFIT countries are either small or medium-sized industrial
countries, or medium and large middle-income emerging market countries.
Countries in the third regime announce an ination objective, but owing to a low
level of credibility do not make a clear commitment. They also are relatively
heterogeneous in the objectives and operation of monetary policy. This regime is
called ination targeting lite because these countries are not able to make a credible
commitment to an explicit ination target. The number of ITL countries is 19, and
all are emerging market countries.
3. Monetary policy objectives
This section looks at self-reported indicators of monetary policy objectives across
the three ination targeting regimes to gauge whether this new classication captures
real-life differences in monetary policy. Table 4 summarizes the within-regime
averages of commitment and discretion on a scale of 0100. These measures are
ARTICLE IN PRESS
Table 3
Ination targeting countries, credibility gauges
Regime Ination S&P long-term
Jan. 1999May 2002 domestic currency Govt.
debt rankings 2001
IPSA 1.2 2.0
IPSA excluding Japan 1.8 1.3
FFIT 4.4 7.2
Industrial countries 2.9 2.9
Emerging market
countries
5.4 10.0
ITL 10.0 14.7
Sources: IFS, Bankscope.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1304
self-reported by central banks in the survey of Fry and others (2000).
4
Central banks
were asked to report their focus on the exchange rate and ination with respect to
their short and medium term policy. The strongest focus, which is termed
commitment in this paper, is 100. The authors of the survey derived the discretion
measure from the scores for exchange rate and ination targeting (Fry et al. (2000),
Table A1).
The survey results are indeed consistent with the denitions of the three ination
targeting regimes. The full-edged ination targeting countries report by far the
strongest commitment to an ination policy and the least discretion. The ination
targeting lite and implicit price stability anchor countries report the most discretion.
Finally, the ination targeting lite countries report a relatively stronger commitment
to ination vis-a` -vis the exchange rate suggesting that an ination objective is the
dening characteristic of their regime. In sum, all of these self-reported policy
indicators are consistent with the denition of ination targeting regimes and
classication of countries into these regimes in the previous section.
However, the strong commitment to ination of the FFIT countries does not
imply that they are, in the memorable words of Mervyn King, ination nutters
(King, 1997). While FFIT central banks are very clear about their commitment to
ination, they also care about stabilizing output, and, further, that in the short run
there is some benecially exploitable short-run tradeoff between real activity and
ination (Faust and Henderson, 2004). Indeed, FFIT countries operate with
ination outside the targeted range much of the time (Roger and Stone, 2005).
ARTICLE IN PRESS
Table 4
Ination targeting central banks, self-reported policy indicators
a
Commitment Discretion Commitment to
to ination exchange rate
Full-edged ination targeting (16 of 18)
Median 91.0 19.0 9.5
Average 79.8 24.6 21.9
Implicit price stability anchor (4 of 5)
Median 19.0 51.5 9.5
Average 26.8 57.8 18.8
Ination targeting lite (13 of 19)
Median 44.0 47.0 25.0
Average 46.9 51.3 23.8
Source: Fry et al. (2000).
a
Number of reporting countries in each regime in parentheses.
4
The survey was conducted in 1999, so there may be some discrepancies between the data and actual
objectives in March 2001. In addition, these data are not available for some of the countries.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1305
4. Structural underpinnings of the different ination targeting regimes
The varying monetary policy objectives and degrees of clarity raise the question
of what explains the revealed preference for different ination targeting regimes.
Insight into this question could be used to help inform the appropriate design
of monetary policy conditional on a countrys circumstances. This section
empirically analyzes the basic structural differences that underpin the three different
regimes.
4.1. Macroeconomic comparisons
The rst set of comparisons is based on simple comparisons of cross-country
macroeconomic data. The use of cross-country data over the same time period is
used to control for time-specic international events. The differences are examined
by comparing the within-regime median value of each indicator across three
regimes.
The IPSA countries have the highest levels of GDP and GDP per capita and ITL
countries have the lowest (Table 5). All of the IPSA countries are industrial, all of the
ITL countries are classied as emerging market, and the FFIT countries are a mix.
The median per capita GDP of the IPSA countries is larger than that of the FFIT
countries by a factor of 4
1
2
, and the median per capita GDP of the FFIT countries, is,
in turn, 4
1
2
times larger than that of the ITL countries. The ITL countries are also
generally smaller than the FFIT emerging market countries. These comparisons
indicate that the revealed preference for different ination targeting regimes can be
explained to some degree by the overall level of development.
The scal position, as measured by the government debt to GDP ratio, is strongest
for the FFIT countries. The IPSA countries have the relatively largest median scal
debt at 59% of GDP (50% excluding Japan). The FFIT countries have the lowest
debt ratio; perhaps because they need to credibly commit to an explicit ination
target. The ITL countries have an intermediate level of debt. Importantly, the
regimes match up with two gauges of restrictions on central bank nancing of the
government.
5
First, the ITL country central banks reported that they provide the
relatively most amount of nancing to the government (Table 5). By contrast, the
IPSA countries and both subgroups of FFIT countries provide little or no nancing
to their governments, as indicated by the value of 100. The second gauge is a more
direct ordinal measure of legal restrictions on central bank government nancing.
Central bank laws with most restrictive limits are assigned lower values. The
emerging market FFIT countries have the strongest limits, and the IPSA countries
have the weakest limits. The central bank nancing restrictions of industrial FFIT
countries and ITL countries are intermediate.
The IPSA countries have well-developed nancial systems, in marked contrast to
the ITL countries. IPSA countries exhibit much higher levels of monetization and
have much larger stock markets than FFIT countries, which, in turn, have deeper
ARTICLE IN PRESS
5
A recent measure of central bank independence is not available for the ination targeting countries.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1306
ARTICLE IN PRESS
T
a
b
l
e
5
M
e
d
i
a
n
v
a
l
u
e
s
o
f
i
n
d
i
c
a
t
o
r
s
a
c
r
o
s
s
i
n

a
t
i
o
n
t
a
r
g
e
t
i
n
g
r
e
g
i
m
e
s
a
P
e
r
c
a
p
i
t
a
G
D
P
U
S
$
G
o
v
t
.
b
a
l
.
G
o
v
t
.
C
e
n
.
b
a
n
k
L
e
g
a
l
r
e
s
t
r
i
c
t
i
o
n
G
D
P
$
U
S
b
i
l
l
i
o
n
s
t
o
G
D
P
d
e
b
t
/
G
D
P

n
a
n
c
i
n
g
c
e
n
.
b
a
n
k
2
0
0
0
2
0
0
0
1
9
9
6

2
0
0
0
1
9
9
9

2
0
0
0
t
o
g
o
v
t
.
g
o
v
t
.

n
.
b
M
a
c
r
o
e
c
o
n
o
m
i
c
i
n
d
i
c
a
t
o
r
s
I
m
p
l
i
c
i
t
p
r
i
c
e
s
t
a
b
i
l
i
t
y
a
n
c
h
o
r
(
5
)
3
3
,
5
8
9
4
7
5
3

0
.
6
3
5
9
.
1
1
0
0
.
0
c
1
.
5
b
F
u
l
l
-

e
d
g
e
d
i
n

a
t
i
o
n
t
a
r
g
e
t
s
(
1
8
)
7
6
9
5
1
4
2

0
.
9
8
3
7
.
5
1
0
0
.
0
0
.
8
I
n
d
u
s
t
r
i
a
l
(
7
)
2
3
,
8
4
4
2
2
7
0
.
6
9
3
7
.
3
1
0
0
.
0
1
.
1
E
m
e
r
g
i
n
g
m
a
r
k
e
t
(
1
1
)
4
7
2
8
1
2
2

3
.
2
7
3
7
.
5
1
0
0
.
0
0
.
5
I
n

a
t
i
o
n
t
a
r
g
e
t
i
n
g
l
i
t
e
(
1
9
)
1
7
2
0
1
9

2
.
8
3
5
0
.
4
6
2
.
5
0
.
9
B
r
o
a
d
m
o
n
e
y
S
t
o
c
k
m
k
t
.
R
e
s
e
r
v
e
m
o
n
e
y
R
e
a
l
i
n
t
.
P
r
v
.
e
x
t
.
d
e
b
t
P
o
r
t
.

o
w
s
t
o
G
D
P
c
a
p
i
t
a
l
t
o
G
D
P
t
o
b
r
o
a
d
m
o
n
e
y
r
a
t
e
t
o
G
D
P
t
o
G
D
P
1
9
9
8

2
0
0
0
1
9
9
8

1
9
9
9
1
9
9
9

2
0
0
0
1
9
9
8

1
9
9
9
1
9
9
8

1
9
9
9
1
9
9
5

1
9
9
9
F
i
n
a
n
c
i
a
l
i
n
d
i
c
a
t
o
r
s
I
m
p
l
i
c
i
t
p
r
i
c
e
s
t
a
b
i
l
i
t
y
a
n
c
h
o
r
(
5
)
1
1
5
.
4
1
6
8
.
2
1
0
.
4
c
5
.
1
c
N
A
N
A
F
u
l
l
-

e
d
g
e
d
i
n

a
t
i
o
n
t
a
r
g
e
t
s
(
1
8
)
5
6
.
2
4
7
.
8
1
1
.
2
8
.
1
N
A
N
A
I
n
d
u
s
t
r
i
a
l
(
7
)
5
3
.
8
9
7
.
1
7
.
1
6
.
0
N
A
N
A
E
m
e
r
g
i
n
g
m
a
r
k
e
t
(
1
1
)
5
8
.
6
3
1
.
8
2
0
.
2
9
.
9
2
0
.
2
0
.
5
I
n

a
t
i
o
n
t
a
r
g
e
t
i
n
g
l
i
t
e
(
1
9
)
4
1
.
6
1
0
.
5
3
2
1
2
.
4
3
.
7
0
.
1
S
o
u
r
c
e
s
:
I
F
S
,
W
o
r
l
d
E
c
o
n
o
m
i
c
O
u
t
l
o
o
k
a
n
d
W
o
r
l
d
B
a
n
k
G
l
o
b
a
l
E
c
o
n
o
m
i
c
I
n
d
i
c
a
t
o
r
s
d
a
t
a
b
a
s
e
s
f
o
r
m
a
c
r
o
e
c
o
n
o
m
i
c
a
n
d

n
a
n
c
i
a
l
i
n
d
i
c
a
t
o
r
s
;
F
a
u
s
t
a
n
d
H
e
n
d
e
r
s
o
n
(
2
0
0
4
)
f
o
r
c
e
n
t
r
a
l
b
a
n
k
i
n
d
i
c
a
t
o
r
s
;
M
i
s
h
k
i
n
a
n
d
S
c
h
m
i
d
t
-
H
e
b
b
e
l
(
2
0
0
1
)
a
n
d
I
M
F
C
e
n
t
r
a
l
B
a
n
k
L
e
g
i
s
l
a
t
i
o
n
d
a
t
a
b
a
s
e
f
o
r
c
e
n
t
r
a
l
b
a
n
k

n
a
n
c
i
n
g
o
f
g
o
v
e
r
n
m
e
n
t
.
a
N
u
m
b
e
r
o
f
c
o
u
n
t
r
i
e
s
w
i
t
h
a
v
a
i
l
a
b
l
e
d
a
t
a
i
n
p
a
r
e
n
t
h
e
s
e
s
.
b
O
r
d
i
n
a
l
i
n
d
i
c
a
t
o
r
w
i
t
h
c
e
n
t
r
a
l
b
a
n
k

n
a
n
c
i
n
g
:
n
o
t
a
l
l
o
w
e
d

0
,
l
i
m
i
t
e
d

1
,
a
n
d
n
o
l
i
m
i
t
s

2
;
a
v
e
r
a
g
e
s
a
c
r
o
s
s
i
n

a
t
i
o
n
t
a
r
g
e
t
i
n
g
r
e
g
i
m
e
s
a
r
e
r
e
p
o
r
t
e
d
.
c
E
x
c
l
u
d
e
s
E
M
U
.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1307
markets than the ITL countries. In a similar vein, real interest rates, an indicator of
the cost and risk of nancial transactions, are relatively high for the ITL countries
and low for the IPSA countries.
In sum, the descriptive comparisons suggest that the revealed preference for the
different ination targeting regimes reects sharp disparities in nancial develop-
ment and the overall level of development.
4.2. Econometric analysis
The structural underpinnings of the different ination targeting regimes are
examined next econometrically. The goal of the econometric analysis is to gain some
understanding of the differences in the economic structure and history across
countries that choose the three different ination targeting regimes.
Each ination targeting regime is assumed to generate some level of utility for the
government: V
IPSA
, V
ITL
, and V
FFIT
. The government chooses the regime which
gives the highest utility: y j if V
j
maxfV
IPSA
; V
ITL
; V
FFIT
g. Each utility is a
function of underlying characteristics and recent history. The estimator is
multinomial unordered probit. Since the dependent variable is unordered, one of
the three categories corresponding to the different ination targeting regimes has to
be used as a baseline against which the others are compared. The FFIT regime is
used as the baseline category so that V
FFIT

FFIT
; V
ITL
Xb
ITL

ITL
and
V
IPSA
Xb
IPSA

IIPSA
.
6
The regressions are based on cross-sectional data, rather than panel data, because
the focus here is on differences across countries rather than over time, and
because the time period over which ination targeting has been in place for most
countries is rather short. The candidate independent variables are those discussed
in the previous section. The ECB is excluded owing to the lack of comparable
macroeconomic data.
Simultaneity is a potential problem with a single equation model of ination
targeting regime choice. The problem is that not only would the explanatory
variables help explain regime choice, but regime choice could also help explain
explanatory variables such as ination performance and nancial depth. Simulta-
neity could cause biased parameter estimates.
A two-step procedure is used here to address simultaneity, as in Rivers and Vuong
(1988). The rst step involves breaking down the variance of the potentially
endogenous right-hand side variables of interest into exogenous and endogenous
components. The second step is estimation of a multinomial unordered probit for
regime choice.
Implementation of the rst step involves OLS regressions of the potentially
endogenous variables on instruments assumed to inuence regime choice only
through their impact via these variables. The potentially endogenous variables (per
ARTICLE IN PRESS
6
In addition, estimation of the model requires several restrictions on the covariance matrix for
EIT
;
ITL
;

FFIT
. In particular, with three regime choices, only two parameters of the covariance matrix can be free
in estimation. The restrictions chosen are s
EIT;ITL
s
EIT;FFIT
.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1308
capita GDP, ination, stock market capitalization to GDP, broad money to GDP
and stock market trading activity) are each regressed on three instruments (property
rights, law and order, and legal origin).
7
The instruments (most notably property
rights) explain a reasonable degree of the variation in the potentially endogenous
variables (Table 6). The unexplained dependent variable variance from these
regressions is presumed to be endogenous to regime choice and is thus used in the
second step regressions.
8
The econometric analysis is meant to gain some understanding of the structural
underpinnings of the revealed preference for the three different ination targeting
regimes. The regimes are differentiated by the clarity and credibility of the central
banks commitment to the ination target. Therefore, the explanatory variables
included in the second step regressions are hypothesized to explain the different
ARTICLE IN PRESS
Table 6
First-step OLS estimates
Per capita Ination Stock market Money to
GDP, Log capitalization GDP
to GDP
Constant 10.59 4.62 66.40 82.59
(19.29) (0.44) (1.91) (4.88)
Property rights 0.9538 8.921 31.18 17.47
(7.72) (3.80) (3.99) (4.60)
Law and order 0.1228 1.711 9.719 2.447
(1.50) (1.10) (1.88) (0.97)
Legal origin 0.1499 2.458 5.205 1.131
(1.88) (1.62) (1.03) (0.46)
No. of observations 41 41 41 41
R
2
0.652 0.357 0.453 0.438
F3; 37 23.1 6.86 10.2 9.61
7
The three instruments used in the rst step regressions are indicators of legal traditions that have been
found in other studies to explain nancial development and growth (e.g. Beck et al., 2001; Acemoglu and
Johnson, 2002). The index of law and order is from the International Country Risk Guide and ranges from
1 to 10 for strong law and order tradition. The property rights index is compiled by the Heritage
Foundation. The legal origin index compiled by La Porta et al. (1998) divides the legal origin of each
countrys company law or commercial code into: English common law, French commercial code, German
commercial code, Scandinavian commercial code, and socialist law.
8
Since the second stage regression includes the generated (residual) regressor its variancecovariance
matrix is adjusted accordingly.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1309
levels of ination targeting regime clarity and credibility. The priors for parameter
estimate signs are as follows:
Per capita GDP may serve as a proxy for structural rigidities (Walsh, 2002) and
central bank independence and is thus expected to enter the regression with a
negative sign for ITL countries vis-a` -vis FFIT countries and a positive sign for
IPSA countries vis-a` -vis FFIT countries. (Since FFIT countries are the baseline
category in the unordered probit regressions the estimated parameters associated
with IPSA and ITL countries are relative to those of FFIT countries.)
IPSA regimes are sophisticated frameworks of monetary policy, which require
ne-tuning of indirect instruments, which, in turn, require deep nancial markets.
Thus, the money to GDP and stock market capitalization to GDP parameter
estimates are expected to be positive for IPSA countries and negative for ITL
countries.
The recent history of ination will help shape the credibility of the commitment to
an ination target and is thus expected to enter with a positive sign for IPSA
countries and negative sign for ITL countries.
An absence of scal dominance is seen as an important precondition for FFIT
(Masson et al., 1997; Mishkin, 2000; Carare et al., 2002) and thus government debt
to GDP can be expected to be negative for ITL countries, while the expected sign
for IPSA countries is not clear.
Restrictions on central bank nancing of the government enhances credibility by
precluding inationary nancing of a government decit and thus enters with an
expected negative sign for the ITL countries and possible a positive sign for the
IPSA countries.
The second step results (Table 7) have a reasonable degree of explanatory
power, judging by the pseudo R
2
and the number of correctly predicted dependent
variable observations. The lack of signicant parameter estimates for the rst
step regressions indicates that biased estimate resulting from simultaneity is not a
problem.
Per capita GDP is a signicant and robust determinant of regime choice,
indicating that the overall level of development is a good indicator of regime choice.
Per capita GDP probably serves as a proxy for the structural ingredients underlying
the revealed preference for different levels of clarity and credibility. The stock market
capitalization parameter estimate is also signicant even after controlling for per
capita GDP. The signicance of the nancial depth indicators is attributable to its
association with nancial stability and the use of the sophisticated framework of
monetary policy needed for IPSA and FFIT.
Perhaps surprisingly, the ination performance in and of itself is not highly
correlated across ination targeting regimes. Nor are the two indicators of the
strength of the scal position signicant, another surprising result. The lack of
explanatory power of ination and the scal position may be because these
indicators are backward looking, and forward looking indicators are more
important, but difcult to introduce empirically.
ARTICLE IN PRESS
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1310
ARTICLE IN PRESS
T
a
b
l
e
7
I
n

a
t
i
o
n
t
a
r
g
e
t
i
n
g
r
e
g
i
m
e
,
m
u
l
t
i
n
o
m
i
a
l
u
n
o
r
d
e
r
e
d
p
r
o
b
i
t
r
e
g
r
e
s
s
i
o
n
r
e
s
u
l
t
s
V
a
r
i
a
b
l
e
/
r
e
g
r
e
s
s
i
o
n
1
2
3
4
5
I
T
L
I
P
S
A
I
T
L
I
P
S
A
I
T
L
I
P
S
A
I
T
L
I
P
S
A
I
T
L
I
P
S
A
C
o
n
s
t
a
n
t
a
1
2
.
3
9

1
0
0
1
0
.
4
7

1
0
0
7
.
8
5

1
0
0
1
1
.
1
9

1
0
0
9
.
5
2

1
0
0
(
1
.
8
5
)

(
2
.
7
6
)

(
2
.
4
3
)

(
2
.
0
3
)

(
2
.
4
8
)

L
o
g
G
D
P
p
e
r
c
a
p
i
t
a
,
1
9
9
0

1
9
9
9

1
.
1
6
8
.
9
8

1
.
2
9
9
.
0
2

0
.
9
8
1
1
.
3
5

1
.
0
3
4
.
1
9

0
.
9
0
5
.
4
0
(

1
.
5
2
)
(
7
.
5
5
)
(

2
.
6
5
)
(
1
8
.
0
6
)
(

2
.
7
1
)
(
7
.
1
1
)
(

1
.
7
8
)
(
0
.
0
2
)
(

6
.
3
3
)
(
0
.
6
0
)
L
o
g
G
D
P
p
e
r
c
a
p
i
t
a
,
1
9
9
0

1
9
9
9
r
e
s
i
d
u
a
l
s
0
.
1
8

0
.
8
4
0
.
7
0

9
.
5
4
(
0
.
2
3
)
(
0
.
1
5
)
(
1
.
3
6
)
(

1
.
7
9
)
I
n

a
t
i
o
n
,
1
9
9
6

2
0
0
0
0
.
0
1
1
1
.
0
0
0
.
0
5
6

9
.
9
6
0
.
0
0
5
1

1
1
.
7
0
(
0
.
4
5
)
(
0
.
9
0
)
(
1
.
4
9
)
(

1
.
0
2
)
(
0
.
2
1
)
(

0
.
0
4
)
I
n

a
t
i
o
n
,
1
9
9
6

2
0
0
0
,
r
e
s
i
d
u
a
l
s

0
.
0
3
8
0
.
9
8
(

1
.
0
0
)
(
1
.
2
9
)
B
r
o
a
d
m
o
n
e
y
/
G
D
P
,
1
9
9
8

2
0
0
0

0
.
0
0
4
0
0
.
0
5
6
(

0
.
2
6
)
(
1
.
3
7
)
S
t
o
c
k
m
a
r
k
e
t
c
a
p
.
/
G
D
P
,
1
9
9
6

1
9
9
9

0
.
0
5
6
0
.
0
7
9

0
.
0
5
6
0
.
6
7

0
.
0
5
9
0
.
2
1
(

1
.
8
4
)
(
0
.
8
7
)
(

1
.
8
6
)
(
0
.
0
3
)
(

2
.
4
1
)
(
1
.
2
6
)
S
t
o
c
k
m
a
r
k
e
t
c
a
p
.
/
G
D
P
,
1
9
9
6

1
9
9
9
,
r
e
s
i
d
u
a
l
s
0
.
0
3
8

0
.
0
4
9
0
.
0
3
8

0
.
5
2
0
.
0
3
6

0
.
0
1
1
(
1
.
3
0
)
(
0
.
4
8
)
(
1
.
4
0
)
(

0
.
0
3
)
(
1
.
3
7
)
(

1
.
5
0
)
G
o
v
e
r
n
m
e
n
t
d
e
b
t
/
G
D
P
,
1
9
9
9

2
0
0
0
0
.
0
0
5
1
0
.
2
5
0
.
0
1
7
0
.
3
1
(
0
.
4
3
)
(
1
.
6
4
)
(
1
.
0
3
)
(
0
.
9
9
)
R
e
s
t
r
i
c
t
i
o
n
s
o
n
c
e
n
t
r
a
l
g
o
v
t
.

n
a
n
c
i
n
g

1
.
1
7
5
.
5
4

1
.
3
2
5
7
.
5
8

1
.
1
6
4
2
.
9
1

1
.
1
3

0
.
0
7
0
(

1
.
1
9
)
(
0
.
8
0
)
(

1
.
5
6
)
(
1
.
0
7
)
(

1
.
2
0
)
0
.
0
1
(

1
.
2
0
)
(

0
.
0
1
)
N
u
m
b
e
r
o
f
o
b
s
e
r
v
a
t
i
o
n
s
4
1
4
1
4
1
4
1
4
1
L
o
g
-
l
i
k
e
l
i
h
o
o
d

1
2
.
1
2

1
3
.
8
5

1
4
.
1
8

9
.
4
4

8
.
9
2
P
s
e
u
d
o
-
R
2
0
.
6
8
7
2
0
.
6
4
2
4
0
.
6
3
3
9
0
.
7
5
6
4
0
.
7
6
9
8
C
o
r
r
e
c
t
p
r
e
d
i
c
t
i
o
n
s
(
p
e
r
c
e
n
t
)
b
E
I
T
(
4
)
7
5
.
0
%
1
0
0
.
0
%
1
0
0
.
0
%
1
0
0
.
0
%
1
0
0
.
0
%
I
T
L
(
1
9
)
8
9
.
5
%
7
8
.
9
%
9
4
.
7
%
9
4
.
7
%
8
9
.
5
%
F
F
I
T
(
1
8
)
8
3
.
3
%
7
7
.
8
%
8
3
.
3
%
8
8
.
9
%
8
8
.
9
%
a
D
u
e
t
o
c
o
m
p
l
e
t
e
d
e
t
e
r
m
i
n
a
t
i
o
n
o
f
t
h
e
E
I
T
r
e
s
p
o
n
s
e
,
w
e
b
o
u
n
d
t
h
e
c
o
n
s
t
a
n
t
f
r
o
m
b
e
l
o
w
a
t

1
0
0
,
s
o
t
h
a
t
t
h
e
m
a
x
i
m
u
m
l
i
k
e
l
i
h
o
o
d
e
s
t
i
m
a
t
e
s
f
o
r
t
h
e
c
o
n
s
t
a
n
t
a
n
d
t
h
e
s
l
o
p
e
c
o
e
f

c
i
e
n
t
o
n
l
o
g
p
e
r
c
a
p
i
t
a
G
D
P
w
o
u
l
d
n
o
t
g
o
t
o
i
n

n
i
t
y
.
W
e
i
m
p
l
e
m
e
n
t
t
h
e
c
o
n
s
t
r
a
i
n
t
t
o
p
r
e
s
e
r
v
e
c
o
m
p
a
r
a
b
i
l
i
t
y
a
c
r
o
s
s
s
p
e
c
i

c
a
t
i
o
n
s
,
p
r
e
v
e
n
t
i
n
g
m
a
g
n
i
t
u
d
e
s
o
f
d
i
f
f
e
r
e
n
c
e
a
r
i
s
i
n
g
a
c
r
o
s
s
s
p
e
c
i

c
a
t
i
o
n
s
.
F
o
r
t
h
e
s
a
m
e
p
u
r
p
o
s
e
t
h
e
s
t
a
n
d
a
r
d
e
r
r
o
r
s
o
f
t
h
e
c
o
n
s
t
a
n
t
s
h
a
v
e
b
e
e
n
o
m
i
t
t
e
d
.
b
E
s
t
i
m
a
t
e
d
p
r
o
b
a
b
i
l
i
t
y
g
r
e
a
t
e
r
t
h
a
n
5
0
%
.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1311
5. Regime switches
The existence of different ination targeting regimes raises the important policy
question of when and how a country should switch from one ination targeting
regime to another. Six switches between the three regimes are possible. However,
switches from IPSA and FFIT to ITL can be precluded from consideration since
countries will not choose to go from a higher credibility to a low credibility regime.
In addition, a low credibility ITL country cannot credibly switch to IPSA. This
leaves three possible regime switches.
No country has made the switch from IPSA to FFIT. However, the US, Japan and
the ECB have all been urged to adopt FFIT; see Krugman (1998), Bernanke et al.
(1999), Mishkin (1999), and Truman (2003) to depersonalize and institutionalize the
high level of credibility and get out of or avoid a disinationary spiral. The efcacy
of a switch from IPSA to FFIT depends on whether the long run gain in ination-
ghting credibility outweighs the loss in exibility to achieve other objectives (Stone,
2003b). Since the US and the ECB enjoy a large degree of credibility already than a
switch would need to be predicated on future developments that could impair
credibility. Such developments could include a marked weakening of the scal
position or the appointment of a new central bank head with less stature. Japan is a
special case owing to its deationary liquidity trap and it could be noted that the
adoption of FFIT will not in and of itself provide a means to get out of this trap.
A switch from FFIT to IPSA is not under consideration by any of the 18 FFIT
countries; although given the relatively short history of FFIT a switch to IPSA may
well happen in the future. A switch would be warranted if the ination-targeting
credibility of the central bank was so well entrenched that it could change to a new
regime where it could temporarily shift its objective away from price stabilitybut
without a shift in the mean or an increase in the variance of ination expectations.
Although FFIT has been adopted to reduce ination, a switch from FFIT to IPSA
should be assessed not only in terms of an increase in inationary expectations, but
also with respect to a decrease in ination, as evidenced by the emergence of
deation in Japan. In practice, a switch from FFIT to IPSA could be implemented
either by reducing transparency and accountability modalities or by keeping these
modalities and announcing multiple objectives. Either way, the ination target
would be much harder for the public and markets to monitor. The industrial FFIT
countries look to be the most likely candidates to move to IPSA as they have more
empirically credible monetary frameworks.
9
The ination targeting regime change of most immediate policy interest is the
switch from ITL to FFIT, which is relevant for emerging market countries only.
FFIT is fast gaining popularity with emerging market countries. From 1997 to 2001
the number of emerging market countries with FFIT rose from zero to eleven, during
ARTICLE IN PRESS
9
Finland and Spain had an FFIT regime prior to joining the EMU on January 1, 1999. Since their
decision to join the euro area means that their economic benets are higher than the costs of joining a
monetary union, and this decision is not related to the costsbenets analysis for changing a monetary
framework we do not document this switch further.
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1312
2002 Peru and the Philippines adopted this framework, and other emerging market
countries are considering the adoption of FFIT (cf. Banerji, 2003). In addition,
emerging market countries with some sort of xed exchange rate regime may view
ITL as a policy option. In fact, ITL can be viewed as a transitional monetary regime
during which the central banks aims to maintain nominal stability long enough for
the implementation of structural reforms in support of a single nominal anchor
(Stone, 2003a). Empirical analysis suggests that ITL countries should aim to boost
nancial system development and their scal position prior to a switch to FFIT
(Carare and Stone, 2003).
The consideration of ination targeting regimes suggests that a new ination
targeting regime halfway between IPSA and FFIT may be in the works. Today
several of the implicit price stability anchor countries are enhancing their
commitment to the ination target in ways that make them look more like full-
edged ination targeters in transparency if not accountability (Stone, 2003b;
Kramer and Stone, 2005). At the same time, the locking in of a high degree of
credibility by full-edged ination targeting countries may give them room to ease
their commitment to the ination target to give them more discretion to pursue
output and nancial stability. The creation and destruction of monetary regimes in
response to changing global and national circumstances, as exemplied by the
emergence of full-edged ination targeting during the 1990s, can be expected to
continue.
6. Conclusion
The message of this paper is that the monetary frameworks of oating exchange
rate countries are now separating into three markedly different ination targeting
monetary regimes: full-edged ination targeting, implicit price stability anchor and
ination targeting lite. The clarity and credibility of the commitment to the ination
target is qualitatively different across the regimes and thus these regimes have
important implications for how to design a monetary policy framework conditional
on a countrys underlying economic structure.
The revealed preference for different ination targeting regimes can be attributed
to different welfare maximizing combinations of policy objectives. The ultimate
purpose of monetary policy is to maximize social welfare by attaining high and stable
growth in the long run. Monetary policy can support long run growth through a
combination of ination in the low single digits, nancial stability, and output
stabilization. The welfare maximizing combination of these three objectives in the
policy framework depends on a countrys level of credibility. The level of credibility,
in turn, depends on the underlying economic structure of a country. Empirical
analysis suggested that the overall level of development, the level of nancial
development, and the severity of restrictions on central bank nancing of the
government are highest for IPSA countries and lowest for the ITL countries.
Finally, the classication of ination targeting countries into separate regimes
raises some potentially fruitful lines of policy application and research in addition to
ARTICLE IN PRESS
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1313
the issue of regime switches. How much scope for discretion, if any, do FFIT
countries lose with greater transparency? How are IPSA central banks held
accountable for their ination performance? What are the lessons from these regimes
for the design of central bank governance? Does the high degree of commitment of
the full-edged ination targeting countries limit the scope of constructive
ambiguity needed to deal with nancial instability?
Acknowledgements
We would like to thank Ashok Bhundia, Mariano Cortes, Haizhou Huang,
Shigeru Iwata, Tim Lane, Akiva Offenbacher, Ole Risager, Lucio Sarno, Andrea
Schaechter, Gabriel Sterne, Alun Thomas, Carl E. Walsh, Mark Zelmer, participants
at the IMF MAE and INS seminars and the Australasian Econometric Society
Meetings in Brisbane, the Bank of Japan, the Bank of England, the Federal Reserve
Bank, the European Central Bank, and the Reserve Bank of Australia. Shane
Sherlund provided invaluable help with the econometrics.
The views expressed in this paper are those of the authors and do not necessarily
represent those of the IMF or IMF policy.
References
Acemoglu, D., Johnson, S., Robinson, J., Thaicharoen, Y., 2002. Institutional causes, macroeconomic
symptoms: Volatility, crises and growth. Mimeo.
Banerji, A., 2003. The choice of a nominal anchor: Is Russia ready for ination targeting? In: Russian
Federation: Selected Issues. International Monetary Fund Country Report 03/146, May.
Beck, T., Demirguc-Kunt, A., Levine, R., 2001. Financial Structure and Economic Growth: A Cross-
Country Comparison of Banks, Markets and Development. MIT Press, Cambridge, MA.
Bernanke, B.S., Laubach, T., Mishkin, F.S., Posen, A.S., 1999. Ination Targeting: Lessons from
International Experience. Princeton University Press, Princeton, NJ.
Blejer, M.I., Ize, A., Leone, A.M., Werlang, S. (Eds.), 2001. Ination Targeting in Practice: Strategic and
Operational Issues and Application to Emerging Market Economies. International Monetary Fund,
Washington, DC.
Carare, A., Stone, M.R., 2003. Ination targeting regimes. IMF Working Paper 03/9, International
Monetary Fund, January 2003.
Carare, A., Schaechter, A., Stone, M.R., Zelmer, M.D., 2002. Establishing initial conditions in support of
ination targeting. IMF Working Paper 02/102, International Monetary Fund, June 2002.
Faust, J., Henderson, D.W., 2004. Is ination targeting best practice monetary policy? Federal Reserve
Board. International Finance Discussion Papers, 2004-807.
Fracasso, A., Genberg, H., Wyplosz, C., 2003. How do central banks write? An evaluation of ination
reports by ination targeting central banks, Geneva Reports on the World Economy Special Report 2,
International Center for Monetary and Banking Studies, Geneva, and CEPR.
Fry, M., Julius, D., Mahadeva, L., Roger, S., Sterne, G., 2000. Monetary Policy Frameworks in a Global
Context. Bank of England, Centre for Central Banking Studies.
Gerlach, S., 1999. Who targets ination explicitly? European Economic Review 43, 12571277.
King, M., 1997. Changes in UK monetary policy: Rules and discretion in practice. Journal of Monetary
Economics 39, 8197.
Kramer, C., Stone, M., 2005. A post-reation monetary framework for Japan. IMF Working Paper,
forthcoming.
ARTICLE IN PRESS
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1314
Krugman, P.R, 1998. Its back: Japans slump and the return of the liquidity trap. Brookings Papers on
Economic Activity (2), 137205.
La Porta, R., Lopez-de-Silanes, F., Schleifer, A., Vishny, R., 1998. Law and nance. Journal of Political
Economy 106 (6), 11131155.
Masson, P.R., Savastano, M.A., Sharma, S., 1997. The scope for ination targeting in developing
countries. IMF Working Paper 97/130, International Monetary Fund, September.
Mishkin, F.S., 1999. International experiences with different monetary policy regimes. Journal of
Monetary Economics 43 (3), 576606.
Mishkin, F.S., 2000. Ination targeting in emerging market countries. NBER Working Paper 7618,
National Bureau of Economic Research, March 2000.
Mishkin, F.S., Schmidt-Hebbel, K., 2001. One decade of ination targeting in the world: What do we
know and what do we need to know? NBER Working Paper Series 8397, National Bureau of
Economic Research, July 2001.
Rivers, D., Vuong, Q.H., 1988. Limited information estimators and exogeneity tests for simultaneous
probit models. Journal of Econometrics 39 (3), 347366.
Roger, S., Stone, M., 2005. On target? The international experience with ination targeting. IMF Working
Paper, forthcoming.
Sarel, M., 1996. Nonlinear effects of ination on economic growth. IMF Staff Paper 43 (March), 199215.
Schaechter, A., Stone, M.R., Zelmer, M.D., 2000. Adopting ination targeting: Practical issues for
emerging market countries. IMF Occasional Paper 202, International Monetary Fund, 2000.
Schmidt-Hebbel, K., Tapia, M., 2002. Monetary policy implementation and results in twenty ination-
targeting countries. Central Bank of Chile Working Paper 166, June 2000.
Stone, M.R., 2003a. Ination targeting lite. IMF Working Paper No. 03/12, International Monetary
Fund, January.
Stone, M.R., 2003b. Greater monetary transparency for the G3: Lessons from full-edged ination
targeters. IMF Working Paper No. 03/218, International Monetary Fund, November.
Truman, E.M., 2003. Ination Targeting and the International Financial System: Challenges and
Opportunities. Institute for International Economics, Washington DC.
Walsh, C.E., 2002. Economic Structure and Monetary Policy Design. Paper prepared for the July 2002
Conference of the East West Center, Korean Development Institute.
ARTICLE IN PRESS
A. Carare, M.R. Stone / European Economic Review 50 (2006) 12971315 1315