Ecological Economics 37 (2001) 39 51

www.elsevier.com/locate/ecolecon

ANALYSIS

Environmental valuation for long-term strategic planning

the case of the Brazilian power sector
Mauricio Tiomno Tolmasquim a,*, Ronaldo Seroa da Motta b,
Emilio Lebre La Rovere a, Martha Macedo de Lima Barata a,
Aline Guimaraes Monteiro a
a
Energy Planning Program, Institute for Research and Postgraduate Studies of Engineering (PPE/COPPE), Room C-211,
Technology Center, Federal Uni6ersity of Rio de Janeiro, C.P. 68565, 21945 -970 Rio de Janeiro, Brazil
b
Institute for Social and Economic Research (IPEA), Brazilian Ministry of Planning, A6. Presidente Antonio Carlos, n 51,
Rio de Janeiro, Brazil

Received 7 December 1999; received in revised form 12 September 2000; accepted 17 October 2000

Abstract

If not properly dealt with over compatible horizons, environmental issues may cause ventures to incur higher costs
and extend over longer implementation periods. Prompted by the uncertainties and risks inherent in these issues, as
well as by the need for long-term plans to incorporate sustainable development principles, the Brazilian Power Sector
is encouraging the inclusion of environmental aspects as a decision variable right from the start of the planning cycle.
The purpose of this paper is to bolster efforts to include in this sectors expansion plans, costs resulting from
environmental degradation of the environment. The paper offers an overview of the methodology developed to assess
and include in long-term planning for this sector, external environmental costs linked to hydro-power and
thermo-power generation. It suggests the use of environmental economic valuation techniques a practice
commonly used to analyze the degradation of natural capital in various countries as well as assessing projects and
programs, adopting the premises and simplifications required for application over the longer term. 2001 Elsevier
Science B.V. All rights reserved.

Keywords: Environmental costs; Monetary valuation for social and environmental external costs; Brazilian power generation;
Long-term planning process; Sustainability

1. Introduction
* Corresponding author. Tel.: +55-21-5608995; fax: + 55-
The Productive Sector plays an important role
21-2906626.
E-mail address: tolmasquim@ppe.ufrj.br (M.T. Tol- in the quest for sustainability. Eco-efficient pro-
masquim). duction is becoming a reality among enterprises in

0921-8009/01/$ - see front matter 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 8 0 0 9 ( 0 0 ) 0 0 2 6 2 - 7
40 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
the developed countries, whereby enterprises join
forces to maximize their combined economic and
environmental results, aimed at the adoption of
cleaner production processes and technologies,
while at the same time maintaining their aggregate
productivity. Nevertheless, it is still rare for major
enterprises particularly those in the Public
Power Sector in developing nations to engage
in these environmental assessments in the initial
phase of setting priorities in their action plans.
This was equally true in Brazil, until recently.
When specific projects and programs do include
environment variables through quantification
of environmental impacts it has been shown
by experience that they are added at a very ad-
vanced phase in the planning cycle, thereby failing
to ensure that the most sustainable projects are
implemented. Within this context, there is a
worldwide trend to expand the concepts and pro-
cedures involved in the Environmental Impact
Assessment of projects in order to assess their
policies, plans and programs, placing environmen-
tal issues at the same level as other parameters
economic, financial, technological, market de-
mands, etc. Keenly aware of this requirement and
in step with worldwide trends, the Environmental
Division of Brazilian Power Utility ELETRO-
BRAS, is striving to include social and environ-
mental variables in its strategic long-term plans,
thereby guaranteeing the inclusion of the princi-
ples of sustainable development when planning its
expansion activities.
The Brazilian Power Sector undertakes its plan-
ning through a series of studies based on long-
term analyses covered by the Brazilian National
Electric Power Plan (NEPP). This plan analyzes
prospects for development of the electric power
market over a 30-year horizon, as well as the
availability of primary power generation sources,
and trends in technological developments. Based
on strategies defined by NEPP, a medium-term
plan is drawn up with a 15-year horizon, which is
then reviewed on an annual basis during the
preparation of the 10-year expansion plans for the
sector. In this ongoing structured planning pro-
cess, expansion strategies are systematically re-as-
sessed and the alternatives reviewed until a
decision is reached on venture implementation.
Within this cycle, decisions are based on energy
production cost/benefit analysis, while environ-
mental issues are only included when the actions
related to a specific project are considered. An
exception to this rule is the planning cycle for
hydro-power plants, where a methodology has
been developed and adopted for considering envi-
ronmental impacts in the initial stage of the plan-
ning process, based on inventory studies of
hydrographic basins (CEPEL/PPE/COPPE/
UFRJ, 1997), working towards maximizing en-
ergy and economic efficiency while minimizing
environmental impacts.
As a matter of fact, with the exception of
degradation costs, all environmental costs in-
curred through the implementation of the venture
are generally already included in the energy cost/
benefit ratings by the sector during its long-term
planning. The study undertaken by the Power
Sector Coordination Committee for Environmen-
tal Activities (COMASE) proposes a social and
environmental cost concept tailored to the charac-
teristics of the Brazilian Power Sector, with tools
and mechanisms to include these costs in the
budget. In turn, this helps to include environmen-
tal variables in the long-term planning for this
sector. The characteristics of the environmental
costs defined for this sector are presented below
(ELETROBRAS/COMASE, 1994):
Control costs are costs incurred by the Power
Sector to avoid the total or partial occurrence
of social and environmental impacts of a
venture.
Mitigation costs are costs incurred by the
Power Sector through actions to reduce the
consequences of social and environmental im-
pacts caused by a venture.
Compensation costs are costs incurred by the
Power Sector through actions which offset so-
cial and environmental impacts caused by a
venture in situations where reparation is
impossible.
Monitoring costs are costs incurred by the
Power Sector through monitoring and evaluat-
ing social and environmental programs.
Institutional costs are costs incurred by the
Power Sector in drawing up the social and
environmental studies covering the planning,
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
implementation and operations stages; prepar-
ing the studies required by the environmental
agencies, and; obtaining environmental
licenses.
Degradation costs are external costs caused by
residual social and environmental impacts,
even after the Power Sector has incurred con-
trol, mitigation and/or compensation costs.
They correspond to social and environmental
external costs, in other words the negative re-
sults caused by power generation that are not
defrayed by power generation utilities but are
rather charged to third parties, due to the lack
of any voluntary action.
In view of the difficulties intrinsic to the nature
of degradation costs, these have not yet been
included in the Accounts Plan of the Brazilian
Power Sector. However, their identification and
valuation would allow the monetary value of
these external costs to be included in the energy
cost/benefit rates (US$/MWh). These procedures
would avoid the risk that energy and economic
criteria are given more importance than environ-
mental criteria when analyzing alternatives for
ventures in the long-term power plan during the
selection process. Environmental ratings may not
be sufficient for planners to opt for alternatives
that maximize economic and energy efficiency
while also minimizing environmental impacts.
Within this context, the inclusion of a monetary
value assigned to external factors in the energy
cost/benefit ratings (US$/MWh) could be an ef-
fective procedure. This could provide the feasibil-
ity for ventures with a reduced impact on the
environment, and/or assigning them higher prior-
ity, either of which would encourage the selection
of ventures maximizing energy and economic
criteria while also minimizing environmental
impacts.
This paper presents an overview of the method-
ological proposals and the respective assumptions
and simplifications presented to the Brazilian
Power Sector, in order for its long-term planning1
1
The Brazilian Power Sector is currently undergoing a
restructuring process (Rosa et al., 1998). In the new sector
model, the long-term plan will lose importance. In spite of
that, it will continue to play an important role as a guide to
regulators bidding at an auction.
41
to include the monetary value of the potential
degradation costs deriving from thermo- and hy-
dro-power generation ventures. For this, we pro-
posed the adoption of the existing environmental
economic valuation techniques (see Freeman,
1979, 1993; Dixon et al., 1994; May, 1999). These
techniques form part of the theoretical framework
of the welfare microeconomics, and are necessary
to determine the social costs and benefits when
public and private investment decisions affect the
well-being of society. They are usually applied
when analyzing the degradation of natural capi-
tal, as well as for assessing projects and programs,
when the ventures to be implemented are known,
as well as their impacts and external factors.
The difficulty in using these monetary valuation
techniques for long-term planning lies in the fact
that at this stage the enterprise characteristics are
not yet known. Additionally, the large number of
potential projects makes it difficult to evaluate the
monetary effect of the environmental impact on
them. The solution adopted to face this problem
was to elaborate a typology of possible projects
thereby limiting the universe of analysis options.
Without a doubt, by adopting this procedure, we
are in danger of over-simplifying the reality. For
the moment, however, this is the most objective
framework there is to evaluate alternative ven-
tures. It is better to work to improve existing and
operational cost benefit frameworks until viable
alternatives are developed (Bucher and Huszar,
1996).
2. A methodology for valuation of degradation
costs in the long-term planning of the Brazilian
Power Sector
In order to assess the value of degradation costs
in long-term plans deriving from thermo-power
and hydro-power generation activities, the follow-
ing stages were followed (Tolmasquim et al.,
1999):
1. Systematic characterization of the impact
caused by each thermo-power and hydro-
power generation activity.
2. Identification of economic effects deriving
from the environmental impacts characterized
42 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951

Table 1
Typology of attributes for the valuation of environmental impacts of thermo-power plants

Technology Fuel Area of influence of pollutant Population density

Conventional steam cycle Brazilian coal

Fluidized bed
Conventional steam cycle with crushed coal boiler Imported coal
Fluidized bed/combined cycle
Conventional Natural gas Good/bad dispersal High/medium
Combined cycle
Gas turbine Standard fuel oil
Conventional Heavy fuel oil

in stage (1), usually not controlled, mitigated
or compensated by the Power Sector.
3. Definition of the categories of thermo-power
and hydro-power generation activities appro-
priate for identifying and incorporating the
degradation cost on long-term planning of the
Power Sector.
4. Selection of the degradation costs for valua-
tion and inclusion in the long-term planning of
the Power Sector, of the categories defined in
stage (3), complying with the following
criteria:
 relevance for the sustainable economic de-
velopment of the country;
 difficulty in eliminating environmental ex-
ternalities through the implementation of
control, compensation or mitigation mea-
sures; and,
 feasibility of the inclusion of valuation in
long-term planning at non-prohibitive costs;
5. Suggest application of the most appropriate
valuation techniques to obtain the degradation
cost selected.
2.1. Definition of 6enture categories appropriate
to the 6aluation of degradation costs in the
long-term planning of the Brazilian Power Sector
In long-term planning, some important vari-
ables for the use of valuation techniques are not
known, implying the need to establish assump-
tions and typologies that allow ventures to be
classified according to variables that will be sig-
nificant for the assessment of degradation costs,
and which are common to the venture classifica-
tion categories.
The proposed typology is based on some ven-
ture attributes which are of vital importance for
the valuation process, as described in Tables 1
and 2. It is noted that for some degradation costs,
not all attributes will be taken under
consideration.
Based on the definition of this typology, cate-
gories were defined for thermo-power and hydro-
power plants, with the main goal to permit the
calculation of average values of the damages
caused by these ventures, which will then be in-

Table 2
Typology of attributes for the valuation of environmental impacts of hydro-power plants

Location Size of flooded surface Population density

Amazon region Virgin lands

Frontier expansion areas Large/small Medium/low
Heavily settled areas
Non-Amazon region
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
corporated into the energy cost/benefit ratings of
ventures covered by the expansion plans for the
Power Sector.
2.2. Proposed techniques for the 6aluation of
degradation costs in the long-term planning of the
Brazilian Thermo-Power Sector
The effects on human health caused by atmo-
spheric discharges of sulfur oxide and particulate
material and the degradation costs of global
warming caused by carbon dioxide emissions were
selected externalities to be valued and included in
the long-term planning of the Power Sector, at-
tending to that established in stage 4 of the
methodology presented.
2.2.1. Damage caused to human health by
atmospheric emissions of sulfur dioxide (SO2) and
particulate material (PM10)
In order to evaluate the damage caused to
human health by thermo-power generation for
long-term planning purposes, the sector could ap-
ply dose/response functions to estimate the poten-
tial cases of morbidity and mortality induced by
the increase of pollutant concentration levels and
then evaluate the monetary value of the associ-
ated impacts.
During the inventory of the power potential of
Brazil, it is possible to list the alternative tech-
nologies and fuels that could be used, but the
exact location of the power plants is not known,
meaning that it is not possible to identify with
accuracy the manner in which the pollutants will
be dispersed and affect human health. Therefore,
in order to evaluate and incorporate degradation
costs in long-term planning caused by the atmo-
spheric emissions of thermo-power plants (EPA,
1997), 30 venture categories2 have been estab-
lished, based on the combination of fuels and
technologies in Brazilian power generation plants,
with the expected demographic density in the
2
As it is not expected that a thermo-power plant fired by
Brazilian coal will be implemented on conventional technology
in a densely populated area, this category has not been consid-
ered.
43
probable area of influence of each power source.
(see Table 1).
To estimate the variations in the concentration
of pollutants coming from the future thermo-
power plants for each thermo-power venture (see
Rosa and Schechtman, 1996), we assume that the
concentration of a pollutant (Cs,in ) in the area of
influence depends on the pollutant emissions
(Es,in ), the average wind speed (u) and the dis-
tance from the emission point of the pollutant(s)
as shown:
Cs,in = (106 Q Es,in )/(2y s u) (1)
where Cs,in corresponds to the annual variation in
the concentration of the pollutant (i ) at point (s)
of the region for each thermo-power venture cate-
gory (n) (mg/m3 per year); Q corresponds to a
conversion coefficient to estimate the concentra-
tion at expected unity; Es,in corresponds to the
annual emissions of the atmospheric pollutant (i )
at point (s) of the region for each thermo-power
venture category (n) (mg of pollutant per year); s
corresponds to the distance from the emission
point of the pollutant (m); and u corresponds to
the average wind speed (m/s).
The effect of the pollutants emitted on human
health is cumulative, which suggests that the in-
crease in the annual concentration of the pollu-
tants should be taken into consideration,
calculated by the numerical integration of the
immediate concentrations of the pollutants
obtained.
The impact on the health of the human popu-
lace (DSs,in ) living in the area of influence of the
thermo-power plant caused by the variations in
the concentration of the atmospheric pollutant (i )
is determined by:
DSs,in = Pops,i % Cs,in bij (2)
s = 1;k
where DSs,in corresponds to deaths or diseases
caused by the variation in the concentration of
the pollutant (i ) on the population living in the
area under influence (s) of a thermo-power cate-
gory (n) (number of deaths or diseases); Pops,i
corresponds to the population exposed to the risk
of the pollutant (i ) in (s) (number of people); Cs,in
44 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
corresponds to the annual variation in the concen-
tration of the pollutant (i ) at point (s) of the
region for each thermo-power venture category
(n) (mg/m3 per year); bij corresponds to the dose/
response coefficient for a variation in the concen-
tration of the pollutant (i ) at point (s) in the
region (response/(mg/m3)); and k corresponds
to the maximum distance traced by the pollutants
from the point of emission, where k =21 km for
low dispersal areas and k =150 km for high dis-
persal areas.
Based on studies of the cases researched, and
contact with specialists in atmospheric dispersion,
the average path (k) is defined for both cases,
indicating the planned area of influence over the
long term.
The dose/response coefficient (bij ) varies ac-
cording to the population surveyed (see Seroa da
Motta et al., 1999). There is no evidence to sug-
gest that the studies of one populace could be
adapted to other cases, as the social, economic
and physical characteristics specific to local com-
munities could influence the results obtained.
2.2.1.1. Monetary 6aluation of morbidity. An ini-
tial estimate of the morbidity costs can be pro-
vided by healthcare expenditures (hospital costs,
medication, etc.) and the loss in productivity
caused by diseases.
It is not possible to identify expenditures on
medical consultations and medications deriving
from diseases caused by atmospheric pollution, as
there are no statistics on this. It is thus suggested
a survey be commissioned in order to obtain
average figures for each meteorologically homoge-
neous region, or for each geographical region.
2.2.1.2. Monetary 6aluation of mortality risk. In
order to evaluate the value of human life, it is
important to distinguish between the statistical
life value and the life value of a specific individ-
ual. It is noted that assigning value to human life
is a controversial issue, for ethical reasons, but
failure to do so is equivalent to assigning nil value
to human life. In order to estimate the statistical
life value, it is suggested that measures be used
which have already been widely estimated for
other countries, particularly those estimated on
the basis of contingent valuation of mortality risk.
These could be adjusted for the social and eco-
nomic conditions in Brazil, as already suggested
in the literature (see Pearce, 1998; Markandya,
1998), using the benefit transfer method.
It is concluded that the degradation cost for
each of the 30 venture categories proposed here
takes into consideration aspects related to the
dispersal of pollutants emitted and the demo-
graphic density of the area where the plant is
located. This corresponds to the sum of monetary
value of morbidity and mortality risk assigned to
each of the venture categories for a set of two
pollutants: SO2 and PM10, as shown:
(VDSHs,n,p )
= VMn*%vu(DSsSO2,n(deaths) + DSsPM,n(deaths))
+ VMBn*%vu(DSsSO2,n(diseases) + DSsPM,n (diseases))
(3)
where VDSHs,n,p corresponds to the total value of
the damage to human health during the working
life of a venture (n) in an atmospheric dispersal
area (s) with a demographic density (p) (on a
monetary basis); VMn corresponds to the statisti-
cal value assigned to life (on a monetary basis);
VMBn corresponds to the value assigned to mor-
bidity (on a monetary basis); DSsi,n(diseases) corre-
sponds to the diseases caused by the variation in
the concentration of pollutant (i ) on the popula-
tion living in the area (s) of influence of a thermo-
power plant category (n) (on a monetary basis);
DSsi,n(deaths): corresponds to the deaths caused by
the variation in the concentration of the pollutant
(i ) on the population living in the area (s) of
influence of a thermo-power plant category (n)
(on a monetary basis).
It is stressed that the degradation costs of
thermo-power plants should be evaluated on the
basis of the average working life of each of the
venture categories proposed here, which is why
the formula gives (vu).
2.2.1.3. Main constraints. Over the long term, not
all variables used to evaluate damage to human
health over the short term are known, whereby
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951 45

some simplifications are adopted and data aver- plant is obtained through:
ages are used. Within this context, it was decided
VDTn = VDSHs,n,p + VDCO2n (5)
to work with 30 categories for Brazil, as this is a
vast country with a wide variety of marked physi-
2.2.2.1. Main constraints. Valuation of the dam-
cal, social and economic differences. This means
ages caused by global warming is a highly contro-
that the results obtained would be more accurate
versial issue. Some worldwide problems, such as
if the categories were established by meteorologi-
global warming, can produce non-marginal alter-
cally homogeneous regions (ideally, but they are
ations in terms of human well-being which are
too numerous).
hard to capture through valuation methodologies.
For each of the regions defined, fresh studies
Despite this, various attempts to evaluate the
should be undertaken to:
damages caused by global warming have been
1. define the new determining radii (k) of the
undertaken in order to help establish targets for
area of influence of power-plants;
reducing the emission of gases causing this phe-
2. define the coefficients of the dose/response
nomenon. This makes the valuation of carbon
function;
emissions by thermo-power plants in Brazil an
3. define the expenditures incurred for medica-
important planning tool for the Power Sector.
tion and hospitalization due to diseases caused
Most monetary estimates of the total damage
by atmospheric emissions.
caused by 1 ton of carbon dioxide emitted today
4. provide specific estimates for the statistical
and accumulated in the atmosphere over time
values based on contingent valuation of mor-
come from the USA, with widely varying results
tality risk.
(UNEP, 1996). However, there is a relatively nar-
row range for calculations of the damage in terms
2.2.2. Global warming caused by carbon dioxide
of the USA and the developing countries, which
(CO2) emissions from thermo-power plants
could lead to a convergence between them. In
The valuation and incorporation in the long-
view of this convergence, it is possible to extrapo-
term planning of the damage due to global warm-
late the USA values to other parts of the world
ing caused by CO2 emissions from thermo-power
with some approximations.
plants could be calculated through the determina-
tion of the product of total emission of pollutant
2.3. Proposed techniques for the 6aluation of
during the venture working life by the monetary
degradation costs in the long-term planning of the
value assigned to the emission of 1 ton of carbon
Brazilian Hydro-Power Sector
dioxide. Thus:
Among the environmental externalities deriving
VDCO2n =(VCO2)(%6uECO2n ) (4)
from hydro-power generation, the degradation
cost of biodiversity3 will be evaluated and in-
where VDCO2n corresponds to the monetary value
cluded in the long-term planning of the Brazilian
of the damage caused by global warming
Power Sector. This is justified by the biological
prompted by carbon dioxide emissions by each
diversity of Brazil and the economic value that
venture category (n) (monetary units); VCO2 corre-
this represents. Additionally, Amazonia still has
sponds to the monetary value of the total damage
much virgin land with a wealth of biodiversity to
of 1 ton of carbon dioxide emitted today and
be discovered, and is viewed as an energy resource
accumulated in the atmosphere over time (mone-
store-house with vast potential, particularly in
tary units per ton of carbon dioxide emitted); and
terms of hydro-power.
ECO2n corresponds to emissions of carbon dioxide
for each venture category during its working life
(tons of carbon dioxide).
The total value of the environmental degrada- 3
The choice of this externality attends to the established
tion (VDTn ) for each category of thermo-power criteria on stage 4, number 2, of the methodology presented.
46 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
The classification of the ventures (see Table 2)
into two types of bio-regions is justified by the
environmental importance assigned to the Amazon
bio-region. Due to the vast extent of this bio-region
and the presence of areas that have not undergone
deforestation, it was decided to divide it into three
areas:
virgin lands, which are areas that have not
been affected by mankind but not protected as
nature reserves (it is important to stress that no
hydro-power plants may be built in nature
reserves);
frontier expansion areas, which are areas of
transition between heavily settled areas and
virgin lands located along the western border
of heavily settled areas;
heavily settled areas are areas that have been
cleared or flooded, used for farming, grazing or
forestry, as well as for the construction of
hydro-power plants, located in Eastern
Amazonia.
The bio-region not in Amazonia contains a
wider diversity of ecosystems, most of which have
been affected by the countrys development. Thus,
despite the diversity of ecosystems, it was decided
to cluster them under a single category. Bio-re-
gions not in Amazonia cover the following ecosys-
tems: Caatinga drylands, Cerrado savannas,
Pantanal wetlands, Atlantic rainforest, Meadows,
and Coastal zone.
The size of the reservoirs will influence the
flooded areas and consequently the physical
amount of species degradation, whereby two cate-
gories of flood areas are taken into consideration
for each bio-region: large flooded areas (over 250
km2) and small flooded areas (less than 250 km2).
The definition of the categories of fauna and
flora population to be affected by the venture and
its proximity to settled areas will be important at
the time of identifying the biodiversity to be
valued.
The procedures suggested for evaluating the
biodiversity affected by each of the 10 venture
categories established is presented here, for inclu-
sion in the long-term planning.
In order to evaluate the biodiversity to be af-
fected by the hydro-power generation ventures,
consideration will be given to:
existence value;
value for direct and indirect use, including
timber products and other extractive assets
(rubber, fibers, nuts, fruits, hunting and
fishing), and
value of the carbon stock function of the
ecosystem.
Listed below are the suggested procedures for
estimating each of the values listed above.
2.3.1. Application of the contingent 6aluation
methodology to estimate the damage caused to
plant and animal species by hydro-power plants
The contingent valuation method (CVM) is used
for estimating the value assigned by the inhabitants
of a country to the preservation of an ecosystem,
as the only method that can capture the value of
existence.
The value of existence of biodiversity for an
individual reflects the value assigned to biodiver-
sity by that individual even if he does not and will
not directly use animals and plants as resources. It
is important to check endemic and non-endemic
species that will be lost. The appropriate way of
handling this type of valuation is to use the CVM
for species affected by each venture. As the exact
site of the dam is not known in the long-term
planning, it was decided to establish the venture
categories for each typology of attributes presented
in Table 2.
In order to characterize the hydro-power ven-
tures for each category, specific questionnaires
should be prepared with scenarios that reflect the
average species loss probabilities. These will be
drawn up on the basis of a survey identifying the
species in each bio-region. It is noted that under
Brazilian law the power sector must implement
mitigatory and compensatory measures offsetting
the environmental impacts deriving from new hy-
dro-power ventures, which should be taken into
consideration when preparing this scenario.
The probability of losses of plant and animal
species (biodiversity) is estimated for each category
(n) of hydro-power venture. Thus, hn will be the
scenario covered by the survey through which the
interviewees will disclose their willingness to pay.
For each magnitude of hn, we will thus have an
average willing-to-pay figure. This means that the
CVM studies will be undertaken for estimating hn.
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
The economic 6alue of the en6ironmental re-
source (on a monetary basis) is estimated on the
basis of the preferences expressed by individuals
based on a demand function disclosing the will-
ingness-to-pay of the interviewees as a function of
the availability of the environmental resource
tagged for preservation.
The willingness-to-pay value (VCn ), being
sought should reflect the probability of the defini-
tive loss of biodiversity (hn ) for each venture
category. This affirmation can be expressed
through:
VCn =f (hn )
Thus, the economic value that an individual is
willing to pay to preserve a group of plant and
animal species found in the categories of the
Amazon bio-region or outside the Amazon bio-re-
gion should reflect the probability of the loss of
these species. This loss may be more or less
marked depending on the size of the flooded area
and the number of species found. In the case of
the Amazon bio-region, this biodiversity loss
varies according to the approximate size of the
area to be flooded and the specific features of the
three different areas: virgin lands, heavily settled
areas, and frontier expansion areas.
For the bio-regions outside Amazonia, the
probability of loss varies as a function of the
approximate size of the area to be flooded and its
proximity to settled areas.
For categories that do not fall into the areas
outside the Amazon bio-region, with an interme-
diate demographic density (more than 100 inhabi-
tants/km2 and less than 1000 inhabitants/km2), as
well as those which are classified as heavily settled
and exploited in the Amazon bio-region, the
probability of definitive loss of species through
building the reservoirs is practically nil in relation
to virgin land with a large number of plant and
animal species. Thus, the economic value assigned
to these species will be nil.
2.3.1.1. Main constraints. The difficulties involved
in applying the CVM to a hypothetical market are
4
See for instance, the classic reference of Mitchell and
Carson (1993).
47
widely discussed in the literature.4 With the
methodology proposed here, these difficulties
would be even greater, due to the information
needed about the biodiversity potential. More-
over, the identification of species subject to loss
would have to be carried out on the basis of
certain ecological parameters that are still very
inaccurate, particularly in the Amazon region.
2.3.2. Damage caused to timber and non-timber
extracti6e products
The valuation of timber and non-timber forest
resources is concentrated on measuring the value
(6)
of direct individual use, associated with the sus-
tainable exploitation of tropical rainforests, which
over the long term corresponds to the income that
will be taken from the users of these resources for
producing timber and non-timber extractive prod-
ucts, due to the flooding of the area covered by
the reservoirs.
The methodology for evaluating this long-term
damage is similar to that used for short-term,
although the exact location of the reservoir is not
known. Thus, the average revenue to be obtained
by area and for each bio-region is inventoried and
then applied to the area to be flooded by each
venture. In order to measure the value of non-tim-
ber and timber resources in the long-term valua-
tion, it is necessary to undertake an inventory.
The latter consists of gathering qualitative and
quantitative data on the estimated species in the
area to be flooded, on the existing resources in the
bio-region to be analyzed Amazon bio-region
(untouched forests, frontier expansion, heavily oc-
cupied) and non-Amazon bio-region establish-
ing a representative sample, checking the annual
output rate for all fruit-bearing trees and palms in
the sample area, checking the current market
value of all species, and establishing their average
economic value for each bio-region.
Thus, the economic 6alue of en6ironmental re-
sources deriving from timber and non-timber
product extraction activities constitutes the total
revenues generated through the sustainable ex-
ploitation of products in each venture category,
with sustainable exploitation taken as that equal-
ing the natural growth rate of the resources
exploited.
48 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
The total net revenue (monetary units per area)
is the result of the total amount of products
harvested in the affected area, multiplied by the
difference between the average price and the gath-
ering, transportation and labor costs.
As these revenues are associated with a sustain-
able production flow, this extraction can thus be
undertaken infinitely, and we must thus calculate
the net present value of the revenues produced
each year on the basis of a discount rate
(d). Seroa da Motta and Ferraz (1999) have
undertaken a discussion of the value of (d) in
Brazil.
Thus, the economic value of the environmental
resource (VEn ) per area (A) for each category of
hydro-power venture (n) is given in:
VEn /A =((PP CT)(DQ/A))/d (7)
where Pp : corresponds to the price of the product
(P) which is taken as constant (monetary units
per quantity of product); CT corresponds to the
gathering, transportation, and labor costs, etc.
(monetary units per quantity of product); DQ/A
corresponds to the stock of products (P) obtained
from the variation in the quantity of the natural
resource (quantity of product per area in km2) in
the affected area; and d corresponds to the social
discount rate (%).
It is important to stress that the environmental
degradation cost for each venture category (n)
corresponds to the product of the economic value
of the environmental resource (VEn /A) in each
area by the surface (A) that will be flooded by the
reservoir of the future venture.
2.3.2.1. Main constraints. The market for some
extractive forest products hampers their sale, par-
ticularly perishable products such as fruits, nuts,
game and fish. No formal market is established,
making it difficult to obtain a market price series
for these products. Even the prices for products
that are widely sold may not reflect their shortage
revenue as they are worked on an open-access
system. For a discussion of this problem with
regard to logging in Amazonia, see Seroa da
Motta and Ferraz (1999).
2.3.3. Damage caused to biodi6ersity
ecosystem functions: carbon uptake
The value of this service providing support
and protection for economic activities based on
biodiversity comes from virgin lands, or those
that are still almost in their natural state.
One of the important ecosystem functions of
tropical rainforests in terms of the total economic
value of this environmental resource is its eco-
nomic value as a carbon stock (or even as a sink,
as some scientists believe), which should be
estimated.
One way of evaluating the amount of carbon
released by forest conversion through economic
activities is to determine the difference between
the value of the carbon stored in virgin forests
and the amount of biomass lost as a result of each
type of land use. In the case of total removal of
plant cover to constitute the reservoir for use
without any fixed carbon in the final product, the
carbon dioxide released is equivalent to the
amount of the original carbon stored in the forest
surface cleared. In the study of the estimated costs
per ton of carbon (US$/tC) stored in forests, the
availability and use of the area should be taken
into consideration, as well as the amount of car-
bon per hectare (according to the woody biomass
types and densities per hectare in the cleared
surface) and the land maintenance costs.
Once the net quantities of CO2 released by
deforestation and the introduction of farming,
grazing and forestry or other activities are known,
a cost coefficient can be established in terms of
monetary units per area, due to this contribution
to the increase of the greenhouse effect. To do so,
it is necessary to research and establish the envi-
ronmental cost caused by the emission of 1 ton of
CO2 into the atmosphere (monetary units per ton
of carbon). This cost must be considered and
decreases the economic return (net revenue on all
operating and capital costs) of the activities to
which the forest will be converted.
This simplified approach can be used as a first
cut of the average environmental cost of the re-
lease of carbon into the atmosphere by the flood-
ing of forests to fill hydropower plants reservoirs.
The VSCn (monetary units) per area (A) due to
the replacement of forests by reservoirs in each
venture category (n) is expressed as follows:
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
VSCn /A= VSCn (DC/A) (8)
where VSCn corresponds to the environmental
cost caused by the emission of 1 ton of carbon into
the atmosphere (monetary units per ton of car-
bon); and DC/A corresponds to the amount of
carbon released through forest conversion in an
economic activity for a specific area (A) in each
bio-region (tons of carbon per area in km2).
The value of the damage caused by the loss of
the ecosystem function due to the reduction in
carbon stocks caused by the filling of the reservoir
corresponds to the product of the surface to be
flooded by the value of the damage per area,
obtained for each bio-region.
2.3.3.1. Main constraints. As a matter of fact, the
building of large hydro-power plants reservoirs in
areas covered with tropical rainforests (generally
difficult to be cleared before filling the reservoirs
due to their remote access) entails a much more
complex process leading to greenhouse gas emis-
sions. A fraction of the carbon stored in the forest
decays in the anaerobic conditions beneath the
water, leading to the emission of methane and
carbon dioxide that climb to the surface and are
released to the atmosphere. This phenomenon
magnifies the contribution of the hydro-power to
the greenhouse effect, as the global warming po-
tential of one ton of methane is more than 20
times the warming caused by 1 ton of carbon
dioxide, in the long term. Moreover, the variation
in the surface of the reservoir between the dry and
the wet seasons throughout the year leads to the
annual flooding of the regenerated biomass grow-
ing in the margins of the reservoir. Both phenom-
ena are particularly intense in the Amazon region,
leading to additional emissions. Finally, how
much of the emissions caused by the reservoir are
net emissions compared to what would happen in
the absence of the project, remains a very contro-
versial issue.
Recent research work in this field performed by
PPE/COPPE/UFRJ (Rosa and Santos, 1999) has
identified a number of factors influencing the
magnitude of emissions from hydro-power plants.
Among others, one can mention:
the type and density per hectare of the biomass
49
flooded;
the average depth of the reservoir;
the age of the dam (changing pattern of emis-
sions over time).
So far, the results obtained from several mea-
surement campaigns in a number of Brazilian
hydro-power plants do not allow for an accurate
modeling of the emissions according to the above
influencing factors. In the future, this may be
possible. It should be stressed, however, that some
factors are project-specific (e.g. the depth of the
reservoir) and will not be available in the long-
term planning of the power sector. Other factors
depend upon the area where the projects are to be
located (such as the biomass density). If the latter
can be used as a proxy to determine the magni-
tude of the emissions, then it will be feasible to
develop a more accurate approach for valuation
of this impact in the long-term planning of the
power sector.
3. Conclusion
This work is the first step towards the internal-
ization of external environmental factors caused
by power generation projects in the long-term
planning of the Brazilian Power Sector.
This analysis is intended to provide inputs for
the inclusion of environmental degradation costs
(as defined by ELETROBRAS/COMASE, 1994)
in the long-term planning of the Brazilian Power
Sector. The non-inclusion of these costs may result
in incorrect allocation of resources. It may also be
included by the power sector in countries where
they used to do long-term planning with sufficient
data for playing the analysis cost-effectiveness and
by sectors with similar technical and environmen-
tal features.
Therefore, the valuation of externalities caused
by the future ventures in the strategic planning of
the Power Sector will contribute to achieving sus-
tainability. This will especially be the case if it is
integrated with other techniques like geographic
information systems (GIS).5
5
More details about cost-benefit analysis and GIS integra-
tion management decisions can be viewed in Walpole and
Sinden (1997).
50 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
As noted, the attempt to introduce environmen-
tal variables into this process requires the adop-
tion of methodological simplifications and the
streamlining of impacts due to the generalized
framework of long-term planning. Within this
context, and despite the importance of assigning
monetary values to environmental externalities in
order to ensure that environmental variables are
properly dealt with during the planning stage and
the decision process in this sector, some criticisms
may be addressed to the application of the meth-
ods proposed here for the long term, including:
In general, environmental valuation method-
ologies are applied to specific cases in order to
identify the cost/benefit of externalities caused
by anthropogenic activities, when and where
the main variables affecting the results are al-
ready known. Here, the adaptations of the
methodological approach required for its appli-
cation in the long-term planning add to the
uncertainties related to assigning values to the
externalities in question. A list of obstacles to
monetary valuation is presented in Tol-
masquim (1995).
In long-term planning, there is the concept of
the variables that will influence externalities,
but the exact scope of their impact and magni-
tude is not known. It is thus necessary to make
some estimates which, added to the uncertainty
inherent in planning, could make the results of
environmental planning seem somewhat un-
trustworthy to skeptics. It is important to stress
that, despite these uncertainties, these valua-
tions provide inputs that help shape decisions
which ensure that environmental variables are
not considered as nil.
Given the uncertainties involved in both ven-
ture location and specific impacts, in which
both the venture benefit cost framework and
the site-specific nature of the physical and so-
cial effects are not exactly known, the adoption
of these techniques in the strategic planing
context of the power sector must be followed
by a sensitive analysis of the probability of
external costs exceeding the net economic
benefits of the venture. This is presented by
Bucher and Huszar (1996) in their response to
David Barkin (1996) a benefit cost analysis
should include both internal and external
benefits costs of the project, as in the mean-
time this is the most objective framework to
evaluate the alternative ventures In the case
of the power sector it is usually part of the plan
which yielded macro-economic benefits, and
the analyst should consider this.
These criticisms do not invalidate the applica-
tion of methods aimed at linking the assignment
of a value for environmental resources to the
well-being of those affected by these ventures, for
inclusion in factors for decision analysis. The
criticisms merely highlight that the results ob-
tained should be taken into consideration with all
due precaution. In the long-term planning of the
Brazilian Power Sector, we observe that inserting
the suggested methodology rather than ignor-
ing or postponing the degradation cost pro-
vides a hierarchical indication of the sustainable
options of energy generation to attend to poten-
tial demand, thereby assisting the decision pro-
cess. It is important to note that with the
sustainability indicators and taking the re-
quired precautions in this type of analysis the
project cost benefit analysis allows projects to
line up in a hierarchy of sustainability.
Despite the methodological limitations, it is
also worthwhile noting that other non-environ-
mental economic variables taken into account
during the inventory phase engineering costs,
for instance are also subject to the same type
of constraints, which are intrinsic to long-term
planning.
Furthermore, it is understood that the applica-
tion of the methods proposed for each of the
typologies presented will allow the Brazilian
Power Sector to fine-tune them over time. This
being the case, the generalizations and simplifica-
tion of value estimations through the proposed
methodology will provide an innovative and pio-
neering procedure for including environmental
variables in the Brazilian Power Sector Planning.
Although useful, it is not the only tool to be used
for the inclusion of environmental variables in
long-term planning; it does represent, however, an
innovative advance in this field.
 M.T. Tolmasquim et al. / Ecological Economics 37 (2001) 3951
Acknowledgements
The authors gratefully acknowledge the finan-
cial assistance of Centrais Eletricas Brasileiras
S.A. ELETROBRAS. The authors gratefully
acknowledge the contributions received from the
atmospheric dispersion specialist Dr. Luiz Fran-
cisco Pires Guimaraes Maia.
References
Barkin, D., 1996. Macro changes and micro analysis: method-
ological issues in ecological economics. Ecol. Econ. 19,
197 200.
Bucher, E.H., Huszar, P.C., 1996. Project evaluation and
economic development: on using benefitcost analysis to
evaluate Hidrovia. Ecol. Econ. 19, 201203.
CEPEL/PPE/COPPE/UFRJ, 1997. Incorporacao da
Metodologia para Analise Integrada de Impactos de Usi-
nas Hidreletricas ao Manual de Inventario Hidreletrico de
Bacias Hidrograficas. Workshop: A Dimensao Ambiental
nos Estudos de Inventario Hidreletrico de Bacias Hidro-
graficas, Rio de Janeiro.
Dixon, J.A., et al., 1994. Economic Analysis of Environmental
Impacts. Earthscan Publications Ltd, London (100 pp).
ELETROBRAS/COMASE, 1994. Referencial para
Orcamentacao dos Programas Socio-ambientais.
EPA, 1997. Office of Compliance Sector Notebook Project,
Profile of the Fossil Fuel Eletric Power Generation Indus-
try. Washington, DC.
Freeman, A.M., 1979. The Benefits of Environmental Im-
provements. Theory and Practice. The John Hopkins Uni-
versity Press, Baltimore.
Freeman, A.M., 1993. The Measurement of Environmental
and Resource Values: Theory and Methods, Resources for
the Future. Washington, DC.
Markandya, A., 1998. The valuation of health impacts in
developing countries. Planejamento e Polticas Publicas 18,
119 154.
May, P.H., 1999. Natural Resource Valuation and Policy in
Brazil: Methods and Cases. Columbia University Press,
New York.
51
Mitchell, R.C., Carson, R.T., 1993. Using Surveys to Value
Public Goods: the Contingent Valuation Method. Wash-
ington, DC.
Pearce, D., 1998. Valuing statistical life. Planejamento e Polt-
icas Publicas 18, 69 118.
Rosa, L.P., Schechtman, R., 1996. Avaliacao de Custos Ambi-
entais da Geracao Termeletrica: Insercao de Variaveis Am-
bientais no Planejamento da Expansao do Setor Eletrico,
Cadernos de Energia, 9. In: Edicao Especial, vol. vol. 2.
(Brazil, 90 pp).
Rosa, L.P., Tolmasquim, M.T., Pires, J.C., 1998. In: Dumara,
R. (Ed.), A Reforma do Setor Eletrico no Brasil e no
Mundo-uma Visao Crtica. (Brazil).
Rosa, L.P., Santos, M.A., 1999. Hydroeletric dams and green-
house gas emission: an introduction to the problem. In:
Rosa, L.P., Santos, M.A. (Eds.), Dams and Climate
Change. Proceedings of International Workshop on Hydro
Dams, Lakes and Greenhouse Gas Emissions, pp. 7 15
(Rio de Janeiro, Brazil).
Seroa da Motta, R., Ferraz, C., 1999. Estimating timber
depreciation in the Brazilian Amazon, environment and
development economics, Brazil, in press.
Seroa da Motta, R.; Ortiz, R.A., Freitas, S.F., 1999. Avaliacao
Economica dos Impactos Causados pela Poluicao Atmos-
ferica na Saude Humana: Um Estudo de Caso para Sao
Paulo. III Encontro da Sociedade Brasileira de Economia
Ecologica, ECO-ECO, Recife, 11 13 November.
Tolmasquim, M.T., 1995. Economia do Meio Ambiente:
Forcas e Fraquezas. In: Cavalcanti, C. (Ed.), Desenvolvi-
mento e Natureza: Estudos para uma Sociedade Susten-
tavel. Editora Cortez, Brazil, pp. 323 345.
Tolmasquim, M.T., La Rovere, E.L., Seroa da Motta, R.,
Monteiro, A.G., Barata, M.M.L., 1999. Planejamento do
Setor Eletrico e a Valoracao das Externalidades Ambien-
tais Provenientes da Geracao Hidreletrica e Termeletrica.
Rio de Janeiro, Brazil, in press.
UNEP, 1996. Climate change 1995 economic and social
dimensions of climate change. Contribution of Working
Group III to the Second Assessment Report of the Inter-
governmental Panel on Climate Change, J.J. Houghton,
L.G. Meiro Filho (Eds.).
Walpole, S.C., Sinden, J.A., 1997. BCA and GIS: integration
of economic and environmental indicators to aid manage-
ment decisions. Ecol. Econ. 23, 45 57.