ARTICLE IN PRESS

Energy Policy 34 (2006) 26–39

www.elsevier.com/locate/enpol

An economic assessment of the Kyoto Protocol application

A.S. Dagoumas, G.K. Papagiannis, P.S. Dokopoulos
Power Systems Laboratory, Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece
Available online 6 July 2004

Abstract

Scope of this paper is to investigate scenarios concerning the economic implications of the Kyoto Protocol (The Kyoto Protocol
to the Convention on Climate change, http://unfccc.int/resource/docs/convkp/kpeng.pdf) in its current version including the clean
development mechanisms (CDMs), the Marrakesh Accords (Seventh session of the Conference of Parties, 29 October–9 November,
Marrakesh, Morocco, http://unfccc.int/cop7/index.html) and the Conference of Parties 9 in Milan (Nineth session of the Conference
of Parties, 1–12 December, Milan, http://unfccc.int/cop9/index.html). The general equilibrium model, GTAP-E was used for the
investigation of the experiments which are focused on the cases of the USA participation and on the role of Russia as a major
emission credits seller. A significant issue in the Kyoto Protocol negotiations is the introduction of sinks in the Marrakech Accords.
This seems to weaken the initial targets by replacing CO2 emissions reduction with forestation activities and reduces the cost of the
Protocol compliance. It is also shown that the absence of the USA may reduce the costs for the other developed countries and may
influence the total costs more than the CDMs. A new scenario is studied by introducing a guaranteed minimum of 60% in the
emission credits sold by Russia. Results show that the profits of Russia are not significantly affected by the guaranteed minimum.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: GTAP-E; Climate change; Kyoto Protocol; Marrakech accords; COP 9; MAC; Emission trading

1. Introduction The signing of the Kyoto Protocol (KP) in December

Climate changes may affect human life and the
ecology of the planet in a variety of ways. Most
countries have taken steps to address climate change
but more has been achieved, so far, in raising awareness
than in bringing about reductions in emissions: globally
energy related CO2 emissions have increased nearly 10%
from 1990 to 1999. Overall growth in emissions from
developing countries has been particularly significant,
although their per capita emission levels remain low.
The analysis of the economic implications of the
climate change mitigation policies has been ongoing for
over a decade by numerous research groups. Several
significant issues, but sometimes with contradicting
results, have been considered like localities, globaliza-
tion, economy, technology etc.
Corresponding author. Tel.: +30-2301-996356; fax: +30-2310-
996321.
E-mail address: dagoumas@auth.gr (A.S. Dagoumas).
1997 is perhaps an historic step in reversing the inexorable
increase in the emission of the greenhouse gases (GHG).
The primary achievement of the Protocol is the commit-
ment of countries referred in the Annex I of the KP to
reduce their emission of GHGs some 5% below their
country specific 1990 level, in the period 2008–2012 with
penalization clauses in case of non compliance.
The Protocol establishes three market based mechan-
isms for achieving emission reductions.
International emissions trading allows for a trade of
emissions among Annex I Parties or international. A
party with emissions lower than its target could sell the
remaining part up to the target (article 17 of the
Protocol).
Joint implementation allows for any Annex I Party
to be credited for emissions reduction achieved by
investing in projects located in other Annex I countries
(article 6 of the Protocol).
Clean development mechanism allows for any Annex I
Party to be credited for emissions reduction achieved by

0301-4215/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.enpol.2004.05.012
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39
investing in projects located in developing countries
under specific conditions (article 12 of the Protocol).
The initial vision of Protocol by the developed
countries was shadowed by the intention of the USA
not to ratify. In the Marrakesh 2001 COP a lot of effort
was invested and some important steps towards the
application of the Protocol were succeeded.
The important issue in Marrakech agreement was the
introduction of CO2 sinks. The key actor in these
negotiations was Russia. With the Marrakech Accords
the following activities can be counted as sinks (IPCC,
2000): Afforestation, reforestation, deforestation, forest
and agricultural management, revegetation and con-
servation activities. The carbon credits from sink-related
activities are about 120 MtC per year, which effectively
weaken the originally stated targets. The new targets
seem to be more feasible.
The cost of the Protocol implementation in each
region (USA, EU, etc.) depends on many factors which
were discussed in COPs. The essential among them seem
to be following:
 The participation of the USA and Russia.
 The CO2 sinks.
 The trading mechanisms and the related trade
restrictions.
 The boycott movements.
A number of important studies (Bahn et al., 2001;
Bernard et al., 2002; Bohringer, 2002; Christopher et al.,
1999; Gusbin et al., 1999; Ellerman and Decaux, 1998;
Hamasaki, 2002; Painuly, 2001; Viquier et al., 2003; den
Elzen and de Moor, 2002; McKibbin and Wilcoxen,
2002; Kuik and Gerlagh, 2003; Criqui et al., 1999;
Paltsev, 2001; Weyant and Hill, 1999; Springer, 2003;
Zhang, 2001: Chen, 2002) have assessed the cost of the
Kyoto Protocol at different times reflecting the con-
sequences of the time relevant COPs decisions. They
were mostly focused in the influence of several factors
but there has not been an analysis of the cost of the
Kyoto Protocol considering all above factors simulta-
neously.
Scope of this paper is to investigate the role of Russia
on the cost of the Kyoto Protocol implementation
taking into account the trading mechanisms and related
restrictions, the participation or not of USA, the
introduction of sinks and of a boycott movement.
2. Modeling
2.1. Available models
There are several ways to analyze the effects of the
climate policy on economic systems. Each model focuses
on certain areas and has its own structure of the
27
economic system. But the models can be classified into
four major categories.
 Computable general equilibrium (CGE) models: These
models are also called ‘‘top-down’’ models and are
either static or dynamic. The main characteristic of
these models is their ability to capture the influence of
energy policy on international trade and the
economy and the fact that the construction of the
economic system is based on the assumption of
perfect markets. Representatives of this category are
GTAP-E, GREEN, EPPA, GEM-E3, G-CUBED,
SGM, GTEM, MERGE, MS-MRT, PACE, WAGE,
WORLDSCAN and NEWAGE models.
 Energy system models: These models represent the
energy sector in much more detail than the CGE
models. They are referred as bottom-up models
because they represent the energy sector by using a
disaggregated data of existing and emerging technol-
ogies. Their main disadvantage is the fact that they
represent only the energy system and do not take into
account linkages of the energy sector with the rest
economy. Typical in this category are MARKAL,
POLES, PRIMES, LEAP and ENPEP models.
 Integrated assessment models: These models consider
human activities, atmospheric composition, climate
change and ecosystems and can be described as
environmental models. They are very useful when
conducting studies that address the problem of
climate change because of the detailed representation
of this change mechanism. Their economic structure
belongs to one of the above categories. Main
representatives of this category are the AIM,
GRAPE, IGSM and RICE models.
 Emission trading models: Models like CICERO, ECN,
MACGEM, ZHANG and ENEA belong in this
category. These models use MAC curves to analyze
International Emission Trading. MAC curves are
usually produced by energy system models, by CGE
models or are estimated econometrically.
In this work the GTAP-E model (Burniaux and
Truong, 2002) has been selected because it simulates the
economy sectors and the energy systems in more details
than the others. GTAP-E is an extension of the basic
GTAP model where E stands for energy. Generally,
GTAP is a widely used, static, multisector, multiregion
applied general equilibrium model (Thomas, 1997). It is
based on a detailed database with a broad coverage of
trade distortions and explicit statistics on transport
margins. Import demand is modeled using the Arming-
ton (1969) assumption of imperfect substitutability
between domestic and imported goods and between
imported goods from different regions. The model
assumes a global bank to mediate between world savings
and investments, and a region-specific set of equations
 ARTICLE IN PRESS
28 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39

for consumer demand that allows for different responses
to price and income changes across regions.
One of its most notable features is its detailed
database. For this study the GTAP 5 database has been
used consisting of 66 regions and 57 sectors (Dimaranan
and McDougall, 2002; Rutherford and Babiker, 1998)
GTAP-E has the same structure as GTAP, but its
production structure includes a more detailed descrip-
tion of substitution possibilities among different sources
of energy.
2.2. Functional description of the GTAP-E model
GTAP-E model in order to incorporate energy
substitution uses the so called bottom-up approach.
The advantage of this approach is in the detailed
specification of the energy technologies, through which
newly developed or future technologies can be incorpo-
rated into the analysis.
2.2.1. Production structure
Energy substitution is incorporated in the structure of
the GTAP-E model both in the production and the
consumption structure. As far as production is con-
cerned, energy is incorporated in the value added nest
(Fig. 1). This is done in two steps. Firstly, there is a
separation of energy commodities into ‘electricity’ and
‘non-electricity’ groups. Some degree of substitution
(Armington, 1969) is allowed within the non-electricity
group (sNELY) as well as between the electricity and
the non-electricity groups (sENER). On the second step
energy composite is combined with capital to produce
energy-capital composite. Finally the combination of
energy-capital composite with other primary factors
produces a ‘value-added-energy’ nest (VAE) through a
constant elasticity structure (CES) as shown in Fig. 2.
It is a zoomed presentation of the item in Fig. 1.
2.2.2. Consumption structure
On the consumption side, the existing structure of
GTAP assumes a separation of ‘private’ consumption
from ‘government’ consumption (consumption by
households of publicly provided goods) and private
savings. Government consumption expenditure is then
assumed to be Cobb–Douglas with respect to all
commodities. In the GTAP-E model, energy commod-
ities are separated from the non-energy commodities
with a nested-CES structure (Fig. 3).
Household ‘private’ consumption (i.e. consumption
of private goods) is assumed to be structured according
to the constant-difference of elasticities (CDE) func-
tional form in the existing GTAP model. If the energy
commodities within the CDE structure have the same
income and substitution parameters, then according to
the theory of the CDE structure, these commodities can
be aggregated into a single composite with the same
parameters as that of the individual components. To
allow for flexible substitution between the individual
energy commodities, the energy composite is now
specified as a CES sub-structure (Fig. 4).
The values of the elasticities have been selected, based
on detailed experiments and taking into consideration
the values of other models such as GREEN (Burniaux et
al., 1992), EPPA (Babiker et al., 2001) and AIM (AIM
(Asian-Pacific Integrated Model)). A stochastic sensitiv-
ity analysis has been carried out so as to evaluate the
significance of every substitution elasticity. For this
reason a lot of experiments have been implemented
based on current policies and the model has been
adjusted so as to justify the targets of policies that have
taken place from 1990 until 2000. The MURE database
MURE Mesures d’Utilisation Rationelle de l’Ènergie
has been also used as a database of measures that have
been implemented in the EU. The values of elasticities
that have succeeded results with great consistency with

Output
σ =0

All other inputs

Value-added- Energy
(Excluding energy inputs but
(including energy inputs)
Including energy feedstock)
VAE σD

Natural Capital-Energy Domestic Foreign

Land Labor Composite
Resource σM
LAB

Region 1 … Region r
Skilled Unskilled

Fig. 1. Production structure in the GTAP-E model used.

 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39 29

Capital-Energy Composite
σ KE

Capital Energy Composite

σ ENER =1.0

Non Electric Electric

σD
σ NELY=0.5

Coal Non Coal Domestic Foreign

σD σ NCOL=1 σM

Region 1 … Region r
Gas Oil Petroleum
Domestic Foreign
… … products
σM
σD

Domestic Foreign
Region 1 … Region r σM

Region 1 … Region r

Fig. 2. Capital-energy composite structure in the GTAP-E model used.

Demand for composite

σ GENNE=0.5

Energy Composite Non energy composite

σ GEN=1 σ GNE=1

Coal … … … … …
σD

Domestic Foreign
σM

Region 1 … Region r

Fig. 3. Structure of government purchases in the GTAP-E model used.

the reality, have been selected for the structure of the
model that was used in this paper.
2.3. Aggregations of regions and sectors in GTAP
GTAP model can be adjusted to the needs of every
study. One important factor is the aggregation of the
database so as to focus on certain regions and sectors.
For the purposes of the present analysis the database
of the model was aggregated in 8 regions and 8 sectors.
The selection of these regions and sectors has been done
based on the regions that are included in the different
scenarios and on the sectors that are used in order
to incorporate energy structure in the model. The
model divides the world into 8 geopolitical regions.
The model region and sector aggregation is shown in
Tables 1 and 2, respectively.
3. Reference case, assumptions
In the reference case, commonly called also the
Business As Usual (BAU) case, we assume that Kyoto
Protocol does not apply. The production of CO2 as a
 ARTICLE IN PRESS
30 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39

Household demand for

Private Goods
CDE

Energy Composite Non energy commodities

σ GEN=1 σD

Coal … … Domestic Foreign

σD σM

Domestic Foreign Region 1 … Region r

σM

Region 1 … Region r

Fig. 4. Structure of household private purchases in the GTAP-E model used.

Table 1
Model region aggregation

EU (European Union) consists of: EEx (Net Energy Exporters) consists of: RoW (Rest of the World) consists of:
AUT Austria IDN Indonesia HKG Hong Kong
BEL Belgium MYS Malaysia KOR Korea, Republic
DNK Denmark VNM Viet Nam TWN Taiwan
FIN Finnland MEX Mexico PHL Philippines
FRA France COL Colombia SGP Singapore
DEU Germany VEN Venezuela THA Thailand
GBR United Kingdom XAP Rest of Andean BGD Bangladesh
GRC Greece Pact LKA Sri Lanka
IRL Ireland ARG Argentina XSA Rest of South
ITA Italy XME Rest of Middle Asia
LUX Luxemburg East XCM Central America and Caribbean
NLD Netherlands XNF Rest of North PER Peru
PRT Portugal Africa BRA Brazil
ESP Spain XSF Rest of Southern CHL Chile
SWE Sweden Africa URY Uruguay
XSS Rest of Sub-Saharan Africa XSM Rest of South America
XRW Rest of the world TUR Turkey
CEE(Central and Eastern Europe) consists of: RoAI (Other Annex I countries) consists of: MAR Morocco
HUN Hungary AUS Australia BWA Botswana
POL Poland NZL New Zealand XSC Rest of SACU
XCE Rest of Central CAN Canada MWI Malawi
European Assoc CHE Switzerland MOZ Mozambique
XEF Rest of EFTA TZA Tanzania, United Republic of
FSU (Former Soviet Union) consists of: CHIND (China and India) consists of: ZMB Zambia
XSU Former Soviet Union ZWE Zimbabwe
CHN China UGA Uganda
IND India USA (United States of America

function of time has been calculated for different regions
and sectors for the period 1990–2010. Emissions growth
in the different regions is due to the changes in the GDP,
population and the carbon intensity of the output. The
average growth in GDP in the Annex I countries in the
baseline scenario is assumed to be 2.3% per annum over
the 20 year period 1990–2010. For the non Annex I
countries, the GDP growth rate is assumed to be 4.4%
per annum over the same 20 year period. Baseline
scenario was based on official reports (EIA (Energy
Information Agency and International Energy Out-
look), 2003; IEA (International Energy Agency); IMF
(International Monetary Fund), 2002; OECD (Organi-
sation for Economic Co-operation and Development)),
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39 31

Table 2
Model sector aggregation

Agriculture: Primary agriculture, forestry and fishing

Paddy rice, wheat, cereal grains nec, vegetables, fruits, nuts, oil seeds, sugar cane, sugar beet, plant based fibers, crops nec, bovine cattle, sheep and
goats, animal products nec, raw milk, wool, silk worm cocoons, forestry, fishing
Coal: Coal mining
Oil: Crude oil
Gas: Natural gas extraction
Gas, gas manufacture, distribution
Oil_pcts: Refined oil products
Petroleum, coal products
Electricity: Electricity
En_Int_ind: Energy intensive industries
Minerals nec, chemical, rubber, plastic prod, mineral products nec, ferrous metals, metals nec
Oth_ind_ser: Other industries and services
Bovine cattle, sheep and goat, meat products, vegetable oils and fats, dairy products, processed rice, sugar, food products nec, beverages and tobacco
products, textiles, wearing apparel, leather products, wood products, paper products, publishing, metal products, motor vehicles and parts, transport
equipment nec, electronic equipment, machinery and equipment nec, manufactures nec, water, construction, trade, transport nec, water transport, air
transport, communication, financial services nec, insurance, business services nec, recreational and other service, public administration and defense,
education, ownership of dwellings

Table 3
CO2 emissions increase for the period 1990–2010
Percentage annually
USA 1.38
EU 1.05
JPN 0.94
RoAI 1.9
EEFSU 0.74
FSU 1.18
CEE 0.45
on national communications (UNFCCC (United Na-
tions Framework Convention on Climate Change)) and
on the results of a neural network that has been
programmed for this reason.
Global anthropogenic emissions of carbon dioxide are
projected to grow by approximately 64.2% between
1990 and 2010. This is equivalent to an increase of
around 14.2 billion tons of carbon dioxide. Emissions
from developing countries are projected to increase
more rapidly than emissions from Annex I countries
over the projection period in the reference case.
Emissions from Annex I countries are projected to rise
by an average of 1.2% per annum over the period
1990–2010, while the corresponding annual increase for
the developing countries is 4.4%. Accordingly, the
Annex I share of world emissions is projected to fall
from 66% in 1990 to 51% in 2010. Table 3 shows the
annual CO2 emission increase for the studied regions.
signing of the Kyoto Protocol. Results obtained were
calculated taking into consideration two basic assump-
tions. Firstly, the Kyoto Protocol targets for the
development of CO2 emissions for the period
1990–2010 and secondly the CO2 sinks as defined in
the Marrakesh Accords in 2001. The development of
CO2 emissions is in Fig. 5 where it can be seen what
happens without and with the application of the Kyoto
Protocol and the required reduction in CO2 production
for this period. The results are region specific and are in
physical values, not in percent. USA and Russia
participate in this case. Fig. 6 shows for each region
the available sinks for the period 1990–2010 as result of
the Marrakesh Accords.
Fig. 7 shows the emission reductions in percentage
that have to be done to satisfy the targets of the Kyoto
Protocol as modified by the introduction of sinks, i.e.
Fig. 7 results out of combination of Figs. 5 and 6. The
cases represented are split into two subscenarios. One
shows the CO2 emission result in compliance with the
Kyoto Protocol in 1997 using as a base value the CO2
emission for the year 1990 and the second subscenario
shows the same quantities as in the first subscenario but
related to the reference case which is dynamic.
Key figure of this paper is the marginal abatement cost
and the total cost in absolute values and as a percent of
GDP. The first one reflects the incremental cost to reduce
the carbon by one unit in the year 2010 and the second
one provides a macroeconomic view of the consequences.

4.1. Evolution of the Kyoto Protocol with the

4. Experiments—results participation of the USA

Several scenarios beyond the reference case have been The cases that have been studied cover all possible
investigated covering all the phases from the initial actions that the participating countries may take. Kyoto
 ARTICLE IN PRESS
32 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39

Fig. 5. Development of CO2 emissions for the period 1990–2010, without and with measures according to the Kyoto Protocol in its initial form.

Fig. 6. CO2 Sinks according to the Marrakesh Accords for the period 1990–2010.

Protocol in its initial phase included domestic efforts
as the major action for the parties to satisfy the targets.
In Fig. 8 it is shown that the marginal abatement
cost is higher for Japan and the rest of the developed
countries. It is zero for Russia due to its emissions
surplus and very small for the CEE countries
which must invest on the reduction of low emission
levels.
In Fig. 9 the total cost for the USA is the highest, due
mainly to the amount of emissions that has to be
reduced as shown in Fig. 5. Expressed as percent of
GDP, this cost remains high although in Fig. 10 the
highest percentage of almost 0.5% of GDP appears for
the Rest of Annex I (RoAI) countries. There are two
main reasons for this. First, even if the absolute Kyoto
target for these countries is low, it represents a high
percentage of their total emissions because of their high
emissions growth as represented in the reference
scenario. The second reason is that these countries show
a relative low GDP.
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39 33

Fig. 7. Emissions reduction following the Marrakesh Accords.

Fig. 8. Marginal abatement cost for 2010 with the inclusion of USA.

The total cost for the implementation of the Protocol, introduced allowing the acquisition of emission credits
based only on domestic efforts, is high for the developed from other countries having a surplus of emission
countries. Therefore several trading mechanisms were credits. Emissions trading has been analyzed as a
 ARTICLE IN PRESS
34 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39

Fig. 9. Cost/Profit for implementation of the Kyoto Protocol for the period 1990–2010 with the participation of the USA. Plus means cost, minus
profits.

Fig. 10. Cost related to GDP for the period 1990–2010 if USA participates.

complementary action to domestic efforts for two shown that the marginal abatement cost is 67$/t C while
different cases. the total costs is almost 40% lower for the developed
According to a first scenario, only Annex I countries countries except for the USA where the cost reduction is
participate in the trading mechanism. From Fig. 8 it is little less. If all countries are participating in the trading
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39
the relevant figures are 33.2$/t C and 50–65% reduction
in the total cost. The problem in this case is located in
the reduction of the gains for Russia which are down
almost to the 13 of the corresponding for Annex I trading.
This is due to the fact that Russia is losing its dominant
role in the trading mechanism as a large amount of
emission credits are available in the trading mechanism
from the rest of the developing countries. The total cost
for the developed countries is low in such a case, as they
tend to buy cheap credits from the market instead of
implementing expensive domestic actions. More specific,
the developed countries manage to reach the Protocol
targets by limiting domestic efforts near 30% and using
the trading mechanisms for buying 70% of the emission
credits that the had to reduce. Developed countries in
fact sustain almost the same levels of emissions as in the
baseline case. Trading mechanism in this case acts as a
tool to cover the obligations towards the Protocol in a
cost effective way.
The Kyoto Protocol does not quantify the percentage
of domestic actions in the total efforts to reduce
emissions, but it declares clearly that domestic actions
should be preferred. Moreover, it was shown that
trading is the preferred action for the developed
countries and this actually counteracts to the Kyoto
targets. To overcome this, a restriction in the trading
quantities was assumed for the credit buying countries.
A limit of 30% of the region target was set for both
Annex I and International trading cases. The marginal
abatement cost is once again significantly lower than in
the case of domestic only efforts except for Russia and
the CEE countries which show a small increase. The
experiments showed that the gains for the developed
countries are substantial compared to the initial scenario
but far away from those that could have been succeeded
in the case that no restriction existed. The gains are 25%
and 50% for the Annex I countries (Fig. 9) compared to
the initial scenario for Annex I and International
trading respectively. The corresponding gains for USA
are a little less while the gains are decreased significantly
for Russia and the CEE countries.
The most important step in the Kyoto Protocol
negotiations was the introduction of sinks which
weakens the initial Protocol and reflects the current
form of the Protocol after the Marrakesh Accords.
Several authors have worked on the quantification of
the sinks for every region. Jotzo and Michaelowa (2002)
estimated the clean development mechanism market
after the Marrakech Accords and helped significantly in
the estimation of the new targets of the revised Protocol.
The Marrakech agreement was a success mainly due to
the satisfaction of Russia in its request of doubling the
sink credits. Sink credits reached 53.13 Mt C while the
total sum is 192.2 Mt C. Also the USA sinks reached
62.9 Mt C and the sum of the sinks of Russia and the
USA exceed 60% of the total sinks. This reflects the
35
willingness of the other parties to offer significant
bonuses to these two major countries in order to succeed
the ratification of the Protocol by these two parties. This
paper examines how the Marrakech Accords will affect
the total costs.
As it can be seen in Figs. 9 and 10 the introduction of
sinks reduces the total costs for the developed countries.
The cases that have been investigated are the same with
those examined for the initial Protocol. The results for
the revised Protocol are almost the same with results
from the corresponding experiments on the initial
scenario. The total cost for the revised Protocol is
nearly 10% less for all countries. The gains for Russia
are slightly increased by 2.5–10%. The most favorable
situation for Russia could be an Annex I trading system
with sinks also taken into consideration. The gains in
such a case exceed 2.5% of the GDP. But also in the case
where a restricted trading system is introduced, the gains
for Russia reach 1.7% of the GDP. A restricted trading
mechanism seems to be the most realistic case at this
period. This happens because EU focuses on domestic
efforts and on the other hand Russia insists on
guaranteed gains, as it was clearly declared in the recent
COP 9 that took place in Milan, (UNFCC (United
Nations Framework on Climate Change)). The intro-
duction of sinks is also a cost beneficial option for the
other developed countries leading to a reduction of their
total costs.
4.2. Evolution of the Kyoto Protocol without the
participation of the USA
The recent developments in the Kyoto Protocol
ratification process related to the participation of the
USA lead to an additional analysis of its evolution. As a
consequence of that a new structure of the cost is
expected differentiating the profits and deficits between
the regions. In this section of the paper we investigated
the evolution of the Kyoto Protocol without the
participation of the USA. The cases that were examined
are the same with those in the previous sections. But
additionally we investigate the case of a boycott
movement against USA because of its unwillingness to
ratify the Protocol.
Figs. 11–13 when compared to Figs. 8–10 show that
the USA absence influences significantly the economic
consequences from the Kyoto Protocol. The absence of
USA, which is the major emissions producing country
limits the emission credits buyers and this leads to a
significant reduction in the marginal abatement cost in
the trading mechanism. Marginal abatement cost
reaches 12.1$/t C in case of International trading and
10.2$/t C when also sinks are considered. Comparing
Fig. 8 with Fig. 11 it is shown that the marginal
abatement costs are a little lower for the other cases,
compared with the corresponding experiments with the
 ARTICLE IN PRESS
36 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39
participation of the USA, and this leads to a reduction
in the loses for the other Annex I countries. The gains
for Annex I countries are almost 50% for the cases
where no restriction exists in trading, while the gains in
the case where there exists a restriction in trading are
35% and lower. On the other hand Russia which is the
major emissions credits seller, sustains significant losses
compared to the case where the USA participates the
trading mechanism. The reduction in the gains of Russia
reaches almost 65% for the whole cases. It can be clearly
understood that the participation of the USA influences
the economic consequences of the Protocol more than
the trading mechanisms.
One possible counter action that could be taken is a
boycott movement against the countries that do not
participate. In this study a boycott movement against
products from the USA was examined, by introducing a
‘tax like’ 10% increase in the price of products from the
USA. Despite the World Trade Organization (WTO),
that could invoke counter measures against such a
movement, the experiment is possible in the sense that
citizens of other countries would not prefer USA
products if they are properly informed about the
negative attitude of the USA in the problem of global
warming. Boycott movement was examined for three
cases. First the case with only domestic efforts is
examined, where results in Figs. 12 and 13 show a
slight increase in the total cost for the Annex I countries.
Fig. 11. Marginal abatement cost for 2010 without the participation of USA.
USA suffers from the boycott movement because of
loses in the exports. In the other two cases, concerning
Annex I and international trading, results show a slight
increase in total cost compared to the respective cases
without the boycott movement. This increase hardly
reaches 0.1% of GDP. Loses in the USA economy are
increased and reach 0.6% of GDP. This result, though
significant, is less than the expected linear projection of
the 10% reduction in the USA economy. This is due to
the fact that trade with developed countries is less
important to the US than is trade with the US to these
countries. So, a boycott movement could become
partially a possible reaction to apply pressure in the
USA to ratify the Kyoto Protocol.
4.3. An estimation of the profits for Russia
The key factor concerning the ratification of the
Kyoto Protocol by Russia is that Russia will retain
certain profits by its participation. The current skeptical
position of Russia is justified by a number of reasons.
The withdrawal of USA, the emphasis of EU in
domestic efforts and the inclusion of other eastern
countries, such as Ukraine and Kazakhstan, in Annex I
group increases the carbon leakage, reduces the credit
buyers and as a consequence it minimizes the marginal
abatement cost and alleviates the dominant role of
Russia. It was attempted to reflect all these cases in this
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39
Fig. 12. Cost/profit for implementation of the Kyoto Protocol for the period 1990–2010 without the participation of the USA. Plus means cost,
minus profits.
contribution by studying the absence of USA, the
restriction in trading and the participation of other
eastern countries in the trading mechanism, respectively.
Therefore, it would be of interest to estimate the
expected incomes for Russia following its participation
with guaranteed gains. To account for this a guaranteed
minimum of 60% is set regarding the credits sold by
Russia compared to its total available credits. Further-
more investigation of the case that USA participates has
shown that Russia will certainly export credits of more
than 60%. In the following experiments it was assumed
that the USA does not participate to the Protocol.
In Fig. 14 it is shown that the consideration of a
guaranteed minimum does not affect in general the gains
for Russia, meaning that the credits sold are more than
60% in most cases. Only in the case of restricted
international trading the inclusion of such a minimum
leads to an improvement in the gains of Russia by 10%
and 23% for the cases without or with sinks, respec-
tively. The rest credit sellers share the losses introduced
by the secure minimum almost linearly.
5. Conclusions
This paper provides an integrated report of strategies,
concerning the implementation of the Kyoto Protocol
37
from its early signing until now adding new elements in
existing publications. The GEM, GTAP-E, was selected
for this contribution because of its detailed representa-
tion of the energy system and its ability to capture the
influence of energy policy in the economy. The different
policies were evaluated and compared with a reference
scenario, which was based on national reports. Com-
parison among different strategies showed the different
economic consequences for every region of interest. The
implementation of the Kyoto Protocol with only
domestic efforts proves to have significant costs for the
developed countries. Trading mechanisms may lead to a
considerable reduction of these costs and provide in
parallel significant gains for Russia. Also the introduc-
tion of sinks weakens the targets of the initial Protocol
and reduces the cost for the developed countries. The
participation of the USA seems to affect considerably
the cost of every partner more than the participation of
any other involved region. The absence of the USA can
not be alleviated by any measure considered in this
report, therefore it influences greatly the viability of the
Protocol. In absence of the USA, Russia must ratify for
the Protocol to be valid. Therefore we compared cases
where a specific minimum trading of the emission credits
is guaranteed for Russia to the case of a free trading
mechanism. Results show that Russia’s profits resulting
by its participation are present even with no warranties
 ARTICLE IN PRESS
38 A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39

Fig. 13. Cost related to GDP for the period 1990–2010 if USA does not participate.

Fig. 14. Profits for Russia under consideration of a secure minimum in its surplus credits sales.
 ARTICLE IN PRESS
A.S. Dagoumas et al. / Energy Policy 34 (2006) 26–39
in the cases examined except few cases of restricted
international emission credits trading. But these cases
seem unlikely to happen.
References
Armington, P.S., 1969. A theory of demand for products distinguished
by place of production. IMF Staff Papers 16, 159–178.
AIM (Asian-Pacific Integrated Model), http://www.nies.go.jp/social/
aim.
Babiker, M.H., Reilly, J.M., Mayer, M., Eckaus, R.S., Wing, I.S.,
Hyman, R.C., 2001. The MIT emissions prediction and policy
analysis (EPPA) model: revisions, sensitivities, and comparisons of
results. Report No. 71.
Bahn, O., Barreto, L., Kypreos, S., 2001. Modeling and assessing inter-
regional GHG Emissions Trading. IEA information paper.
Bernard, A., Reilly, J., Vielle, M., Viguier, L., 2002. The Effects of the
US Withdrawal from the Kyoto Protocol on International
Emission Trading. Paper prepared for the International Energy
Workshop jointly organized by the Energy Modeling Forum
(EMF), International Energy Agency (IEA) and IIASA.
Bohringer, C., 2002. Climate politics from Kyoto to Bonn: from little
to nothing?. The Energy Journal 23 (2), 51–71.
Burniaux, J., Nicoletti, G., Oliveira-Martins, J., 1992. GREEN: A
global model for quantifying the cost of policies to curb CO2
emissions. OECD Economic Studies 19, 123–140.
Burniaux, J.-M., Truong, T.P., 2002. GTAP-E: An energy- environ-
mental version of the GTAP Model. GTAP Technical Paper No. 16.
Chen, W., 2002. Carbon quota price and CDM potentials after
Marrakech. Energy Policy 31, 709–719.
Christopher, N.M., James, A.E., Son, H.K., Ronald, D.S., 1999. The
economics of the Kyoto Protocol. Special issue. ‘‘The Cost of the
Kyoto Protocol’’. The Energy Journal, 25–72.
Criqui, P., Mima, S., Viguier, L., 1999. Marginal abatement costs of
CO2 emission reductions, geographical flexibility and concrete
ceilings: an assessment using the POLES model. Energy Policy 27,
585–601.
den Elzen, M.G.J., de Moor, A.P.G., 2002. Analyzing the Kyoto
Protocol under the Marrakech accords: economic efficiency and
environmental effectiveness. Ecological Economics 43, 141–158.
Dimaranan, B.V., McDougall, R.A., 2002. Global trade, assistance,
and production: The GTAP 5 data base, Center for Global Trade
Analysis, Purdue University, http://www.agecon.purdue.edu/.
Ellerman, A.D., Decaux, A., 1998. Analysis of Post-Kyoto CO2
emissions trading using marginal abatement curves. MIT report 40.
EIA (Energy Information Agency, International Energy Outlook),
2003. http://www.eia.doe.gov/oiaf/ieo/.
Gusbin, D., Klaassen, G., Kouvaritakis, N., 1999. Costs of a ceiling on
Kyoto flexibility. Energy Policy 27, 833–844.
Hamasaki, H., 2002. The Economic and Environmental Impact of the
US Withdrawal from the Kyoto Protocol.
39
IPCC (Intergovernmental Panel on Climate Change), 2000. Special
report on land use, land-use change and forestry, http://
www.ipcc.ch/pub/srlulucf-e.pdf.
IEA (International Energy Agency). Climate change. http://
www.iea.org/.
IMF (International Monetary Fund), 2002. World Economic Outlook,
http://www.imf.org/.
Jotzo, F., Michaelowa, A., 2002. Estimating the CDM market under
the Marrakech Accords. Climate Policy 2, 179–196.
Kuik, O., Gerlagh, R., 2003. Trade liberalization and carbon leakage.
The Energy Journal 24 (3).
McKibbin, W.J., Wilcoxen, P.J., 2002. The role of economics in
climate change policy. Journal of Economic Perspectives 16,
107–129.
MURE (Mesures d’Utilisation Rationelle de l’Ènergie), http://
www.mure2.com/.
OECD (Organisation for Economic Co-operation and Development).
Climate change, http://www.oecd.org/.
Painuly, J.P., 2001. The Kyoto Protocol, Emissions Trading and the
CDM: An Analysis from Developing Countries Perspective. The
Energy Journal 22 (3), 147–169.
Paltsev, S.V., 2001. The Kyoto Protocol and sectoral contributions to
the carbon leakage. The Energy Journal 22 (4).
Rutherford, T.F., Babiker, M.H., 1998. GTAP-E: A global energy-
economy dataset. Technical Report, Economics Department,
University of Colorado.
Springer, U., 2003. The market for tradable GHG permits under the
Kyoto Protocol: a survey of model studies. Energy Economics 25,
527–551.
Thomas, H., 1997. Global Trade Analysis: modeling and applications.
Cambridge University Press, Cambridge.
UNFCCC (United Nations Framework on Climate Change). The
Kyoto Protocol to the Convention on Climate change, http://
unfccc.int/resource/docs/convkp/kpeng.pdf.
UNFCCC (United Nations Framework on Climate Change). COP 7.
Seventh session of the Conference of Parties, 29 October–9
November, Marrakesh, Morocco, http://unfccc.int/cop7/in-
dex.html.
UNFCCC (United Nations Framework on Climate Change). COP 9.
Ninth session of the Conference of Parties, 1–12 December, Milan,
http://unfccc.int/cop9/index.html.
UNFCCC (United Nations Framework Convention on Climate
Change). National Communications from Parties included in
Annex I to the convention, http://unfccc.int/resource/natcom/
nctable.html.
Viquier, L.L., Babiker, MH., Reilly, J.M., 2003. The costs of the
Kyoto Protocol in the European Union. Energy Policy 31,
459–481.
Weyant, J.P., Hill, J. (Eds.), 1999. The cost of the Kyoto Protocol: A
Multi-Model Evaluation. Special issue, 1–390.
Zhang, Z.X., 2001. An assessment of the EU proposal for ceilings in
the use of Kyoto flexibility mechanisms. Ecological Economics 37,
53–69.