Assessment of international mitigation costing studies in developing countries

Kirsten Halsnęs

UNEP Collaborating Centre on Energy and Environment

Risų National Laboratory, Denmark

1 Introduction

The establishment of the Framework Convention on Climate Change has motivated a large number of mitigation costing studies for developing countries. A variety of modelling approaches and input assumptions have been employed, and studies have been carried out by a range of institutions, including international research institutions, consultants and national organisations.

This paper attempts to bring together the main results and characteristics of some of the major recent studies. In spite of the diversity of teams involved in the studies, the similarities in scenario definitions and in the bottom-up approach used for the energy sector analysis, enable some key macro indicators and results of the studies to be compared.

Two main coordinated country study efforts have been carried out: by the Lawrence Berkeley Laboratory (LBL) and by UNEP Collaborating Centre on Energy and Environment (UCCEE). The results and methodological framework of these country studies are assessed in relation to similar country study activities for China, West Africa and South East Asia in order to evaluate possible common conclusions.

2 The assessed mitigation studies for developing countries

National climate-change mitigation studies for developing countries carried out so far have been primarily bottom-up studies. These studies have either assessed the abatement potential of individual technologies, or performed an integrated analysis for main emitting sectors such as the energy system. Some of the main studies are listed below.

Mitigation studies for developing countries.

The national studies for developing countries have focused on energy system analysis and CO2 reductions. In most cases the analysis is carried out for a scenario period from 1988/90 to 2020/30. Emission reductions are generally calculated as an increment to a baseline case which may be defined to reflect reference economic development and energy requirements (UNEP, 1994a), or as a high emission case for the energy sector (Sathaye and Goldman, 1991).

Not all studies have evaluated the costs of achieving abatement. Only the UNEP studies (UNEP, 1994a), the LBL study for India (Mongia, 1991), and the ADB study for China (Asian Development Bank, 1993) have assessed costs. These include direct investment costs, O&M and fuel costs. The Indian and Chinese country studies estimate such abatement costs for a specific abatement target at different points in time. The UNEP studies have estimated abatement costs for a range of target reductions spanning 12.5% to 25% reduction in 2005/10 and 25% to 50% reduction in 2020/30. The levelised abatement costs for these targets have been estimated following a cost curve format, in which individual technical options and comprehensive abatement strategies are ranked according to marginal costs (UNEP, 1994c).

3 Methodological approaches of the country studies

Comparability in bottom up studies has often been understood as synonymous with the use of a uniform modelling approach. For example two major series of studies conducted under the auspices of the IEA (ECN, 1994) and the CEC (COHERENCE, 1991) employed specific linear optimisation models, MARKAL and EFOM, respectively. The studies involved the implementation of the models for a number of countries and the use of common databases, requiring considerable financial resources and time. Such uniform modelling approaches have certainly contributed to capacity building in the involved countries, but have not always been closely related to ongoing national planning activities and existing national models in the countries concerned.

National mitigation cost assessment in connection with the FCCC will involve analysis for countries with a wide variation in energy resources and in other GHG emitting sources and with large differences in economic development level and objectives. It is likely that, in order to be comprehensive and reflect such national specific planning objectives, any uniform model would be very large, and easily become unmanageable and too complicated for national country teams to use. This is one important reason for the view that it is unrealistic to aim at using the same uniform model for mitigation cost assessment for a comprehensive range of country types at different stages of development, and in very different geographic, political and resource settings.

The coordinated country study programmes conducted by UNEP and LBL instead defined an analytical framework for mitigation studies comprising uniform assumptions and analytical structure allowing the use of different national modelling tools. These two coordinated country study efforts critically discussed below, evaluating key assumptions, critical issues and methodological guidelines for studies of developing countries.

The UNEP study defined a methodological framework for national abatement costing analysis, comprising a common analytical structure and a basic set of uniform assumptions (UNEP, 1994c). The analytical structure consists of four basic elements:

Standardized input assumptions defined average global economic growth, fuel prices, and emission factors, basic economic concepts were defined such as costs, prices and discount factors, and common energy units and reporting formats were specified. Uniform abatement targets and focal years were also defined to enable intercountry comparison of emission reductions and cost assessments, although some flexibility was permitted to facilitate harmonisation with existing national planning time horizons.

Two key areas were left for the national teams to decide: (a) the reference scenario for economic growth, consistent with the specified average global growth, and (b) the details of energy system development and technology data used in the reference and abatement energy scenarios. The rationale of allowing this freedom was, (a) that national economic and development aims and expectations should be reflected in the scenarios and (b) that the appropriateness and performance of different technologies should be evaluated in a specific national context. Thus both the identification of appropriate abatement technologies and the establishment of scenarios were part of the national studies. The notion of comparability therefore becomes quite complex. Comparability is more a reflection of the transparency and quality of the national studies than of "uniformity".

The UNEP study recommended that national teams define reference scenarios on the basis of a critical evaluation of official national macroeconomic and energy system forecasts. Assumptions should be made on expected efficiency improvements in the reference energy system development. Three different types of reference scenario were defined: (1) The "efficient" baseline, which assumes all profitable efficiency improvements in the energy system to be implemented; (2) The business as usual (or "inefficient") baseline, which assumes that the energy system will continue to include major inefficiencies due to capital constraints, lack of information and other barriers; and (3) The "most likely" development, representing a compromise between (1) and (2).

Another group of studies (Sathaye et al., 1991; Mongia et al., 1991; Davidson, 1993) used a uniform analytical structure and scenario approach. Compared with the UNEP studies these studies put more emphasis on uniform technology assumptions. Standardized international data for physical energy requirements for industrial production processes, household energy requirements for cooking, heating and cooling and for different modes of transportation have been used to project end-use energy demand.

The national scenarios were assumed to follow an S-curve which saturates at different internationally determined "activity levels". Statistics on current income distribution and energy consumption, together with the international framework, were used to project the future stock of household appliances and vehicles assuming saturation levels for the different technologies. In the same way the industrial structure was generally assumed to approach a determined pattern. Some countries like Brazil in a new study (La Rovere et al., 1994) used an input-output model to ensure consistency between the international aggregate development assumptions and the detailed production structure of Brazil. Finally, the energy projections in the LBL studies have been corrected for expected changes in national energy prices.

The Indian country study (Mongia et al., 1991) described two alternative reference cases. One reference scenario freezes the efficiency of demand and supply technologies at 1985 levels and assumes that fuel supply only changes through a small increase in the use of renewable energy sources. Another reference scenario assumes efficient technologies to be introduced in 2005 and 2025, and allows more fuel switching than the first scenario. The abatement scenario is assessed as an increment to these two alternative scenarios, producing two sets of abatement potentials and cost results. It should be noticed that the absolute emissions and marginal costs in the abatement scenarios are defined as being independent of the reference scenarios.

The different methodological approaches used in mitigation studies for developing countries have implications for the main study results. This is particularly evident for the projected reference energy requirements and emissions, and consequently for abatement potential and costs. The results of studies for developing countries are discussed in more detail below.

4 The development of GDP, primary energy consumption and CO2

emissions in the country studies

Two important indicators in the comparison of national abatement potentials and costs are assumptions on the development in:

The projected value of primary energy/GDP intensity is a measure of how much energy a country will require, in relative terms, to satisfy the projected economic growth. Besides representing the extent to which energy is used in the economy, the indicator also implicitly incorporates any expected changes in the conversion efficiency of energy technologies.

The CO2 intensity of primary energy consumption reflects the dependence of the energy system on fossil fuels. If the CO2 intensity of the baseline is high then energy efficiency improvements will clearly have a high value in terms of CO2 savings. If the CO2 intensity is projected to decrease in the future, then abatement measures relative to the baseline would, in principle, be correspondingly difficult to accomplish through efficiency improvements.

The abatement scenario is characterized by the above two indicators in the same way as the reference. In addition, the abatement scenario must include a number of options which lead to lower CO2 emissions. These options are either reductions of end-use energy demand or technical changes in the energy system comprising end-use efficiency improvements, improved conventional power supply technologies, fuel substitution, and the adoption of new supply technologies such as renewable energy sources or advanced power generation methods with low or zero CO2 emissions.

Growth rates of GDP, primary energy consumption and CO2 emissions are shown in Table 1 for the reference and abatement scenarios. There is a general tendency for higher GDP growth assumptions in the UNEP studies compared with the LBL studies. This may be a consequence of the UNEP link to official national forecasts. For Brazil there is a variation between 4.7% and 3.2% in the UNEP and LBL study, and for India the variation is between 6.3% and 5.0%. Venezuela on the other hand has a slightly higher GDP growth assumption in the LBL than in the UNEP study.

Table 1.Growth rates of population, GDP, primary energy consumption and CO2 emissions in reference and abatement scenarios for 1990 to 2020/30.

Country            Population  GDP   Primary               CO2                
                                      energy            emissions             

                                     Reference Abatement Reference  Abatement  

Argentina (1)         1.1      2.0     1.8       1.1       1.8        1.1     
Brazil                1.4      4.7     3.5       3.7       5.3        3.2     
Brazil (1)            1.2      3.2     2.5       1.0       2.8        0.8     
China (6)             0.8      6.0     3.1       2.7       2.6        2.1     
Egypt                 1.7      5.0     3.2       1.4       3.8        1.4     
Ghana (5)             2.9      4.0     3.6       3.0       5.4        5.0     
India                 1.8      6.3     4.2       3.8       3.8        2.9     
India (1)             2.0      5.0     4.6       4.3       4.6        4.3     
Indonesia (1)         1.2      3.0     3.6       3.2       4.0        3.5     
South Korea (4)       0.4      5.0               3.1                  2.8     
Mexico (1)            1.7      4.4     2.7       2.1       2.8        1.9     
Nigeria (5)           3.0      4.0     2.6       1.9       4.6        3.8     
Senegal               3.0      3.2     3.2       2.4       3.5        2.6     
Sierra Leone (5)      2.3      3.0     2.2       1.7       3.5        3.0     
Thailand              0.9      4.8     4.8       4.2       5.5        4.6     
Thailand (4)          2.0      8.0               7.8                  7.8     
Venezuela             2.1      3.8     2.5       1.9       3.1        2.1     
Venezuela (1)         1.7      4.0     3.4       3.2       2.5        1.9     
Zimbabwe              2.4      4.2     2.8       1.9       3.2        1.5     

Latin America (2)     1.5      3.3     2.7                 3.2                
Latin America (3)     1.3      3.2               0.9                 -0.6     

Southeast Asia(2)     1.4      4.6     3.9                 5.1                
Southeast Asia(3)     1.6      4.1               1.5                 -0.2     

(1) Energy & Environmental Division, LBL; CO2 Emissions from Developing Countries: Better Understanding the Role of Energy in the Long Term; July 1991.

(2) IPCC , 1992; Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment.

(3) Stockholm Environment Institute - Boston CentreTowards a Fossil Free Energy Future; April 1993.

(4) Asian Energy Institute; Collaborative Study on Strategies to Limit CO2 Emissions in Asia and Brazil; 1992.

(5) Davidson Ogunlade; Carbon Abatement Potential in Western Africa ; in P. Hayes, K. Smith, eds.,The Global Greenhouse Regime: Who Pays?; 1993.

(6) Asian Development Bank; National Response Strategy for Global Climate Change: People's Republic of China; Nov. 1993.

Unnumbered studies were part of the UNEP GHG Abatement Costing project.

The elasticities of primary energy to GDP, CO2 emissions to primary energy and CO2 emissions to GDP are shown in Table 2. Most of the countries expect primary energy consumption to grow more slowly than GDP in the reference as well as in the abatement scenario. Many of the countries have already in the reference case a primary energy/GDP elasticity between 0.6 and 0.8, which is low for a developing country. The reference case for China has the lowest primary energy/GDP intensity among the studies amounting to 0.5%. This is explained by the low efficiency of the present energy system and the introduction of nuclear power plants.

Table 2.Key elasticities in the abatement studies.

Country          Reference                    Abatement                       

                 Energy/GDP C/Energy  C/GDP   Energy/GDP C/Energy    C/GDP    

Argentina (1)       0.9       1.0      0.9       0.6       1.0        0.5     
Brazil              0.8       1.5      1.1       0.8       0.9        0.7     
Brazil (1)          0.8       1.1      0.9       0.3       0.8        0.3     
China (6)           0.5       0.8      0.4       0.5       0.8        0.4     
Egypt               0.6       1.2      0.8       0.3       1.0        0.3     
Ghana (5)           0.9       1.5      1.4       0.8       1.7        1.3     
India               0.7       0.9      0.6       0.6       0.8        0.5     
India (1)           0.9       1.0      0.9       0.9       1.0        0.9     
Indonesia (1)       1.2       1.1      1.3       1.1       1.1        1.2     
South Korea (4)                                  0.6       0.9        0.6     
Mexico (1)          0.6       1.0      0.6       0.5       0.9        0.4     
Nigeria (5)         0.7       1.7      1.1       0.5       2.0        1.0     
Senegal             1.0       1.1      1.1       0.8       1.1        0.8     
Sierra Leone(5)     0.7       1.6      1.2       0.6       1.8        1.0     
Thailand            1.0       1.2      1.1       0.9       1.1        1.0     
Thailand (4)                                     1.0       1.0        1.0     
Venezuela           0.7       1.3      0.8       0.5       1.1        0.6     
Venezuela (1)       0.9       0.7      0.6       0.8       0.6        0.5     
Zimbabwe            0.7       1.2      0.8       0.4       0.8        0.4     

Latin America(2)    0.8       1.2      1.0                                    
Latin America(3)                                 0.3       -0.7      -0.2     

Southeast Asia(2)   0.8       1.3      1.1                                    
Southeast Asia(3)                                0.4       -0.1       0.0     

(1) Energy & Environmental Division, LBL; CO2 Emissions from Developing Countries: Better Understanding the Role of Energy in the Long Term; July 1991.

(2) IPCC , 1992; Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment.

(3) Stockholm Environment Institute - Boston CentreTowards a Fossil Free Energy Future; April 1993.

(4) Asian Energy Institute; Collaborative Study on Strategies to Limit CO2 Emissions in Asia and Brazil; 1992.

(5) Davidson Ogunlade; Carbon Abatement Potential in Western Africa ; in P. Hayes, K. Smith, eds.,The Global Greenhouse Regime: Who Pays?; 1993.

(6) Asian Development Bank; National Response Strategy for Global Climate Change: People's Republic of China; Nov. 1993.

In most of the studies the CO2/primary energy elasticity is above unity in the reference scenario. A remarkably high value is projected for The UNEP Brazil study and for Ghana, Nigeria and Sierra Leone in the reference. These high values are a consequence of an extensive replacement of biomass by fossil fuels in the reference. The resulting CO2/GDP elasticity in the reference scenarios for most of the country studies is close to unity.

The different assumptions on the development in GDP, primary energy and CO2 emissions in the reference case implies that the imposition of abatement targets related to future baseline emissions would have quite different implications for the absolute development in emissions in the abatement scenarios compared with today's level. This variation is illustrated in Figures 1 to 4 which show reference and abatement development in CO2 emissions in the UNEP and LBL country studies respectively.

The UNEP reference emissions show a large variation ranging from a six to eight-fold projected increase, for Thailand and Brazil, and to a two to three-fold increase for the other developing countries. In the abatement scenario, after the 30% reduction from baseline, Thailand still projects a six-fold increase in emissions, and Brazil project a three-fold increase after a 50% reduction. The rest of the developing countries project emission increases up to a doubling in the abatement scenario after a 40%-50% reduction in 2020/30. The high emission projections for Thailand and Brazil in the reference and abatement scenario can be explained by a large contribution of fossil fuels in the supply system. In addition the Thailand study includes very low expectations for energy efficiency improvements.

The LBL high emission (reference scenario) shows a variation from a high six-fold increase for India, five-fold increase for Indonesia and an increase of between two and three times for China, Argentina, Brazil, Mexico and Venezuela. In the low emission (abatement scenario) the emissions are projected to grow by only about 50% from present levels in Brazil after a 54% reduction, and in Argentina after about 25% reduction. China, Mexico and Venezuela project emission increases between a doubling and tripling of present today's level after 20% to 30% baseline reductions. India and Indonesia project after the high reference emissions also a persistent high abatement emission level amounting to a four to five-fold increase from present after very small reductions of 13% and 18% from baseline, respectively.

A striking difference between the UNEP and LBL approaches is evident in the case of Brazil. The UNEP Brazil study assumes high reference growth in primary energy consumption and increasing CO2/primary energy intensity. This is partly a consequence of a low projected fuel price increase in the UNEP study, which makes the present biomass energy supply unprofitable seen from a pure "fuel-economic" point of view. The LBL study for Brazil projects a much slower increase in CO2 emissions in the reference and abatement case due to lower economic growth assumptions, lower CO2/primary energy intensity and a larger potential for end-use demand efficiency improvements in the abatement scenario. A new Brazilian study (La Rovere et al., 1994) has been carried out as a compromise between the assumptions of the UNEP and LBL study.

5 Abatement costs

The UNEP country studies include an estimation of abatement costs for emission reductions between 12.5% and 25% from baseline in the short-term (2005/10) and on 25% to 50% from baseline in the long-term (2020/30). Cost is defined as direct cost including investment, operation and maintenance and fuel costs (UNEP, 1994c). Cost curves containing a number of target reductions in one year should be regarded as snapshot pictures of the levelised cost of achieving a given reduction in that year. Figures 5 and 6 show the cost curves for a number of participating countries in the short-term and long-term target years, respectively.

The short-term cost curves for Brazil, Thailand and Zimbabwe exhibit a number of similarities. A particular feature is the large potential for negative cost abatement options up to 10% to 15% reductions from baseline. The curves are also similar in shape: the first part of the cost curves up to about 5% reductions indicates very low abatement costs, followed by a long interval up to about 25% reductions in which abatement costs fall within a relatively narrow range between - $10 and + $30 marginal cost per tonne CO2 reduced. Few options were included in the analysis after this interval. The options that do occur tend to be expensive supply-side options, thus implying a steep increase in marginal costs the reduction targets increase.

Estimated abatement costs for the participating industrialised countries Denmark and The Netherlands are also presented in the figure for illustrative purposes. In Denmark a substantial negative cost potential is identified, while the Netherlands estimate much higher abatement costs. The divergent results can be explained to a certain extent by different assumptions regarding the inclusion of "no-regret" options in the reference case.

The long-term marginal cost curves shown in Figure 6 show many similarities for the developing countries in line with the short term curves. Venezuela may be regarded as an exception due to entirely positive curve which arises from the inclusion of all "no-regret" options in the baseline.

Senegal has a negative marginal cost interval up to 8% reduction from baseline in 2020/30 followed by very low costs amounting to less than $10 up to the maximum 50% reduction. The potential for negative marginal cost abatement is expanded to about 30% and 40% reduction for Egypt and Zimbabwe respectively.

As in the short-term curves, the long term marginal cost falls within an interval of - $10 to + $25 per tonne CO2 for a broad emission reduction interval from about 10% to 40% reduction.

Table 3 shows marginal, average and total abatement costs for the short and long term reduction targets related to GDP of the countries. In the long term, the average abatement costs for reductions from baseline between 25% and 50% range between -$5 and +$37 per tonne CO2 with France as an exception. The total abatement costs are negative for Egypt and Senegal.

Figure 5. UNEP studies: marginal abatement costs for 2005/10.

Figure 6. UNEP studies: marginal abatement cost in 2020/30.

Table 3.Marginal (MAC), Average (AAC) and Total Abatement Costs (TAC) for maximum reductions, comparison with GDP.

(a) Short term                                                                             

Country    Year  Baseline  Reduction    Reduction    MAC1      TAC     AAC      GDP  TAC/GDP                 emission  (Mt.CO2)         %    ($/t.CO2)  ($mill.) ($/t.CO2) ($bn)    %                
(Mt.CO2)Brazil     2010      767.8     101.3      13.2        25     -1881   -18.6      848   -0.22 
Thailand   2010      359.8      99.3      27.6       171      1027    10.3      250    0.41 
Zimbabwe   2010       32.7       6.9      21.1       289       221    32.0       35    0.64 
Denmark    2005       53.4      11.3      21.2       529       -70    -6.2      154   -0.05 
France2    2005      407.0      50.9      12.5       -        9725   191.2     1463    0.66 
Netherlands2010      185.0      46.3      25.0       121      1440    31.1      379    0.38 

(b) Long  term                                                                             

Country    Year  Baseline  Reduction    Reduction    MAC11      TAC     AAC      GDP  TAC/GDP                 emission  (Mt.CO2)         %    ($/t.CO2)  ($mill.) ($/t.CO2) ($bn)    %                

Brazil     2025     1611.2     741.2      46.0       29        9339    12.6  1740    0.54 
Egypt      2020      253.0     141.7      56.0        2        -732    -5.2   118   -0.62 
Senegal    2020       15.4       7.7      50.0        3         -16    -2.0    14   -0.12 
Thailand   2030      751.4     221.7      29.5      171        2089     9.4   538    0.39 
Venezuela  2025      189.4      50.4      26.6       56         685    13.6   177    0.39 
Zimbabwe   2030       57.4      22.2      38.7      289         205     9.3    72    0.28 
Denmark    2030       59.3      29.0      48.9      614         168     5.8   211    0.08 
France2    2030      615.0     153.8      25.0        -       11756    76.5  2400    0.49 
Netherlands2030      204.3     102.2      50.0       78        3775    36.9   548    0.69 

1. The marginal abatement cost (mac) here corresponds to the level at the maximum reduction achieved for the country.

2. The average abatement cost for france is set equal to the rate of carbon tax necessary to achieve the reductions.

The Indian country study (Mongia et al., 1991) estimated abatement costs for two abatement scenarios. One assumes a least cost supply system requiring foreign exchange and capital resources. Another abatement scenario assumes a more expensive supply solution with greater reliance on domestic renewable energy sources The first (and cheapest) abatement case estimates average abatement costs to be -$9.5 per tonne of CO2 for a 45% reduction from an inefficient reference in 2025. Measured in relation to an efficient reference in the same year the average abatement costs for a 22% reduction is assessed to be +$10.9 per tonne of CO2. It should be noted that the absolute amount of CO2 reduced in the two alternatives is the same.

6 Technical options in the national studies

The country studies have estimated abatement potential, and in some cases costs, taking an evaluation of technical options as the starting point For the mitigation costing studies, such as the UNEP study, the abatement potential can be related to individual technologies or to comprehensive packages of technologies. The cost and abatement potential of the options have been assessed using energy system models to evaluate end-use efficiency improvements, fuel substitution and new supply technologies in an integrated way. This analysis has in practice been done with different degrees of sophistication (UNEP, 1994a).

A summary of the main technical abatement options in the UNEP country studies is presented in Table 4. The national options are listed in aggregated form and thus represent classes of options rather than individual technologies. The level of aggregation varies, depending on the level of detail in the country studies, ranging from the assessment of 57 options in the study of Egypt, to less than ten options in the Brazil, Senegal and Thailand studies.

Table 4.Main categories of CO2 reduction options and related marginal abatement cost (MAC) in the participating developing countries for the long-term target.

Country   CO2 reduction option                     %    cost   

Brazil    Electricity savings (industry. services  7    -66    
          and residential)                                     

          Solar uses in agriculture                1    22     

          Fuelwood and charcoal for afforestation  21   24     

          Ethanol,  bagasse and electricity        19   29     
          generation from bagasse                              

          Total CO2 reductions                          48     

Egypt     Fuel switching in households             6    -21    

          Efficient industrial equipment and       10   -12    

          Transportation                           2    -12    

          Heat recovery and new industrial         9    -8     

          New raw materials                        5    -3     

          Efficient household appliances           5    -3     

          Electricity generation                   8    -1     

          Efficient stoves                         7    2      

          Total CO2 reductions                     52          

Senegal   Early hydropower implementation          0.1  -210   

          Agriculture intensification              5    -28    

          Energy conservation in  industry         0.4  -4     

          Dissemination of improved stoves         11   0      

          Improve carbonization efficiency         13   1      

          LPG-charcoal substitution                15   2      

          Biomass from afforestation               6    3      

          Total CO2 reductions                     50          

Thailand  Efficient air conditioners               2    -36    

          Electronic ballast                       1    -27    

          Compact fluorescent lamps (service       6    -14    

          Compact fluorescent lamps (residential   2    -9     

          Nuclear electricity                      18   20     

          Highly efficient gasoline cars           1    92     

          Total CO2 reductions                     30          

Venezuela Reduced flaring and leakage of methane   7    3      

          Efficient boilers and kilns              10   10     

          Freight transport                        1    13     

          Efficient electric motors in the         2    17     
          industrial sector                                    

          Passenger transport                      4    21     

          Electric sector                          0.6  35     

          Other energy savings in the industrial   2    39     

          Efficient electrical appliances          0.4  52     

          Total CO2 reductions                     27          

Zimbabwe  Efficient boilers                        23   -9     

          Energy savings in the industrial sector  4    -2     

          Efficient motors and power factor        2    -2     

          Increased hydropower                     5    5      

          Efficient furnaces                       2    66     

          Central photovoltaic power               1    153    

          Coal for ammonia                         1    289    

          Total CO2 reductions                     38          

One general similarity between the country studies is that the least expensive part of the cost curve contains energy end-use savings in households and/or industry. Another common result is that electricity supply options first appear in the second third of the reduction potential of the cost curves.

The highest level of detail in the country studies generally occurs in the first part of the cost curve relating to end-use savings. The countries typically concentrated on one sector and one main category of technologies, which were considered to be important for that country, such as industry in Egypt, Venezuela and Zimbabwe, and households in Senegal. The studies were not exhaustive and it is therefore likely that further study would lead to the identification of more low-cost options, and thus an extension of the first low-cost segment of the cost curves.

On the supply side, most of the studies assessed traditional energy-supply technologies and few included more advanced technologies and/or renewable energy technologies. Consequently the cost curves either increase very sharply or simply do not consider any further reduction possibilities after the exhaustion of these options.

The exhaustion of low-cost options (i.e. marginal cost less than $20-30 per tonne CO2) is reached at about 30% reduction from baseline in 2025 for Brazil, about 10% reduction for Thailand, about 25% reduction for Venezuela and at about 35% reductions for Zimbabwe. It is likely that the exhaustion of such low-cost options could have been followed by a more gradual increase in costs if more abatement technologies with higher costs were included. Examples of such intermediate-cost options might be renewable energy technologies and additional end-use savings. Consequently a larger reduction interval could probably be achieved without such a steep increase in marginal abatement costs.

In the Egypt study the marginal cost develops fairly linearly up to the maximum estimated reduction potential of 56% in 2020. The considerable potential for low-cost abatement is associated with a large capacity for low-cost energy conservation, particularly in industry, combined with optimistic assumptions regarding the implementation of such measures.

The studies done by LBL have in general included more details on end-use efficiency measures than the UNEP studies. The two coordinated sets of studies thus complement each other with regard to coverage of all relevant abatement technologies. Nearly all the LBL studies assessed a large number of end-use options, amounting to a significant potential for energy-efficiency improvements in households, industry and transport. The contribution of end-use demand efficiency improvements to abatement can be measured by comparing end-use demand reductions with emission reductions in the abatement scenario. Values of the final energy reduction/CO2 reduction ratio are shown in Table 5 for LBL studies and UNEP studies.

Table 5.The ratio of final energy savings to emission reductions in LBL and UNEP studies (for long term emission reductions in 2020, 2050 or 2030)

Study/country             Reduction (%) Final energy reduction/CO2reduction            

LBL studies                                                               
India                           13                1.2               
Indonesia                       18                0.8               
China                           20                0.7               
Argentina                       26                0.7               
Brazil                          54                0.7               
Mexico                          30                0.7               
Venezuela                       21                0.4               

UNEP studies                                                              
Brazil                          48                0.2               
Egypt                           52                0.9               
Senegal                         50                0.4               
Thailand                        30                0.2               
Venezuela                       27                0.6               
Zimbabwe                        38                0.6               

All the LBL country studies have a final energy demand reduction/CO2 reduction ratio on at least 0.7, with Venezuela as an exception. The UNEP studies show a larger variation with a high final demand reduction/CO2 reduction ratio of 0.9 for Egypt, which is the UNEP study with the highest number of end-use efficiency options assessed. A remarkably low ratio is found for Brazil, Senegal and Thailand. The abatement achieved in these country studies is to a great extent a consequence of substitution to low carbon intensive fuels.

In conclusion there seems to be a tendency in the LBL studies to put most emphasis on end-use efficiency improvements, while most of the UNEP studies have assessed a limited number of end-use options, but included more radical changes in the supply system structure in the reference and abatement scenarios. Although the economics and efficiency of technologies relating to end-use and supply may be highly site and system specific, it is likely that the most of the country studies could be supplemented with more technologies. In the case of the UNEP study this would result in an expansion of the first part of the cost curves containing end-use efficiency improvements followed by a cost curve segment with more low-cost renewable technologies and other supply technologies. For the LBL studies more far extensive emission reductions could probably be achieved through further structural changes in the supply structure.

7 Implementation issues

One general similarity between all the mitigation studies for developing countries is that they include a significant potential for energy end-use savings in industry/or households. These options have been assessed to have negative or very low abatement costs.

In order for the CO2 reduction to be realised in practice however, such options must be implemented by industries or private households, requires widespread information and access to financial sources by the many individual decision makers.

No formalized procedure for estimating implementation costs has been defined in any of the developing country studies. Implementation problems were treated in some of the studies by defining expected penetration rates for end-use technologies, assuming for example that 25% of an identified technical potential could be implemented in the short term and a larger share in the long term. These assumptions are made as exogenous input to the bottom up calculations.

Some lessons on penetration rates and regulation efforts can certainly be learned from the experience with demand-side management (DSM) programmes in industrialized countries, and this could contribute to the establishment of a more formalized evaluation of implementation processes. It must however be recognized that any implementation process is nationally unique, and that any response from decision makers is dependent on the entire national regulatory system and on a broader range of lifestyle and cultural factors.

Consequently, the identified abatement potential and the estimated "cost elements" in the UNEP study only address part of the resources required to put an abatement options into place in a specific country. A further step in the identification of an implementation strategy should involve the establishment of a dialogue with national governments, public utilities and interest groups representing industry and private consumers in order to find appropriate regulatory instruments and to make the necessary capacity building.

8 Comparability of studies using different methodological frameworks

The mitigation costing studies for developing countries considered here exhibit similarities with regard to the assessed potential for negative or low-cost abatement. In general these options comprise end-use energy efficiency improvements, energy supply efficiency improvements and introduction of fuels with lower carbon-intensity. These technologies cover the first and cheapest part of the abatement potential and can especially in the longer term be supplemented with renewable energy technologies, more far-reaching end-use savings and advanced combustion technologies. The long term abatement potential will consequently be extended and as estimated in the UNEP studies also cheaper compared with short term reduction targets.

The individual country studies are difficult to compare quantitatively with regard to detailed assessments of abatement potential and costs. This is a consequence differences in methodological approach and in scenario assumptions on economic growth, energy requirements and abatement potential and costs.

A uniform methodological framework has been used in the UNEP studies and LBL studies, respectively, and some conclusion can therefore be made on the comparability of different methodologies on the basis of these studies. The major differences between the methodological approaches used in the UNEP and LBL studies relate to the baseline definitions for the economy and energy sector development and to the detailed technology assumptions.

Economic development scenario. In the UNEP framework the national teams were recommended to take official macroeconomic forecasts as the starting point for energy demand projections. In contrast the LBL framework used international economic focal points to describe a possible economic structure and income distribution which could be achieved in a developing country within a given time horizon. It could for example be assumed that a country like Brazil would approach an economic structure like Spain in a given time frame.

The advantage of using national macroeconomic projections is obviously that it reflects national planning efforts. Formalized macroeconomic models and long-term economic projections, however, do not exist for many of the developing countries, which makes it difficult to make valid projections. Available national forecasts may therefore sometimes be only partially consistent and realistic. Using international economic data can especially help to establish an internally consistent set of data. Studies made on such a common well documented background can also be somewhat easier to compare than studies using different national forecasts. The use of international economic data to make national forecasts may, however, de-couple the mitigation studies from national development and priorities, thus reducing the relevance of the study to application within the context of the Climate Convention.

Baseline energy scenarios. Countries participating in the UNEP study effort were recommended to make assumptions on expected efficiency improvements in the reference energy system on the background of national economic development programmes and availability of capital and foreign exchange. The country teams were free to make judgements on the most likely implementation rate for profitable efficiency improvements in the reference. This meant that a country like Venezuela assumed all profitable efficiency improvements to be implemented, while other countries like Brazil and Thailand assumed a relatively "inefficient" reference scenario. Another difference in national definitions of reference scenarios relates to assumptions on large structural changes in fuel supply. Major supply system changes were included in the Brazil study, where the reference case assumes replacement of present biomass use, including ethanol, with fossil fuels.

The LBL studies defined the reference energy scenarios in order to reflect a continuation of historical trends. This approach can be said to be in line with the "business as usual" approach used in many energy system analyses for industrialized countries. These reference scenarios in general assume gradual changes in supply structure and a slow rate of implementation of efficiency improvements. Taking Brazil as an example again, this means that LBL's reference scenario assumes biomass to continue in Brazil.

Comparing the UNEP and LBL approaches, it appears that the UNEP approach has a tendency to impose a larger variation in reference emission projection with an upward "push" if countries assume biomass and hydropower to play a low role in the reference and with a downward "push" if many profitable efficiency improvements are assumed to be implemented in the reference.

Technology data. National specific investigations on technologies were used in the UNEP studies, while the LBL studies used international standardized data on efficiency and penetration rates.

The real difference between the two approaches is difficult to assess because national technology choice and assessment in the UNEP studies was influenced by available international technology data as used in the LBL studies. A major difference does however arise from the technology menu assessed in the two group of studies especially regarding end-use technologies. LBL, through a standard menu of end-use options, achieved a more comprehensive coverage of possible end-use energy savings than the UNEP studies. The technologies assessed in the UNEP studies vary in number and coverage between the countries. Thus Egypt covered all relevant end-use technologies and sectors, while others primarily covered either industry or the household sector alone.

In conclusion the different approaches to the definition of reference scenario and technology assumptions can lead to national emissions forecasts, abatement potentials and costs which diverge in absolute terms. Differences may especially arise from reference scenario assumptions on the implementation of efficiency improvements.

If a reference scenario is defined to include all profitable efficiency improvements, then all negative cost options are assumed to be included in the reference and only positive cost options are left for abatement. The negative-cost options are however scheduled for implementation in the scenario period, regardless of whether they are referred to as profitable efficiency options or negative-cost abatement options. This means that the reference and abatement scenarios consider the same types of technologies, but that these can be assigned to be part of different scenarios depending on assumptions on efficiency improvements in the reference.

A given emission reduction from baseline will in general be more expensive and will imply the use of more costly and advanced technologies, if it is measured in relation to an "efficient" reference energy system than to an "inefficient" one.

9 Conclusion

The climate change mitigation costing studies carried out for the energy sector until now for developing countries exhibit, despite differences in methodological approach, some common empirical results:

A key methodological issue for mitigation studies for developing countries is to define a common approach for defining reference scenarios for the energy sector and other main GHG emitting sectors. In this context it should be recognized that the GHG reduction objective should be evaluated in relation to a comprehensive national development context, and that the implementation of any abatement option will be interrelated with general economic development and performance. This means the link between energy forecasts and macroeconomic projections must be improved, and that a broader range of national institutions and interests must be considered in the evaluation of abatement options and strategies.


Asian Development Bank (1993). National Response Strategy for Global Climate Change: Peoples Republic of China.

Asian Energy Institute (1992). Collaborative Study on Strategies to Limit CO2 Emissions in Asia and Brazil.

COHERENCE (1991). (JOULE CEC DGXII) Cost-effectiveness analysis of CO2-reduction options; synthesis report. CEC DGXII report.

Davidson, O. (1993). Carbon Abatement Potential in Western Africa; in Hayes, P., and Smith, K.eds.(1993) The Global Greenhouse Regime: Who Pays?

ECN (1994). Boundaries of Future Carbon Dioxide Emission Reduction in Nine Industrial Countries. Netherlands Energy Research Foundation, ECN-C-94-025.

IPCC 1992 (1992). Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment.

La Rovere, E., Legey, L. F. L. Miguez, J. D. G. (1994). Alternative Energy Strategies for Abatement of Carbon Emissions in Brazil. (to be published in Energy Policy)

Mongia, N., R. Bhatia, R, Sathaye, J. and Mongia, P. (1991). Cost of reducing CO2 emissions from India, Energy Policy vol. 19, no. 10, p. 978-987.

Sathaye, J and Goldman, N. eds. (1991). CO2 Emissions from Developing Countries: Better Understanding the Role of Energy in the Long Term. Volume II: Argentina, Brazil, Mexico and Venezuela; and volume III: China, India, Indonesia and South Korea. Energy & Environment Division, Lawrence Berkeley Laboratory.

Stockholm Environment Institute (1993). Towards a Fossil Free Energy Future. A Technical Analysis for Greenpeace International. SEI-Boston.

UNEP (1994a). UNEP Greenhouse Gas Abatement Costing Studies. Part One: Main Report, UNEP Collaborating Centre on Energy and Environment, Risų National Laboratory, Denmark, May 1994.

UNEP (1994b). UNEP Greenhouse Gas Abatement Costing Studies. Part Two: Country Summaries. UNEP Collaborating Centre on Energy and Environment, Risų National Laboratory, Denmark, May 1994.

UNEP (1994c). UNEP Greenhouse Gas Abatement Costing Studies. Annex: Guidelines. UNEP Collaborating Centre on Energy and Environment, Risų National Laboratory, Denmark, May 1994.

Back to Contents