Issues in conducting GHG mitigation assessments in developing countries

Jayant A. Sathaye

Lawrence Berkeley Laboratory, Berkeley, California

1 Introduction

The Framework Convention on Climate Change, signed by more than 150 governments worldwide, calls on parties to the Convention to undertake inventories of national sources and sinks of greenhouse gases and to develop plans for responding to climate change. Several institutions, including UNEP, have initiated programs to assist developing countries and countries with economies in transition to meet this obligation. For instance, the U.S. government has committed $25 million to support climate change country studies. This program is supporting experts in 56 countries to complete their inventories, and assessments of vulnerability and adaptation opportunities and mitigation options.

This paper describes a mitigation methodology that is being used for these country studies, and discusses issues that have arisen in conducting mitigation assessments for developing countries in the past. The material in this paper is largely based on a presentation made by the author at a United Nations Environment Programme (UNEP) Conference in June 1994 at Copenhagen on mitigation methods and costs of reducing carbon emissions.

2 Structure of a mitigation assessment

The structure of a mitigation assessment will vary depending upon its goals and scope. A basic structure for a comprehensive national assessment is illustrated in Figure 1. Conducting an assessment involves incorporating different types of data into several types of analysis. A key analytical step involves a screening of potential mitigation options to select those that will receive further analysis. Examples of criteria that may be used for screening include:

Potential for large impact on greenhouse gases (GHGs)

Direct and indirect economic impacts

Consistency with national development goals

Consistency with national environmental goals

Potential effectiveness of implementation policies

Sustainability of an option

Data availability for evaluation

Other sector-specific criteria

Figure 1. Structure of a mitigation assessment

Once options have been selected for analysis, it is necessary to characterize technologies and mitigation options with respect to their costs and benefits. Various models and methods may be used to assist in this evaluation, depending on the sector that is being analyzed. Data gathering for evaluation of options will also provide general data on technologies, which can be incorporated into scenarios.

The construction of scenarios also requires data on the activities that produce GHG emissions. The types of data vary among sectors. They include output of key production industry, number of urban and rural households in the residential sector, number of vehicles in transportation, and demand for land and forest products. Development of scenarios requires a projection of the future levels of each kind of activity. Such projection in turn draws on assumptions made about growth in population, GDP, and other macro variables.

A mitigation assessment should include at least two scenarios for each sector considered. A "baseline" or "reference" scenario is a description of a plausible future in which no specific actions are taken to reduce GHG emissions or to store carbon. A baseline scenario could include some degree of adoption of technology improvements or other actions that improve the efficiency of resource use. A "mitigation scenario" describes a future that is essentially similar to that in the baseline scenario with respect to overall economic and social trends, except that it assumes that efforts are made to encourage adoption of measures that will reduce GHG emissions or store carbon.

In selecting mitigation options for inclusion in a mitigation scenario, standard techniques of benefit-cost analysis are usually applied, with some modification. Some of the impacts of a mitigation option may be difficult to quantify or to incorporate in a common measurement. Thus, benefit-cost analysis should be supplemented with assessment of other criteria such as complementary environmental effects, secondary economic effects (e.g., employment creation), and social and political considerations (e.g., the impact on societal equity).

The combination of benefit-cost analysis and assessment of other criteria results in a ranking of mitigation options, which may be used to construct a mitigation scenario. A mitigation scenario may reflect not only the technical potential of various options to reduce GHG emissions or to store carbon, but also the part of that potential that is actually achievable. An assessment of the achievable potential requires identification and consideration of policies that could be used to encourage adoption of mitigation options in each sector.

The analysis of individual mitigation options provides an e stimate of the micro-economic impacts of each option. To assess the macro-economic impact of a set of mitigation options, modeling of the interaction between particular sectors and the overall economy may be conducted. Various models may be used to analyze the interaction between the energy sector and the economy. Similar models have not been used to assess the impact on the economy of mitigation options in non-energy sectors, but models used for general analysis of agriculture and forestry could be modified for analysis of macro-economic impacts of mitigation options in those sectors.

The general structure described above applies for both the energy and non-energy sectors. However, the analysis of the energy sector differs from the analysis of the non-energy sector in two ways. First, analysis of the energy sector requires an integration of energy demand and supply in order to estimate the GHG impacts of mitigation options in the energy demand sectors, and to compare demand-side and supply-side options. Second, various tools are available that can be used to assess the impact of energy-sector mitigation options on the overall economy. Thus, the analytical process for the energy sector involves these additional steps of integration and energy-economy modeling.

3 Issues in mitigation assessment

We review several issues that were highlighted in earlier developing-country mitigation studies supported by the US Environmental Protection Agency through the Lawrence Berkeley Laboratory and by United Nations Environment Programme (UNEP) through its Collaborating Center for Energy and Environment at the Riso National Laboratory (Sathaye and Christensen, 1994).

Estimating incremental cost

Several articles of the FCCC refer to incremental costs of actions. The incremental costs are broadly defined as additional costs on countries beyond the costs that are necessary for achieving their development goals, but which generate additional global benefits. For sector analysis, it is important that a baseline scenario be defined and agreed to in order to estimate the incremental costs of a mitigation scenario.

Baseline scenario

The baseline scenario may include projections for several future years or for a single future year. Creating a baseline involves choices dictated by analytical and other considerations. In transition countries, a baseline will be quite different from historical trends. The baseline may include some efficiency improvements or none. Selecting the baseline is an important step in the scenario process, since the incremental costs and benefits of a mitigation option will depend on the definition of a baseline scenario.

Defining a baseline will be more difficult at the sector level, where many interacting forces come into play, than at the project level. Furthermore, in most sectors, the baseline used for determining incremental cost will be open to negotiation between recipient and donor countries.

Mitigation scenario

A mitigation scenario may be based on either the technical potential for reducing GHG emissions or storing carbon, or on the achievable potential, which would consider the many factors - institutional, cultural, legal, etc. - that may limit the implementability of the technically available options. Ideally, both the technical and the achievable potential should be reported.

The choice of a baseline is also influenced by the analytical methods used for estimating costs. The "topdown" analytical method assumes a macroeconomic perspective, wherein mitigation costs are defined in terms of losses in economic output, income and GDP. A key assumption underlying many topdown analyses is that the baseline scenario represents the economy in equilibrium, with all factors of production employed efficiently given prevailing prices. Thus mitigation options are interpreted to cause deviations from an equilibrium situation which results in positive costs to the economy. By contrast, "bottomup" models focus on individual processes such as enduse energy consumption, which do not have to assume a market equilibrium, thus intervention in the market place may be achieved at a negative cost.

A mitigation scenario may include only those options that have been analyzed in some detail. In addition, a mitigation scenario may consider the overall potential for GHG mitigation in a sector or in the country, including options that have not been specifically evaluated. Thus, for some countries a mitigation scenario might include a relatively small amount of well-defined GHG abatement, while for others it may also include a large amount of abatement that has been more roughly estimated.

The estimation of incremental costs is sensitive to several factors that can easily change the costs by orders of magnitude and from positive to negative. Table 1 presents some of the salient assumptions used to illustrate the sensitivity of incremental costs (Table 2) to different assumptions about the capital cost of power plants. For this illustration, we chose a wind farm and a coal fired power plant which generate the same amount of electricity annually. A wind farm may displace the electricity generated by peaking (oil and gas), intermediate (coal) or base load (nuclear) units (Grubb M. and Meyer N., 1993) depending on how the daily load curve matches the characteristics of the wind-farm site. For our illustration, the wind and coal power plants have the following characteristics (Table 1).

Table 1.Illustrative wind and coal power plant characteristics.

                                      Coal                Wind Farm       

Plant Size                            1 MW                 2.4 MW         
Capacity Factor                        60%                   25%          
Capital Cost                        2000 $/Kw          See table below    
Thermal Efficiency                     35%                                
Coal Carbon Content                 1 tC/toe                              
Discount Rate                          10%                   10%          
Lifetime                            20 years              15 years        
Fuel Cost                           35 $/tce                              

Table 2 illustrates the sensitivity of the incremental cost per tonne of carbon to changes in the capital cost of the wind power plant. For this illustration we increase the capital cost of the wind machine from $800 to $1300 per kW which is well within the normal variability of capital costs for different types of machines and wind sites (Cavallo, Hock and Smith, 1993). These are compared with a coal fired plant whose capital cost is assumed to be $2000/kW. The annualized cost is shown in $/yr and not $/kWh since a 2.4 MW wind farm displaces a 1 MW coal plant. The incremental cost is the difference between columns (3) and (4) divided by the annual carbon savings.

Table 2.Incremental cost.

 Capital Cost             Annual Cost                  Incremental Cost     
    ($/kW)                   ($/yr)                          ($/tC)         

     Wind        Coal        Wind          Coal                             
     (1)          (2)         (3)          (4)                              

     800         2000         257          296               -32            
     900         2000         289          296                -6            
     1000        2000         321          296                20            
     1100        2000         353          296                47            
     1200        2000         385          296                73            
     1300        2000         417          296                99            

The incremental cost varies over a wide range from -$32/kW to as high as $99/kW. What the wide range illustrates is that within a range of reasonable assumptions regarding the capital cost of a mitigation option, it is possible to arrive at widely different incremental costs. Thus it may be premature to claim a single threshold as the basis for weeding out expensive technologies.

Why do incremental costs vary so much when the capital and annualized cost vary by only 62%? The primary reason quite simply is that the incremental cost reflects the difference in capital costs of the two power plants, which varies much more than the variation in the capital cost of the wind farm alone.

Data availability

Using energy more efficiently and switching to less-carbon-intensive fuels (such as renewable energy) are the primary options for reducing emissions from the energy sector. Data Systems for the collection of data on energy efficiency and renewable energy are often the weakest in most countries. Data on conventional sources of supply are gathered from fuel supply companies, but those on end-use have to be collected through customer surveys, which are not conducted regularly or on a uniform basis in most countries. Similarly, in the household sector, non-traditional wood use is the norm and documenting and collecting data on these is not an easy task.

This situation is further complicated because mitigation analysis requires that data for the energy, forests, agriculture, etc. sectors be matched with those on the socio-economic impact of a mitigation option. Estimating the cost of a mitigation option entails that data on both the technical performance of a technology or a program be matched with its costs and other socio-economic attributes. Such matching poses a problem since different systems are used to collect data which collate them by different categories. In addition, inter-disciplinary analytical skills are needed, since a mitigation analyst has to be familiar with both types of analyses.

Most collected data are limited to a particular sector. Energy data report on fuel use by type and fuel losses by technology and sometimes on the performance of the technologies used for fuel supply, transformation and use. But they rarely report on the number of vehicles, their distance traveled, the economic output of an industry, etc. Thus matching requires better data collection and consistent categorization aided by some judgement on part of the analyst. It is also incumbent on the analyst to supply adequate documentation for both sets of data so that the reader is in a position to judge the matching for himself.

In "bottomup" analysis, cost data are collected for a technology option of a nominal size and with characteristics typical of an average option. Because a single technology characterization is often used to represent a range of technologies with varying characteristics, the average cost is only indicative of that which might be incurred in actual applications.

Discounting carbon

Since the benefits of atmospheric stabilization are not known, the "bottomup" approaches rank options on the basis of their cost effectiveness for reducing carbon emissions. Most studies use $/tC (or CO2) as a measure of the costeffectiveness of restraining carbon emissions. Future costs of reducing emissions are discounted in estimating the unit cost, while future carbon emissions are not. By not discounting carbon, these studies assume that the value of carbon damages in the future will increase at the real rate of discount. Given our limited knowledge about how fast damages caused by carbon emissions might increase in the future, this is an appropriate assumption to make at this point in time. As better scientific data and information becomes available, this assumption will need to be revisited in the future.

Macro-economic implications

In addition to knowing the incremental costs of mitigation options, policy makers are interested in understanding the macro-economic impacts of mitigation options on GDP, capital and foreign exchange flows, etc. Mongia, Bhatia, Sathaye and Mongia, 1991, highlight the tradeoffs between capital investment and foreign exchange requirements, which are both scarce commodities in a developing country. Importing fuels (natural gas) as a substitute for coal will reduce carbon emissions and decrease capital investment but at a higher foreign exchange outflow. Using renewable will also achieve carbon emissions reduction but with higher capital investment. Thus a careful balance will need to be struck between capital and foreign exchange implications of each option.

A study for Egypt examined the impact of the cost of reducing nearterm (to 2002) carbon emissions, with energy conservation measures, on GDP, using a "topdown" model. The study shows that, contrary to an earlier analysis for Egypt, emissions reduction through energy conservation will lead to an increase in GDP.

Additional analyses of this type are needed since the flexibility of an economy to afford emissions reductions will vary. An earlier analysis has shown for instance that the longrun GDP impact of emissions reduction is lower in Brazil, because of its abundance of bio and hydro resources, than in India, where energy resources are scarce.

Energy prices

Assumptions regarding energy prices can affect the incremental cost of one scenario compared to another. For example, La Rovere, Legey, Miguez 1994 show that the mitigation cost becomes negative in 2025 although they are positive in 2005 in their low carbon scenario for Brazil. The reason is the high 2025 world oil price of $37 per barrel (in 1990 $) assumed in the analysis. Renewable energy options become costeffective at this price in 2025, leading to a negative cost relative to the baseline scenario.

Historical and future trends

Changes in historical trends, for example in energy intensity improvement, will play a critical role in determining future emissions. For example over the last ten years, China's energy/GDP elasticity has hovered around 0.5; a remarkable achievement in itself. Holding it at that level for another 50 years will require continued strong efforts at energy efficiency improvements. If this can be achieved, China's emissions may only double at a much higher standard of living (Wu et al., 1994).

Combining forest and energy sector mitigation analysis

Afforestation, forest protection and conservation and improved logging methods are three types of mitigation options whereby net carbon emissions can be reduced. If wood from renewable sources is used to displace other fossil fuels then the benefit of afforestation increases. Studies show that technical land availability is not a constraint for afforestation (IPCC, 1995). Implementing the three types of options will cost less than $10/tC to capture between 50 to 75% of the carbon that trees could sequester.

The forest sector mitigation costs are likely to be higher than those for the energy efficiency options which hold the promise of negative costs for mitigation options. But, the forest sector

mitigation costs are much lower than those reported for carbon taxes of the order of $100/tC (Rubin et al., 1992).

4 Summary

Several programs are ongoing to assist developing countries and countries in transition to meet their obligations to the FCCC. The US Country Studies program has cooperative agreements with 55 countries, of which 36 are conducting, or planning to conduct, mitigation assessments. Mitigation studies supported by the US EPA and UNEP point to several issues, incremental cost, inadequate data, choice of discount rates, oil price assumptions, importance of macro-economic impacts, linking historical and future trends and interconnecting different sectoral assessments, which deserve careful attention in a mitigation assessment. We highlight the importance of these issues and illustrate that the incremental cost is very sensitive to a realistic range of costs of a mitigation option. Analysts and decision makers should therefore be cautious about accepting point estimates of incremental costs, and should expect a wide band of uncertainty around sectoral estimates of incremental costs.


Cavallo A., Hock S. and Smith D. (1993). Wind Energy: Technology and Economics. Chapter 3 in Johansson T., Kelly H., Reddy A. and Williams R. Renewable Energy: Sources for Fuels and Electricity. Island Press, Washington DC.

Grubb M. and Meyer N. (1993). Wind Energy: Resources, Systems, and Regional Strategies. Chapter 3 in Johansson T., Kelly H., Reddy A. and Williams R. Renewable Energy: Sources for Fuels and Electricity. Island Press, Washington DC.

Inter-governmental Panel on Climate Change (IPCC) (1994). Chapter III.F. Establishment and management of forests for mitigation of CO2 emissions. Working Group II, IPCC 1995 Assessment. Second Order Draft. December.

La Rovere E., Legey L. and Miguez J. (1994). Alternative Energy Strategies for Abatement of Carbon Emissions in Brazil: A Cost-Benefit Analysis. Special Issue of Energy Policy, November.

Mongia N., Bhatia R., Sathaye J. and Mongia P. (1991). Cost of reducing CO2 emissions from India: Imperatives for international transfer of resources and technologies. Special Issue of Energy Policy, November, pp.978-986.

Rubin E., Cooper R., Frosch R., Lee T., Marland., Rosenfeld A. and Stine D. (1992). Realistic mitigation options for global warming. Science, Vol. 257:148-149, 261-266.

Sathaye J. (1992). Energy Demand and Carbon Emissions from Brazil, China and India. Presented to the EMF-12 Meeting in Washington, D.C., May.

Sathaye J. and Christensen J. (1994).( Guest Eds.) Methods in Mitigation Analysis. Special Issue of Energy Policy, November.

Wu Z., He J., Zhang A., Xu S., Zhang S. and Sathaye J. (1994). A Macro-Assessment of Technology Options for CO2 Mitigation in China's Energy System. Special Issue of Energy Policy, November.

Back to Contents