A country framework for analyzing climate change mitigation

Ken King, Senior Environmental Specialist

Global Environment Facility, The World Bank

1818 St., NW, Washington DC 20433, USA


Once the global community accepted the threat of global climate change as a serious one, many countries began to study its possible consequences and the scope for national action that could help mitigate it. Following the entry-into-force of the Framework Convention on Climate Change (FCCC, 1992), which sets out country commitments to communicate information about proposed actions to mitigate climate change, many more countries expect to begin studies and will seek donor support for them.

Governments will find that national climate change studies are useful quite apart from helping them to fulfill commitments under the FCCC. All countries, developed and developing, will benefit from an understanding of the sacrifice they need to make for the greater benefit of climate change mitigation. This sacrifice - that is, the set of trade-offs or the "added burden" is what economists refer to as the "incremental cost" of climate change mitigation actions. All countries will benefit from knowing what these costs are and from knowing how much their national action will matter. Using measures of cost and mitigation, they will be able to choose from among their alternative national actions those with the greatest effect and the lowest incremental cost. Eligible countries can also use these measures to negotiate the amount of financial assistance they need from the financial mechanism of the FCCC or other sources.

A country framework for medium-term considerations will yield results useful for investment planning, policy choice, and financial reimbursement (where appropriate). This is the focus of the framework outlined in this paper. This framework does not explicitly cover longer term and global level assessments, although one could use an essentially similar framework for those as well. This paper outlines a framework, not a methodology, and as such it is compatible with many different specific methods. The reason for preserving this flexibility is that there are many methods, reflecting somewhat different traditions in modeling and country expertise, and each of these may have its own field of appropriate use. These methods are also evolving rapidly. It is therefore useful to identify, in the midst of these varied circumstances and rapid evolution, which particular general features are essential, or at least highly desirable, and to state these explicitly.

Section 1 is an introduction to climate change studies. Section 2 sets out a framework for analyzing the issues systematically at the country level. Section 3 describes some additional features, which are very desirable and which could be added wherever possible. There are, however, few specific features that the FCCC requires for the communication of country information under Article 12. So the paper makes no suggestion that any of these features described, or even the framework itself, satisfies a formal or legal requirement. It sets them out as tools that countries may wish to use to sharpen their analyses.

1 Country studies

1.1 Background

Once the global community accepted the threat of global climate change as a serious one, many countries began to study its possible consequences and the scope for national action that could help mitigate it. These studies had various objectives. Some aimed to provide general background, some to elucidate policy and investment options, and others to estimate the costs of proposed mitigation and adaptation efforts.

Several governments and some multilateral organizations have already financed such studies in developing countries and economies in transition. Following the entry-into-force of the Framework Convention on Climate Change (FCCC, 1992), which sets out country commitments to communicate information about proposed actions to mitigate climate change, many more countries expect to begin studies and will seek donor support for them.

Governments will find that national climate change studies are useful quite apart from helping them to fulfill commitments under the FCCC. All countries, developed and developing, will benefit from an understanding of the sacrifice they need to make for the greater benefit of climate change mitigation. This sacrifice - that is, the set of trade-offs or the "added burden" - is what economists refer to as the "incremental cost" of climate change mitigation actions. All countries will benefit from knowing what these costs are and from knowing how much their national action will matter. Using measures of cost and mitigation, they will be able to choose from among their alternative national actions those with the greatest effect and the lowest incremental cost. Eligible countries can also use these measures to negotiate the amount of financial assistance they need from the financial mechanism of the FCCC or other sources.

The United Nations Environment Programme (UNEP) started a process of consultation on the economics of climate change in 1990. UNEP held two workshops in 1991 that set out the main issues of methodology and the factors that analysts need to address for their studies to gain broad acceptance (UNEP, 1992a). UNEP set up a project, carried out by the UNEP Collaborating Centre on Energy and Environment (UCCEE) in Denmark, to prepare a methodology for costing GHG mitigation. UCCEE has since published reports on its costing studies (UNEP, 1992b) and on the evolving methodology (UNEP, 1993). UCCEE uses an iterative, pragmatic approach broad enough to encompass many different costing methods while providing room for methodological growth through a process of learning by doing and learning from examples of best practice. This approach has kept the studies clearly focused on possible investment and financing decisions and on the development of a national skills base. Although the studies suggest what improvements could be useful in future iterations, this pragmatic approach has ensured that the studies outstrip neither the data nor the analytical tools available at any given time.

The Global Environment Facility (GEF), as the interim operating entity of the financial mechanism of the FCCC, will finance country studies. GEF need not, however, be the only source of funds for such studies. Because UNEP is one of the three Implementing Agencies of the GEF, some of these funds will flow through UCCEE, which will continue to play a major role in coordinating future studies. GEF will also finance the agreed incremental costs of measures undertaken by eligible countries to mitigate climate change (GEF, 1994). Although it is not a requirement of the FCCC, some countries might therefore wish to use their country studies to set investment priorities and estimate incremental costs. GEF will work closely with UCCEE to study how one should do this. The foundation for this work would be the evolving methodology developed through the efforts of UCCEE. The specifically incremental cost concepts would be adapted from other work forming part of a wider GEF exercise known as the Program for Measuring Incremental Costs for the Environment. See King (1994) for a description of the issues being addressed in this program.

Diversity of approach has been and continues to be extremely valuable, particularly in a field with as many conceptual subtleties and uncertainties as climate change mitigation. The challenge must be to coordinate approaches only to the extent that one can eliminate wasteful duplication and avoid needless divergences in assumptions and terminology. [To begin laying the basis for such an assessment, the Program for Measuring Incremental Costs for the Environment commissioned a survey of the country studies complete or on-going in 1993 (Fuglestvedt et al., 1994)]. Beyond that, it is worth preserving as much diversity of approach as possible. This paper draws attention to the features that would be either essential or desirable in any analysis of climate change mitigation. However, it is general enough that one could use any of several different methods within it.

There is also a diversity of funding sources; GEF forms only a small part of the effort so far. Developed countries of course finance their own studies and, on a bilateral basis, those of developing countries as well. Probably the largest bilateral program is the $ 25 million US Country Studies Initiative (U.S. Department of State, 1993), although it is by no means the only one. The Asian Development Bank has also executed a program of studies in Asia. We hope that UCCEE, the bilaterals, and other country study programs will continue to share experiences of best practice and that the desiderata mentioned in this report will also be useful to others.

1.2 Types of country studies

As mentioned above, there are many different planning and modeling approaches. Studies of mitigation also vary in purpose, and this will also affect the choice of approach. Some designed their studies to provide background information or to build institutional capacity rather than to estimate the costs and impacts of mitigation measures.

System boundaries

Perhaps the clearest difference among studies is the "system boundary," or the extent of the effects analyzed.

Global level. Studies at the global level are useful for estimating

- estimating the aggregate impact of measures on atmospheric concentrations of GHGs;

- estimating the overall amount of finance that needs to be mobilized in a given period to pay for these measures;

- revising research strategies;

- determining the relative priority of mitigation and adaptation;

- proposing international burden sharing.

Such studies may have been useful for advising participants in the Intergovernmental Negotiating Committee of the FCCC. The IPCC, which is now preparing its Second Assessment Report (due in 1995), will also draw on such models. These will help IPCC take a view about the adequacy of commitments in the FCCC (among other matters).

Country level. Studies at the country level usually identify the mitigation and adaptation measures available to governments, along with their priority, likely effects, and costs. The studies help governments select measures and negotiate financial assistance. In most cases, one takes as given in country studies certain matters which global studies treat as issues to be addressed.

Sector level. Sometimes it is too difficult to integrate studies macroeconomically, but one can study sector investment plans and sector policy responses. The paradigmatic case is the energy sector (the focus of much of UCCEE's Phase I effort).

This paper focuses on country studies rather than global studies, even though some of the observations will also apply at that level. Although one could attempt country studies at the macroeconomic level, if this proved infeasible one could concentrate on priority sectors (such as energy).

Time frames

Studies with different objectives may have different time scales too. Rather than give precise definitions of the time scales, it is better to refer to these different objectives:

Long term. Uncertainties about technical change affect all long-run models. The large discrepancy between the costs and impacts of different long-run models is often due to different (yet still reasonable) assumptions about technical progress. (See for example IIASA, 1993.) Such long-term studies may be able to inform us whether we are able to avert climate change. They might also provide valuable inputs to medium term studies about (i) the value of information to reduce scientific and economic uncertainty; (ii) the likely magnitude of future costs; and (iii) the future benefits of research, organizational learning and economies of scale. "Long term" thus typically refers to a period in excess of 15 years, since it often takes this long to commercialize unanticipated breakthroughs.

Medium term. In medium term studies, costs are given but investment choices are still open. The investment cycle is several years, so "medium term" typically refers to periods of five to ten years.

Short term. In the short term, one assumes that investments are also fixed but that some adjustment can take place in response to price changes. "Short term" typically refers to periods of a year or two.

Country studies set out investment choices, so they would be classified as "medium term." Long term studies are less useful at the country level because the parameters that affect the long term have global dimensions. Technical progress and the availability and price of fossil fuels and their substitutes are prime examples of such parameters. Of course, outputs of long term studies would be necessary as inputs to (that is, as assumptions in the scenarios of) country studies.


There are many methods and modeling techniques, the most important dichotomy being between the "top-down" and "bottom-up" approaches. To date, possibly the most operationally fruitful approach for country studies has probably been the bottom-up approach, for several reasons. One is that it provides a wealth of detail on medium term options (specific investments). A second reason is that it does not assume that sectors are currently operating efficiently, and therefore helps to focus attention on such things as market barriers, transactions costs, and the "negative cost options." There are also two reasons that apply more to developing countries. The first is that it is simply easier to negotiate, agree, and monitor the "engineering" costs of mitigation measures and therefore easier to obtain financing for the incremental costs incurred. The macroeconomic costs in top-down models would be so contentious that there is little likelihood of their being used for purposes of resource transfers in the near future. The second is the common feeling that developing countries, in course of mitigating a global problem not principally of their making, should not have to change their development goals. This is easier to show if the output mix under mitigation is as close as possible to that under no-mitigation.

Consider, for example, the case where a country proposes to substitute new and renewable energy for fossil fuel energy sources. It would be relatively easy to propose such a package of mitigation measures if it maintained energy services at the same level as the country would enjoy in the absence of the package. If the country were reimbursed for the incremental costs of these measures, this financing would effectively subsidize the costs of energy to the point where the cost of production was the same. The pre-existing price of energy services could be maintained and neither the consumption of energy services nor any other aspect of the economy should be affected. It would be much harder to propose a more efficient set of measures based on a top-down approach. In this approach the country would propose to adjust economically to higher cost alternative energy sources. It would do so through the pricing mechanism, thereby reducing consumption of energy and bearing any other attendant macroeconomic adjustments such as inflation and (possibly) increased unemployment. Even if the country were satisfied that it was adequately reimbursed for GDP losses, this approach would be more difficult for its citizens to accept and for the government to administer effectively.

1.3 Country framework

Need for a framework

All signatories to the FCCC, developed and developing, are contemplating national actions they can undertake to mitigate global change. When they do so they will confront a number of issues at the national level, including

Also, if FCCC accepted a national climate study, it would facilitate normal development work. This is because it would then not be necessary to view the GHG emissions of any individual project in isolation. It would not be reasonable to condemn an individual project as "bad" for emitting some GHG or to deem it "good" if it does not. If a project forms part of an overall least-cost plan (meeting both development and GHG reduction goals) then it should be acceptable in terms of climate change effects. Many developing countries will have baselines of development activity that are rising over time and, in the medium term, rising levels of GHG emissions too. A well thought-out country study with a well-described baseline will show the extent to which this baseline rise is necessary, and will protect countries from arbitrary criticism that their projects emit GHGs.

Focus of a country framework

A country framework for these medium-term considerations should yield results useful for investment planning, policy choice, and financial reimbursement where appropriate. This is the focus of the framework outlined in this paper. This framework does not explicitly cover longer term and global level assessments, although one could use an essentially similar framework for those as well. This paper outlines a framework, not a methodology. It is compatible with many different specific methods. The reason for preserving this flexibility is that there are many methods, reflecting somewhat different traditions in modeling and country expertise, and each of these may have its own field of appropriate use. These methods are also evolving rapidly. Yet it is useful to identify, in the midst of these varied circumstances and rapid evolution, which particular general features are essential, or at least highly desirable, and to state these explicitly.

Section 2 of this paper sets out the essential aspects of country framework for analyzing these issues systematically. Section 3 describes some additional features, which are very desirable even though not strictly essential. There are, however, few specific aspects that the FCCC requires for the communication of country information under Article 12. So this paper makes no suggestion that any of these aspects or even the framework itself satisfies a formal or legal requirement. The paper sets them out as tools that countries may wish to use to sharpen their analyses.

The examples are cited only as "best practice." I have not cited specific "bad examples," although there are many. There are three reasons for this. One is that this paper is not a critique or a description of existing studies (such as that of Fuglestvedt et al., 1994): it is a normative procedure. Another is that country study approaches have been evolving so rapidly that more recent studies already reflect several lessons from previous "bad examples." The third is that studies have had different objectives (background description, capacity-building, abatement costing, etc.) and one should not judge the method of one by the standards of another.

2 Comparative analysis

Comparative analysis is the framework for climate change mitigation studies. This framework captures the essence of decision-making that influences the emission of GHGs and the costs of those decisions.

2.1 Comparisons

Decisions make a difference. An action causes a change in the future state of affairs from that which an alternative action would have caused (or from that which have been the case in the absence of any action). The action results in additional (incremental) benefits, usually the achievement of what one set out to do, and additional (incremental) costs.

A country with alternative courses of action could therefore analyze them in a comparative framework. A strategy, or course of action, is a group of actions that are consistent with one another, but alternative strategies are mutually exclusive. It is this exclusivity that makes decision necessary. It is the existence of more than one possible strategy that makes the comparative analysis of outcomes essential. That is, it is necessary to see how much better one course of action is than another, or how much more it costs than another. This is comparative analysis. It can be applied to any differences, such as

shifts in strategy;

increments in cost or losses of opportunity; and

reductions in GHG emissions or additions to benefits.

To maintain consistency, one should measure the shifts, increments, and reductions under the same conditions, described below, of baseline, alternative, and scenario.

2.2 Baseline strategies

In the case of climate change mitigation, a country may have many alternative courses of action. One obvious strategy is not to do anything specifically for mitigation but to simply optimize economic behavior domestically. Let us call this strategy a baseline. The baseline will lead to a particular emission of GHGs (GB), will cost a particular amount (CB), and will yield a particular benefit to the country in terms of goods and services (BB). The baseline reflects national priorities in the absence of any special effort to implement the FCCC. Given the financial constraints at the time, questions of equitable distribution, and whatever other internal constraints exist, we would not normally expect any other course of action to be able to provide the same BB yet cost less than CB. For if it did, the country would have made it the baseline strategy instead.

Some country studies have avoided the use of explicit baselines, with mixed results. In some early studies, the authors took the costs of mitigation to be the aggregate cost of the technology needed to supply energy without using fossil fuels. This is fallacious, and the explicit consideration of a baseline would have shown it to be so. The baseline would have included costs, including both capital and fuel costs, to produce the required energy using fossil fuels. The real cost of mitigation in this strategy is the incremental cost. This is the difference between the cost of non-fossil fuel energy supply and baseline supply, and not the entire cost. The fallacy of assuming that the cost of mitigation is the total cost is that avoided costs (baseline costs not incurred as a result of the strategy adopted) are ignored.

Other studies have aggregated the costs deemed to be incremental without explicitly considering a baseline. For example, the incremental costs of each technological substitute are added together to obtain a country incremental cost. While obviously an improvement over the above, this process may not be internally consistent. It is as if each component carried its own little baseline with it, but these may not in aggregate constitute a coherent baseline strategy. This is particularly evident where there are system effects in energy supply.

2.3 Alternative strategies

However, there may be other more desirable courses of action if the objectives change. If the country would like to mitigate climate change, for example, it could consider an alternative strategy. In order to make a difference, this alternative strategy should lead to a lower level of GHG emissions than the baseline strategy (otherwise there is no point considering it). Note that the judgment is about the difference, not the absolute amount. Even if the total amount of GHG emissions is rising in both the baseline and the alternative, the alternative could still be efficacious. The measure of the effectiveness of the strategy is the amount by which the emissions in the alternative (GA) are less than those in the baseline (GB). Likewise, the cost of reducing the emissions is the difference in cost measured between the same two strategies (CA CB). This is the reason one usually refers to the cost as the incremental cost. If the domestic benefits are similar (BA BB) then one would expect the incremental cost to be positive, that is, the alternative to cost more.

2.4 Increments

Comparison matrix

A simple matrix shows the comparative nature of the analysis.

Table 1.Comparison matrix.

                    Cost         Domestic benefit        Global benefit    
                ($ million)                           (avoided emissions)  

Alternative          CA                 BA                    - GA         
                     CB                 BB                    - GB         
strategy        (CA - CB) >             0                (GB - GA)  > 0    
                     0        when benefits are held                       
Increment                            constant                              

We are searching for mitigation alternatives so, by hypothesis, the global benefit would be positive; that is, the change in emissions would be negative, a reduction. Since the country prefers the alternative less than the baseline, either the benefits will be less or the costs will be greater. It is common to suppose that countries will not undertake alternatives that require them to sacrifice economic development. In this case we can further assume that the domestic benefits in the alternative are not less than in the baseline and therefore that the incremental cost is positive.

Other baseline strategies

The baseline is a counterfactual situation when the country follows an alternative strategy instead. This raises the question about which baseline is the real one, or even whether there can be more than one. Only internally consistent descriptions of possible courses of events can be strategies, so any coherent strategy could be baseline. Where multiple strategies are possible, the analyst could regard any of them as the baseline for comparing any other. The choice of appropriate baseline is a matter for policy. For financing incremental costs, one is usually interested in a baseline comprising what a country could have done in its own interests. However, one could conceive of a sub-optimal "business-as-usual strategy," an economic strategy, and a mitigation strategy in order to describe incremental costs between all sets of alternatives. Table 2 shows what this implies:

Table 2.Comparisons using different baseline strategies.

                                Cost         Domestic     Global Benefit   
                            ($ million)      Benefit         (avoided      
                                                          million tonnes   

Alternative strategy             12             B               -2         
Baseline strategy                8              B               -4         
(economic)                       4              0               +2         
 Increment (alt.-econ.)          10             B               -5         
Baseline strategy                -2             0               +1         
(business-as-usual)              +2             0               +3         
 Increment (econ.-BAU)                                                     
 Increment (alt.-BAU)                                                      

Let us suppose that the three strategies yield essentially similar benefits in terms of energy services, B. The business-as-usual (BAU) strategy is inefficient, and releases a lot of GHG. The economic (Econ.) strategy reflects the optimal use of energy conservation, which reduces costs to $ 8 million and reduces GHG emissions from 5 units to 4 (because less fossil fuel is needed for the same energy service). Suppose that the alternative (Alt.) strategy costs more ($ 12 million) because it uses some advanced renewable energy technologies, but emits much less GHG (2 units). The importance of baseline is now clear. The incremental cost of moving from an inefficient baseline to the alternative is less than that of moving from an efficient baseline. The decomposition of costs above shows that the incremental cost of $ 2 million in moving from BAU to Alt. is actually $ 4 million, partly financed from economic savings in moving (notionally, at least) from BAU to Econ. If the incremental cost is to be a measure of the loss of economic opportunity, then the figure of $ 4 million is the appropriate one. If international resource transfers are based on incremental cost, then it is in the country's own interest in situations like the above example to use an economically efficient baseline. The $ 2 million saving in moving to the efficient baseline is a saving that the country could have appropriated anyway. The use of the BAU effectively uses these savings to cofinance the global environment benefit.

Other alternative strategies

One could use other alternative strategies to provide either different global benefits (in which case one would weigh the incremental cost against those incremental benefits) or additional domestic benefits.

Consider the case in Table 3 where there are two independent, mutually exclusive, alternative strategies for reducing GHGs. Supposing one regards as cost-effective all those mitigation actions that cost less than $ 2.50 per tonne of CO2 avoided. (This might be because of the availability of options below this ceiling in other countries.) Which alternative should one adopt? At first sight, both appear to fall within the limit. The unit incremental cost of Alt-2 with respect to the baseline is $ 2 / t and that of Alt-1 with respect to the baseline is $ 1 / t. But notice that the unit incremental cost of Alt-2 with respect to Alt-1 is $ 4 / t. The reason for this is that the increments between discrete strategies are averages, whereas the figure required for comparison to a cost-effectiveness criterion is the marginal incremental cost. Thus one should decompose the strategies as far as possible so that one can estimate the cost of each intermediate increment and thereby choose the optimal strategy.

Table 3.Comparisons using different alternative strategies.

                                 Cost       Domestic      Global benefit   
                             ($ million)     benefit         (avoided      
                                                          million tonnes   

Alternative strategy - 2          12            B               -2         
Alternative strategy - 1           8            B               -3         
Baseline strategy                 6             B               -5         
 Increment (Alt-2 - Alt-1)        4             0               +1         
 Increment (Alt-2 - base)         6             0               +3         
 Increment (Alt-1 - base)         2             0               +2         

Consider now another type of situation where other alternatives are important (Table 4). In this case, the first alternative generates global environmental benefits (reduced emissions) whereas the second alternative generates the same global environmental benefits and some additional domestic benefits. There may be several strategies that yield the global benefit but have different domestic implications. The simplest illustration is at the project level, but the same principle could apply at the country level. Suppose that a run-of-river hydro plant can substitute for a diesel generator. It supplies the same electricity but reduces greatly the GHGs. The incremental cost of $ 5 million is, let us suppose, the least incremental cost obtainable. Suppose also that a hydro dam could supply the electricity and eliminate the GHGs at an incremental cost of $ 6 million. This would not be least incremental cost but may be interesting from the country perspective because for the additional $ 1 million the hydro dam can also supply irrigation water (a domestic benefit only).

Table 4.Comparisons using a second alternative which generates additional domestic benefits only.

                               Cost         Domestic     Global benefit   
                           ($ million)      benefit         (avoided      
                                                         million tonnes   

Alternative strategy - 2        10        Electricity          -2         
Alternative strategy - 1        9              +                          
                                           Irrigation          -2         
Baseline strategy               4             (I)              -5         
 Increment (Alt-2 -             1              E                0         
Alt-1)                          6              E               +3         
 Increment (Alt-2 -             5              I               +3         
base)                                          I                          
 Increment (Alt-1 -                            0                          

2.5 Scenarios

The analysis of differences above is quite general. It could apply at the project level (for mitigation projects see Ahuja, 1994), the sector level (for the power sector see King, 1994b) or the country level. In all cases though, it is necessary to define which things the decision affects and which it does not. The separation is known as the system boundary. The things that the decision does not affect comprise the scenario, the background against which the drama of the decision and its consequences is played. For example, the level of world oil prices is almost always part of the scenario for small countries. Their use of oil is unlikely to have a major effect on oil supply or demand. Oil prices would, however, be an outcome for decisions taken by large OPEC producers or by OECD countries collectively.

Note that, like the outcomes, the scenario will describe future events and to that extent will be uncertain. One could consider alternative scenarios in order to test the importance of assumptions for the outcome. For example, one could assume alternative oil prices and note their effects on the overall costs.

2.6 Cases

It is possible to consider various baselines, various strategies, and various scenarios. To avoid confusion, it is necessary to have a clear terminology and to keep track of the effects of alternative assumptions within the comparative framework. (Unfortunately, this is not always done and the effects of a scenario change are sometimes merged with those of a change of strategy or a shift in baseline.)


One can use different scenarios to test the sensitivity of the results to uncertainties. One could model the effect on the baseline strategy of different oil prices, weather patterns, or global responses to climate change. Likewise one could model the effect of these uncertain parameters on the outcome of the proposed alternative strategy. The effect on the increment, of course, requires that the same scenario be applied to both the baseline and alternative strategies. Note that the difference between a baseline in one scenario and an alternative in another scenario is meaningless as it mixes two different influences. The difference in cost, in such a confused calculation, would be the sum of the costs imposed by events not under the control of the decision-maker and the incremental cost resulting from a specific decision.

Typically, a scenario will include the assumptions on

Usually, one defines the scenarios only on the basis of the principal country-specific uncertainty. This is the growth in demand for the relevant services (usually energy), often linked to the rate of economic growth.

High growth scenario

Reference growth scenario

Low growth scenario

The other factors are international rather than country-specific. One could use them to define scenarios and standardize assumptions with other groups working on similar questions. Usually, one seeks the opinions of technical experts about technical availability, and tests alternative cost assumptions in sensitivity testing around the reference scenario. Ahmed (1994) made estimates for some generic costs of renewable energy technologies.


One could use different baselines to test different assumptions about the outcome foregone as a result of undertaking the alternative. That is, because the baseline will be counterfactual once the country has implemented an alternative, it is possible to consider various baselines. It will be useful to make the underlying assumptions explicit, since some baselines may be more plausible or have more policy relevance than others. For example, if the financial policy requires it, the baseline may need to incorporate any of the following: (i) the most economic way of achieving the identified domestic benefits, (ii) consistency with national priorities, and (iii) reasonable environmental protection standards. Alternatively, it may simply be the most likely outcome, including any inefficiencies, distortions, or non-economic considerations that the country is likely to have permitted.

One generally uses one of two baselines. One is the "business-as-usual" baseline, which represents the continuation of existing policy. The other is the "optimal" baseline, which includes assumptions about whatever pricing policy reform, industrial restructuring, and environmental amelioration would be in the county's own best interests. The impact of this choice could be substantial, because many economically attractive but unfinanced opportunities could also benefit the global environment (such as energy conservation, in relation to the global benefit of reduced GHG emissions).

Many developed countries are working towards targets expressed in terms of base year emissions. They seek to limit emissions by a certain target date (such as the year 2000) to an amount equal to emissions in a base year (such as the year 1990). They also seek to reduce future emissions reduced by a fixed percentage of these (20 per cent has been suggested, perhaps by the year 2020). This does not obviate the need to evaluate alternative strategies in a comparative framework; the baseline strategy may not be able to meet the target while the alternative should. The cost of the alternative is still the incremental cost of the alternative relative to the baseline. In any case, developing countries have no specific commitments under the FCCC. It is likely that they will measure their GHG emission reductions against normal trends rather than base year emissions in order to provide for continued industrialization to increase living standards. That is, the baseline would include "normal" growth in emissions over time.

Policy responses will be very important in defining the baseline. For one thing, efforts to control the release of GHGs will span many decades and may even be indefinite. Long time horizons make it more likely that baselines will (eventually) incorporate economic policy reforms and industrial restructuring. For another, the costs of GHG reduction strategies are likely to be very high and inefficiencies will not be affordable.

Alternative strategies

One would normally evaluate different strategies reflecting (i) the severity of the GHG emission constraint, (ii) the degree of economic efficiency of the response, or (iii) additional domestic considerations.

In theory, there could be a continuum of alternatives. In practice, one would use two or three GHG constraints to estimate the incremental cost parametrically. For example, a 25 per cent reduction may imply an average incremental cost of $ 20 a tonne of CO2, whereas a 10 per cent reduction might imply an average incremental cost of $ 5 a tonne, and a 1 per cent reduction only 50 cents a tonne. There are various ways of constructing alternative strategies like this. One would be simply specifying the substitute technologies or actions. Others would be imposing a GHG emission constraint and optimizing the investment plan or using a non-zero accounting price for GHG and optimizing the investment plan.

The usual alternative considered will, like the baseline, be economically optimal. That is, it will represent the least-cost way of obtaining the emission reduction.

Number of cases

There are thus many possible cases one can evaluate. The number depends on how many scenarios S (reflecting external uncertainties) one needs to consider; how many baselines B are of interest to policy-makers; and how many alternatives A (strategies) are under consideration. The maximum number of cases C is multiplicative, i.e.,

C = S x B x A

although the actual number of cases that one needs to consider may be smaller if certain combinations are unlikely or unnecessary.

The number of alternatives itself depends on the number of emission reduction constraints (ERC) and the policies (P) under consideration:

A = ERC x P

There is no prescription for the number that one should consider - this will be a matter of judgment in each case.

2.7 Using the framework

Shifts in strategy

One needs to see the outcomes of actions to mitigate climate change as differences between baseline actions and the alternative actions. Some examples:

The imposition of a carbon tax should be seen against the background of baseline taxes or subsidies. Also, if the government offsets the carbon tax by reducing taxes elsewhere (through a commitment, say, to revenue neutrality) then the analyst must draw the system boundary widely enough to include these effects.

Energy conservation activities will reduce average GHG emission relative to a base year when little attention was paid to energy conservation. But energy conservation does not represent a shift in strategy unless one could reasonably project a baseline where the economy does not adopt energy conservation to the same extent as in the alternative proposed. In other words, an increase in energy conservation may be part of a reasonable baseline rather than a special shift in strategy. Confusion is possible because the word "increase" is ambiguous: it can refer to change over time or to the policy-induced shift at any given time between baseline and alternative action. It is necessary to be clear which meaning one intends. A thorough application of the comparative framework will help to maintain this distinction.

Likewise, economic reform, structural change, technological change, and the uptake of previously unrealized opportunities for cost-effective energy conservation and substitution may be part of the baseline rather than part of an alternative strategy. The longer the time period, the more likely that economic reforms become part of the baseline assumption. The relevant shift in strategy is the difference between a specific mitigation policy (alternative) and the consequence of "normal" economic forces (baseline). It is not the mere change in policy and investment over time.

Increments in cost

The incremental cost calculation is of particular interest because it is the reference point for determining financial compensation for eligible countries from the financial mechanism of the FCCC. Even non-eligible countries will be interested in their incremental costs because they may have alternative strategies that yield the same global benefit.

There are some practical issues.

Sensitivities. The estimated incremental cost, being (as we have seen) a difference between two projected costs, is sensitive to the various assumptions made. Alternative, but still reasonable, assumptions about baseline actions and costs of key inputs (such as oil and gas) can have large effects on the estimated incremental cost. Nevertheless, despite the difficulties, asking for the incremental cost is posing the right question. This is because the incremental cost is the cost of the strategy and therefore a measure of the sacrifice made for a global benefit.

Counterfactuals. The baseline, by the very fact that the country implements the alternative, is counterfactual. This means that one cannot directly observe the baseline, unlike the alternative strategy, which is actual. One can, however, monitor certain key variables. These key variables include the price of oil and capital items and monitoring them will narrow the range of baseline cost estimates.

Uncertainties. Future costs, in both the baseline and the alternative, will be uncertain to some extent. Hence, ex ante agreement on incremental cost imposes risks. The recipient risks being undercompensated if a financial transfer for estimated incremental costs turns out to be too small. This could happen if baseline costs turn out to be smaller (due, say to a lower than expected international coal price) or if the alternative costs turn out to be higher (due, say, to unexpected difficulties in commissioning advanced renewable energy technologies). Conversely, the financial mechanism risks overcompensating if matters turned out otherwise. One can address this issue by continuously re-appraising incremental costs and making transfers annually.

Incentives. The estimation matters to all parties - the recipient country, the financial mechanism, and the financier of the non-incremental costs. Therefore each will have an interest in ensuring that it suffers no disadvantage from the technical uncertainties. This means that one should emphasize a transparent and rigorous framework for technical negotiations on incremental cost. The framework of comparative analysis presented here would help to structure such negotiations, which might otherwise depend more on the forcefulness of the interlocutors and on vague understandings of the lost opportunities being compensated. The framework would need to be supplemented by a number of additional tools, such as "tests of reasonableness" drawn from broader sector work and internationally accessible data.

Reductions in emissions

The efficacy on national actions on GHG emissions is measured by the reduction in emissions compared to the baseline emissions, not by reductions compared to earlier levels of emission. One should not make the mistake, possibly encouraged by the adoption of targets expressed in terms of base years, of attributing to the policy shift only those reductions that take place in total emissions over time. Consider a situation of rapid growth in GHG emissions for development purposes. A reduction relative to the high expected baseline growth is still a positive achievement, even if total net emissions still rise. Similarly, a slight reduction in GHGs over time where baseline trends showed a decline anyway may not be an achievement at all.

2.8 Some conclusions and observations

Comparative analysis. The essence of planning requires a comparative analysis of the impacts and costs to highlight changes, shifts, and differences. In principle, one could do this at any level - macroeconomic, sectoral, or project. However, not every analyst applies comparative analysis or makes its use explicit.

The relevant comparisons. Changes highlighted by a comparative analysis are those that result from a specific mitigation policy. They are not the changes that may occur over time, and should not be confused with them. In particular, the outcome attributable to a shift in policy is the difference between the outcome with the new policy (alternative) and the outcome without it (baseline) for any given time. It is not the difference in outcome at one time and some historical outcome (base year outcome). Nor is it the difference between the outcomes of two different countries. Comparisons with other times and other countries may be useful, but only to the extent they provide proxies for the relevant baseline.

Consistent comparisons. One could use various scenarios, alternatives, and baselines. However, for consistency, the same assumptions should underly any given case or comparison. For example, one should use the same baseline for measuring policy shifts, emission reductions, and incremental costs. One should also use the same scenario for the baseline and the alternative.

3 Desirable additions to the framework

The previous section described the essence of mitigation analysis: comparative analysis. This section describes some other features one could strive for in order to make the analysis more useful. One could add these features progressively to a country framework.

3.1 Net analysis

Sources and sinks

Because it is accumulation of GHGs in the atmosphere that is the cause of the risk of climate change, it is net emissions (emissions less reabsorption) that is important. To the extent that policy influences reabsorption, a net analysis will be useful.

This is different from the analysis of other atmospheric pollutants, such as sulfur dioxide. In that case it is deposition that is damaging, and this in turn is controlled at source. Emissions and deposition do not cancel and a net analysis has no meaning.

IPCC guidelines for GHG inventory analysis require consideration of both sources and sinks. Because the uncertainties about sink data are considerably higher than those about source data (especially energy-related sources), it is useful to maintain separate tallies for emissions and absorptions. Nevertheless, it is net emissions that are ultimately important.

Discounted emissions

Another type of issue is the netting of emissions and absorptions that occur at different times. This issue has not been fully resolved. In the case of sulfur dioxide, for example, one can discount the future damage of acid precipitation in the usual way. But in the case of CO2 this is not possible because one cannot use the physical emissions as a proxy for the damage. The damage function depends non-linearly on accumulation and possibly rate of accumulation. One can infer discount rates from different time horizons for the accumulation target. IPCC is still discussing the issue.

Although not completely resolved, this is not an operationally urgent issue. While it is necessary to develop good estimates of incremental cost (since reimbursements will depend on these estimates), one can express the global benefit as a time profile of net emission reductions. The only occasion where it may make a difference is where one is finely discriminating between alternative actions on the basis of a cost-effectiveness criterion. However the financial mechanism will judge interventions on multiple criteria, including the value of the information obtainable, the available finance will not permit it to operate even close to the margin. Therefore such discriminations are largely academic. When in doubt, use sensitivity analysis.

Given the uncertainty, it is inadvisable to aggregate emission reductions into a single "present value" or to add emissions of different gases according to their global warming potential (which would imply a discounting). Separate accounts for each gas, expressed as a time profile, would be more useful.

3.2 System analysis

The analyst will need to judge the appropriate system boundary for the calculations. It may be desirable to analyze the whole economy as an integrated system, with the change in GDP as the incremental cost. However this may be too ambitious, and an aggregation of sectoral and project analyses may be all that is initially achievable.

Integrated investment plans

In some sectors, the costs and emissions of individual projects are interdependent. This means one cannot analyze them separately and then aggregate. In the power sector, for example, there are numerous systems effects and an integrated sector investment plan is the simplest system that can be analyzed. In other sectors, the projects may be more independent and it will be a matter of judgment what level of analysis one requires.

Fuel cycles

Another kind of system that is important is the fuel cycle, because actions in one part of the cycle will influence the emissions of GHGs elsewhere. Therefore baselines and alternatives should include systems defined widely enough to include these ramifications. An example is the proposal to develop natural gas as a substitute for coal. The basis of this proposal is the fact that natural gas releases less CO2 per unit of useful energy provided. However the provision of natural gas has its own risks in the form of pipeline leaks. Because natural gas is a much more potent GHG than CO2, even small leaks could cancel the beneficial effect of coal to gas conversion.


Usually, mitigation will not affect adaptation at the country level because it is global trends that matter. While adaptation and mitigation need to be integrated at the global level, one can present them as separate activities at the country level. One exception is the protection of low-lying coastal forests (carbon sinks) from sea level rise.

3.3 Economic analysis

Economic costs are the real resource costs of mitigation. Looked at from the outside, it is the economic costs that are important for the adjustment and which would be the basis for any international financing. At the country level, the costs are economic costs and the analyst nets out the transfer payments. Two practical problems may arise though.

Redistribution. Redistribution of the adjustment cost may not be practicable. One group of financial agents might bear financial incremental costs while another group obtains financial benefits.

Pricing policy. Existing pricing policies may make economically attractive GHG reduction options financially unattractive. In some developing countries, for example, electricity for rural water pumping is provided at well below cost. High efficiency electric pumps could reduce the amount of electricity needed and hence the emission of GHGs, but the additional cost of these pumps cannot be financially justified at the prevailing subsidized prices.

3.4 Integrated analysis

The system effects considered above arise from physical links between the activities. One needs system analysis to take these effects into account. Another type of link is economic, operating through changes in relative prices. Where these effects are important, one needs an integrated analysis.

One could use integrated country analysis to estimate the least cost path of climate change mitigation. Economywide adjustment to GHG constraints through the price mechanism is desirable because it is more efficient than sector-by-sector or project-by-project adjustment. This efficiency is the result of additional flexibility. Consider the case of energy suppliers that make the necessary changes to their systems. Their costs will rise (by the incremental cost). If, on the one hand, suppliers pass this cost on to consumers, they in turn will adjust optimally by investing in conservation and substitution and reducing their demand for energy services. If, on the other hand, the incremental costs are financed as and where they are incurred by the government or the financing mechanism of the FCCC, the costs, output, and prices would be left unchanged.

Fully integrated analysis of the adjustments in developing countries may be less important operationally in the short term because of the form international financing is likely to take. International financing of the incremental costs of eligible countries is more likely to take place at project and sector levels than at the country level for several reasons. One is that incremental costs will be easier to estimate at these levels. It seems very unlikely that the parties would ever agree on macroeconomic costs in view of the difficulty of defining and monitoring a baseline. A second reason is that adjustments of output will not be easy to agree on. This is because developing countries expect that their priorities and development opportunities would be left undisturbed by their commitments to climate change mitigation. Costs such as unemployment and missed macroeconomic targets will be more difficult to accept politically than reimbursed increases in investment costs.

3.5 Comprehensive analysis

Ideally, one should include all sectors and all gases. However, the methodologies applicable to some sectors are more sophisticated than others and one could make progress there without having to wait for similar analyses elsewhere.

3.6 Standardized analysis

Standard terms and assumptions make it easier to compare the results for one country with another. This would facilitate the choice of the lowest cost interventions first, irrespective of where they occur for financing under GEF or joint implementation. It would also permit aggregation of results and facilitate assessments of the adequacy of existing commitments under the FCCC.

Generic data

Some data items are international because they concern tradable goods such as oil, coal, and capital equipment. Standard assumptions on such items reduce the differences in cost estimates that are simply due to such assumptions, making the remaining cost differences more meaningful. Not all costs are generic though. The incremental costs always depend on the costs of the alternative and the baseline and so are country-specific. This is why incremental costs can seldom be standardized in an "indicative list," the way categories of eligible expenditure can.

Test ranges and assumptions

Although countries may have their own preferences for parameters like discount rates, it is useful to include one or two input values as standards that would be run as well. This would again facilitate comparisons. Standard assumptions could be made about:

Discount rates

Baseline strategy (economic, business-as-usual)

Alternative strategy (least incremental cost, severity of constraint on GHG, percentage reduction, shadow value)

Scenarios (world energy growth, availability of fossil fuels)

Sensitivity ranges


The various study authors have used terms in different ways. Although there is no single accepted standard, it is advisable to be explicit about the terms used to avoid confusion. Terms have caused confusion in two ways. One is the obvious way, in that terms used in different studies have different definitions and this makes misidentification more probable and comparison of country results more difficult. The second way is more subtle and dangerous. Terms not clearly fixed in meaning within the same study can inadvertently shift in their connotation and lead to error. This is particularly important in comparative analysis where there are many different elements that one needs to control for the analysis to make sense.

Here is a summary of terms used in this report and the concepts with which they are commonly misidentified:

Scenario. I have reserved the term "scenario" to apply only to the context of the problem, not the outcome or the strategy. The scenario embodies the relevant exogenous parameters that are not subject to policy control and which remain the same no matter which strategy one adopts. (Some analysts have used "scenario" to mean "strategy," and sometimes to embody strategy as well as fixed variables.) There may be various scenarios, however, reflecting the uncertainty about the external context.

Baseline. I have used "baseline" in the sense required by comparative analysis to denote a strategy that does not take global environmental considerations into account. (In some other studies, "baseline" or "base case" means the reference scenario.)

Reference. I have used "reference" to qualify "scenario." (In some other studies, it qualifies "strategy" or "case." It sometimes qualifies "baseline," since several baselines are possible.)

Change. Words indicating change such as "increase" or "reduction" are usually meant in the comparative sense, that is, to indicate differences between alternative strategies at a given point in time. (It is quite usual though for these words to mean changes that occur over time within a given strategy. Study authors should make the intended meaning clear, as this confusion is most serious.)

Case. I have used the word "case" to describe a complete set comprising of choices of baseline, alternative, and scenario. One typically evaluates many cases, corresponding to the range of assumptions tested.

3.7 National analysis

Country studies serve operational needs. It is therefore important that the analysis be a national one, nationally undertaken and linked to planned action rather than curiosity. A national study will help to build institutional capacity and facilitate ultimate implementation. The UNEP series of studies coordinated by the Collaborating Centre on Energy and the Environment is a good example of the national approach. GEF's Program for Measuring Incremental Costs for the Environment, puts a similar emphasis on its work. In this program, Regional Centers of Excellence in developing countries carry out methodology studies, case studies, and dissemination of results.


Ahmed, K. (1994). Renewable Energy Technologies: A Review of the Status and Costs of Selected Technologies. World Bank Technical Paper, no. 240, Energy Series. Washington, D.C.: World Bank.

Ahuja, D. (1994). The Incremental Cost of Climate Change Mitigation Projects. Working Paper 9. Washington, D.C.: Global Environment Facility.

Bernstein, M. (1992). Cost and Greenhouse Gas Emissions of Energy Supply and Use. Environment Department Divisional Working Paper No. 1993-40. Washington, D.C.: World Bank.

FCCC (1992). United Nations Framework Convention on Climate Change. Opened for signature in Rio de Janeiro on June 4, 1992; entered into force on March 21, 1994; UN Doc. A/AC. 237/18, Pt. II/Add. 1 and Corr. 1 (1992); text in International Legal Materials vol. 31 (1992) pp. 851-873.

Fuglestvedt, J., T. Hanisch, I. Isaksen, R. Selrod, J. Strand, and A. Torvanger (1994). GEF Working Paper No. 7. Washington, D.C.: Global Environment Facility.

Global Environment Facility (1994). Instrument for the Establishment of the Restructured Global Environment Facility. Report of the GEF Participants Meeting, Geneva, Switzerland, March 14-16,1994. Washington, D.C.

Grubb, M., J. Edmonds, P. ten Brink, and M. Morrison (forthcoming in 1994). The Costs of Limiting Fossil Fuel CO2 Emissions: A Survey and Analysis. London/Norwich: Centre for Social and Economic Research on the Global Environment.

IIASA (1993). International Workshop on Integrative Assessment of Mitigation, Impacts and Adaptation to Climate Change, October 13-15, 1993. Laxenburg: International Institute for Applied Systems Analysis.

King, K. (1994a). Issues to be Addressed by the Program for Measuring Incremental Costs for the Environment. GEF Working Paper No. 8. Washington, D.C.: Global Environment Facility.

King, K. (1994b). Global Environmental Considerations in Power Sector Development. Interregional Training Course on Power Sector Planning. April 4, 1994. Argonne, Ill.: Argonne National Laboratory.

Pearse, J. (1991). Technology Choice. Environmental Considerations in Energy Development (ed. K. King), Chapter IV. Asian Development Bank: Manila.

U.S. Support for Country Studies to Address Climate Change. Information packet. Washington, D.C.: Office of Global Change.

UNEP (1992a). Workshop on the Costs of Fossil Fuel CO2 Emissions and Workshop on the Economic Analysis of Limiting Energy-Related Greenhouse Gas Emissions. Summary Record. Environmental Economics Series, Paper No. 5. Nairobi.

UNEP (1992b). Analysis of Abatement Costing Issues and Preparation of a Methodology to Undertake National Greenhouse Gas Abatement Costing Studies: Phase I Report. UNEP Collaborating Centre on Energy and the Environment, Risų National Laboratory, Denmark.

UNEP (1993). Methodological Framework for National Greenhouse Gas Abatement Costing Studies. Appendix I: Phase II Guidelines. Appendix II: Country Summaries. UNEP Collaborating Centre on Energy and the Environment, Risų National Laboratory, Denmark.

Back to Contents