Key Assumptions

[ Home ] [ Up ]

When evaluating the meaning of modeling studies, it is important to consider the nature and value of assumptions that make up the external inputs to the model and the assumptions embedded in the models.  Modeling assumptions affect the predicted economic impacts of carbon emission reduction.  For example, a recent study by the World Resources Institute (WRI) identified a list of eight assumptions (included in the table below) as the key assumptions that differentiate the model predictions of the costs of implementing a carbon tax.  Values of these assumptions, along with assumptions about predicted emissions reduction, were found to account for 80% of the variation in predicted economic impacts of emission reductions.  The table below summarizes the key assumptions identified in the WRI study, the Second Assessment Report of the Intergovernmental Panel on Climate Change, and the Manne and Richels paper, "The Costs of Stabilizing Global CO₂ Emissions: A Probabilistic Analysis Based on Expert Judgments" [1, 9, 12].   Follow the links for a more in-depth discussion about a given assumption, what factors determine its value, and the effects it may have on predicted climate change mitigation costs.

Emissions Targets

The first, and most obvious difference to check for when comparing mitigation costs studies is the target assumption used.  There is no conventional benchmark for reduced carbon emissions that has been adopted for estimating costs [18].  Some studies predict the cost of attaining carbon emission levels that are some fraction of the future level that would occur in the absence of any emission controls (business as usual).  Other studies consider the cost of achieving a past level of emissions, and still others look at the costs of stabilizing the global CO₂ concentration at an arbitrary, predetermined level [11].  For example, the United Nations Framework Convention on Climate Change chose to explore the cost of maintaining a global concentration of 550 parts per million by volume [10].  The economic costs will depend on the target, as well as the manner in which that target is reached [16].  Although the studies look at emissions, or, in the case of carbon sequestration, in the quantity of stored carbon, keep in mind that ultimately, the goal of climate policy should be to stabilize the atmospheric concentration of carbon dioxide and other greenhouse gases.

Intuitively, the target level that will require the greatest emission reduction will be the most costly.  The benchmark options described above will not all result in the same level of required emission reductions.  If an emissions limit based on past emissions levels is used, as is proposed by the Kyoto Protocol, the overall emission reduction required will be greater than if a percentage of predicted future emissions is used as the target.  This is because the future emissions level accounts for economic and population growth that will occur over time and naturally lead to increasing emission levels.  

  Back to top.

Baseline and Future Emission Levels

Obviously, the predicted level of future emissions is a very important predictor of emission reductions costs.  Reference cases used for models project future emissions in the absence of control measures, and, to some extent, the costs of emissions reductions are determined by this emissions baseline [1, 9].  The higher the baseline, the larger the amount of carbon that will have to be removed to meet the target and the more it will cost [9].   The following equation is used to illustrate why these emissions level projections differ [1, 9]:

Growth rate in emissions =             growth rate in GDP

-    the rate of decline of energy use per unit of output

-    the rate of decline of CO₂ emissions per unit of energy use

Two parameters that are often considered separately and that have an effect on future emissions levels are population and economic growth.  Holding all else constant, an increase in population will result in an increase in greenhouse gas emissions.  Likewise, increased economic growth increases activities that use energy which will lead to increased emissions.  However, the increased of economic growth also increases the turnover of capital which will allow more efficient technologies to penetrate the market and somewhat offset the increase emissions caused by the activity growth.

  Back to top.

Energy Demand (Structural change and Technological change)

Assumptions about the degree to which the structure of the economy changes will impact the cost estimates because different sectors of the economy have different energy intensities and thus structural changes will significantly impact overall energy use.  For example, if there is a shift in the economy away from manufactured goods and toward more services, energy consumption may decline. 

Technological changes can also lead to reductions in energy demand.  Specifically, non-price efficiency improvements can occur, for example, through changes in public policy.  Detailed, disaggregate models usually capture this potential through engineering studies or by endogenizing a variable that represents the evolution of technologies [10].  Econometric models use a parameter called the autonomous energy efficiency improvement (AEEI) to represent changes in technology.  AEEI is a proxy for the non-price factors that affect the tendencies of technology change and suggests the rate at which the new technology will penetrate the economy and lead to reductions in the economy-wide energy intensity per unit of output [1, 9].  The higher the value of the AEEI, the lower the required amount of energy and the lower the level of emissions which leads to lower costs of achieving a given level of emissions.

  Back to top.

Energy Supply  (Short-term and Backstop technologies)

The treatment of backstop technology is another aspect of cost estimation models that has the potential to significantly affect results.  As prices of fossil fuels increase, alternative energy sources such as wind and solar energy will become more attractive.  The short-term availability of alternative supplies determines the potential for fuel substitution.

The backstop technology refers to an alternative energy source available in unlimited quantities at some price [12].  Models include numerous assumptions about the non-fossil energy sources, such as availability, initial cost, and subsequent cost changes.  Some models exclude backstop technologies even though ever-increasing prices of fossil fuels are contained in the models [4].  Other models recognize non-fossil sources, but assume limited availability, which leads to rising prices for alternative energy as well.  Still other models assume that backstop energy sources will be available at non-increasing prices [12].  The cost at which an infinite alternative supply of energy becomes available sets an upper bound on the cost of mitigation. 

  Back to top.

Timing

The  time frame considered is very important.  Assuming that the ultimate target is to achieve some atmospheric concentration, the world is working with a set carbon budget.  Depending on the timing of cutbacks, the costs associated with reduction in emissions will be higher or lower. 

The first consideration is the flexibility of consumption patterns [1]. Some changes are near impossible or at the least very costly to make in a short amount of time.  For example, changing transportation infrastructure may be an effective solution for reducing CO₂ emissions, but would not be economically feasible except in the extreme long term.

Another reason it might be cost-effective to use up the carbon budget now and make reductions later is that any changes will necessitate changes in capital stock.  Because capital stock is durable and a lot of equipment and building will ultimately need to be changed, adjustment will be a costly process [12].  If time is allowed for the transition, capital stock can be replaced when it wears out, and would need to be replaced anyway.  This will lower the overall costs associated with emission reductions. 

Another advantage to postponing transition is that research and development may yield more energy and cost efficient technologies.  The availability of technologies for alternative fuels and for efficiency options will increase over time.  This will mean the difference between an expensive change in capital stock that will lead to minimal energy efficiency gains and a delayed change in equipment that will lead to a substantial efficiency gain. 

A potential problem with near-time action is that changes in development patterns may create irreversible processes of technical change [1].  The risk is making a decision that is not fully informed.  The effect of early action is detrimental when a particular tract of change leads to a development pattern that makes once available alternatives and choices impossible.  These irreversibilities can lead to given socioeconomic development paths that exclude the potential for alternative development paths [1].

Finally, assuming a positive discount rate, future expenditures are favored over present expenditures [1].

  Back to top.

Price and income elasticities of energy demand

The price and income elasticities of energy demand measure the relative change in energy demand, given relative changes in energy prices and in incomes [1].  A higher elasticity means a larger change in energy use.  The degree of substitutability will affect the economic losses from energy scarcities and from price increases.  Elasticity of substitution allows for a measurement of the degree to which capital or labor can be substituted for energy and thus gives insights as to how expensive it is to decouple energy consumption from GDP growth during a period of rising energy prices [7, 9]. 

  Back to top.

Revenue Recycling

One debate in the literature revolves around the costs associated with the use of carbon taxes to reduce emissions.  Most macroeconomic models have focused on carbon or energy taxes as a means to induce emission reductions [1].  Models assessing costs must make assumptions regarding what is done with the tax revenue and what, if any, potential affects these revenues will have on the overall cost of emission reductions.  The extent to which the costs of carbon taxes can be reduced by recycling of the revenues from such taxes has been at the heart of the climate policy debate [11].

A carbon tax would generate a significant amount of revenue [12].  These revenues can either be returned to the economy in a lump sum or used to finance cuts in the marginal rates of existing income taxes [11].  Studies have confirmed the fact that gross costs of carbon taxes can significantly be reduced when the revenue is used to reduce the marginal income tax rate.  The real controversy surrounds the question of whether revenue recycling can make the gross costs of revenue-neutral carbon tax become zero or negative [11].

Some economists claim that there is a possible economic double dividend, or positive side-effect on growth or employment, from the recycling of carbon tax revenues [1].  The current tax system reduces private incomes by more than a dollar for every dollar of tax revenue collected by penalizing work, savings and investments [12].  The logic is that the use of tax revenues to reduce existing tax rates benefit society by reducing the costs of distorting taxes while at the same time correcting a market failure in energy use, thus creating an economic double dividend.  Some models even suggest that substituting a carbon tax for other taxes could provide net economic benefits irrespective of the environmental gain [12].  This is not always true however, and many economist argue against this strong double-dividend theory.

Recent work indicates that carbon taxes tend to be less efficient sources of revenue than income taxes.  This is because carbon taxes are applied to intermediate inputs or consumer goods.  Thus, these taxes raise output prices and lower the real returns to factors like labor and capital [11].  Simply put, this will translate to an increase in income because of the reduction in income tax, but will also result in an increase in prices of goods, so consumer spending power does not actually increase.

  Back to top.

Joint Implementation

Joint implementation in the climate change context would involve a country or firm paying for an emission reduction or carbon sequestration project in another country or firm rather than reducing its own emissions [17].  The idea behind joint implementation is that, by allowing for the trading of emission rights, emission reductions can be made where it is least costly to do so.  Assuming the trade is a one for one trade, then the gain from the trade is this lower cost emissions reduction.  In some scenarios, the trade is greater than a one for one trade, which corresponds to net benefits of a lower cost emission reduction plus a change in net emissions.  This discussion will focus on joint implementation, as it is described above.  Recently, however, joint implementation has been interpreted more broadly to include the flexibility of emission rights over time [17]. 

Thus far, joint implementation has been attempted only in pilot projects [12, 17].  There are several inherent difficulties in setting up an international trading system.  First, countries would have to agree to a baseline from which net emission reductions would be measured.  Second, monitoring and enforcement would be necessary to ensure that agreements were carried out.  This would require some type of international standards and an international enforcement body.  Problems exist because such entities interfere with national sovereignty  [13, 12].

From the perspective of an individual participating country or region, assumptions about the possibility of international trading can significantly impact the bottom line.  Because trading can essentially lower the actual emissions reduction that must be made by any one country, the underlying assumptions must be examined closely.  For example, the Clinton Administration analyzed the costs of complying with an emissions reduction scenario similar, but not identical, to the scenario posed by the Kyoto treaty [8].  The treaty, which has not been ratified, is an international treaty agreement that addresses the need for global reduction in greenhouse gases and does contain vague provisions for a trading mechanism [12].  The Second Generation Model results predicted the cost of emission reductions in terms of dollars per ton of carbon.  Without trading of emissions permits the implicit cost of emissions reduction came to $108 per ton of carbon reduced; with trading among industrialized nations, the cost came to $72 per ton; and with worldwide trading the costs were only $26 per ton [8].  This price discrepancy explains why the Clinton Administration is insisting on meaningful participation by the developing countries.

Meaningful global participation also has the advantage of deterring the occurrence of leakage.  When abatement actions are limited to a given region or subset of the population, leakage can occur in a number of ways [1].  For example, production of energy-intensive products can be shifted to regions that are not abating [1]; a decline in the domestic coal market could promote exports, which would complicate the predictability of CO₂  emission reduction [3]; and the reduced demand for fossil fuels could lower prices and encourage increased energy consumption in nonabating regions [1].  Scenarios that do not allow for global participation in emission abatement should account for such leakage effects.