Page images
PDF
EPUB

investment. As a result, investment will fall and the growth of the economy will be reduced for a considerable period of time. This does not happen in developing countries that do not adopt emission limits. Therefore there will be a shift in investment, and rather than investing in the United States, individuals and institutions will instead invest in developing countries where returns are higher.

This shift would be particularly strong for energy-intensive industries. We estimate that with industrial country emissions limited to 1990 levels from 2010 onward, investment in the Non-Annex 1 countries would increase by $85 billion in 2010, with over half going to energy-intensive industries. Investment in the U.S. economy would fall by over $100 billion in the same year.

Environmental Benefits of an Agreement Limiting Emissions in Industrialized Countries

Q14. In your opinion, what would an agreement limiting emissions in industrialized countries do for the environment?

A14. An agreement limiting emissions in industrialized countries alone would do very little for the environment. Over the next century, the time frame that matters for climate change, most emissions will come from developing countries. Limiting emissions just in the industrial countries will encourage energy-intensive industries to relocate to the developing countries. Typically, energy-intensive industries in these countries are already much less efficient than their counterparts in the industrial countries. Moreover, reduced demand for oil by the industrial countries will depress the world price of oil, encouraging developing countries to be even less efficient in their use of energy. All these factors lead to the conclusion that increased emissions in the developing countries will negate a large share of the emission reductions paid for at high cost in the industrial countries.

FOLLOW-UP QUESTIONS FROM DEMOCRATIC MEMBERS

Economic Model Assumptions

Q1.

Al.

In their study for World Resources Institute, Repetto and Austin found that just 8 assumptions accounted for 80 percent of the differences in the various model projections. What do each of the models that you rely upon in your testimony assume with respect to the following factors:

Q1.1. the benefits of avoiding global warming and other environmental benefits of reducing fossil fuel consumption (e.g., SO2 reduction);

Q1.2. the economic benefits of “recycling” any tax revenues back into the economy;

Q1.3. the rate of innovation and adoption of new efficiency or non-fossil technologies?

First, the WRI study confuses the discussion of climate policy by failing to observe the distinctions we make in ordinary language between costs and benefits, and by mixing categories in a way that reduces the information they convey to policymakers. Benefits and costs should be analyzed separately, and weighed against each other in any decision process. Our study, like many others, examines costs. It is particularly important that independent studies like our discuss costs since the Administration has not provided such estimates in its recent statements on climate policy, and has emphasized instead some highly conjectural benefits of reducing emissions. Benefits are more uncertain, occur over a different time frame, and are far harder to estimate. Keeping these two categories separate provides more information than lumping them together, as Repetto and Austin suggest. Taken literally, our study does include all the benefits of avoiding global warming that could be expected to accrue to the United States over the next 30 years, because the policies under consideration would not produce any such benefits over that time period.

When Repetto and Austin claim that including benefits of avoiding global warming reduces costs, confuse model structure with modelers' assumptions. Only a few analysts have attempted to include benefits in their models, and some of them (primarily William Cline) also assume costs far below the average of the literature. The Repetto and Austin "regression analysis" pays no attention to the actual methodologies used in different studies or to the basic economic theory of costs and benefits. Thus what Repetto and Austin have found is not any agreement among analysts that properly measured benefits outweigh costs, but rather that some analysts who include benefits also make very optimistic assumptions about costs relative to the norm among analysts.

Second, we assume that tax revenues are fully and costlessly recycled. We follow the general equilibrium literature (Goulder, Bovenberg, Parry, Bovenberg and Rutherford) in assuming that at best recycling can offset the "tax interaction" effect, but cannot reduce

the basic deadweight loss of energy taxation (see Goulder and Parry, recent RFF discussion paper). WRI fails to cite this literature, and its conclusion that including recycling through reducing burdensome taxes in a model would lead to lower estimates of costs is incorrect and a gross distortion of the literature. Bovenberg and Rutherford include pre-existing taxes and recycling through reduction of those taxes in a model closely related to ours (of which Rutherford is one of the principal authors) and find that costs actually increase through inclusion of recycling in a theoretically consistent way.

Third, we have included optimistic assumptions about how much new efficiency will occur in the baseline (using a parameter called Autonomous Energy Efficiency Improvement that is set equal to 0.75%, compared to Energy Modeling Forum guidelines of 0.4%) and about how use of fossil fuels will respond to policies that increase energy prices (called the elasticity of substitution between energy and non-energy goods and set equal to 0.5). As recent theoretical work on induced technical progress by Goulder and Wilcoxen makes clear, these two parameters capture everything there is to say at the aggregate level about how technical progress will respond to climate policies. I believe that our assumptions imply greater responses to price increases than we have seen historically when energy prices have changed.

Energy-Intensive Industry Trends

Q2. The analyses discussed by Mr. Montgomery and Mr. Roberts indicate significant decreases in output, deterioration in balance of trade, lower employment, and relocation of energy-intensive industries from the U.S. to developing countries. The Argonne Report concludes a similar pattern will occur, but also indicates that some of these industries have, and are continuing, to show this pattern even without any changes in energy prices.

A2.

Q2.1. Do your projections under a scenario of higher energy prices reflect a change in trends for these industries or an acceleration of current trends?

Q2.2. What are the other economic factors that account for the trends we see in these industries?

We began by projecting trends in the output of different industries in different countries, based on perpetuation of current different differences in energy costs across countries. This leads to some shift in industries across regions. We then modeled emission limits in the industrial countries, and found that they would raise costs and depress the output of energy intensive sectors in the industrial countries and lower costs and increase output of energy intensive sectors in the developing countries. These changes are relative to baseline levels that already incorporate the basic trends in each region and industry.

"Top-Down” vs. “Bottoms-Up" Models

Q3.

A3.

We have heard the terms "top-down" and "bottom-up" used to describe different types of models used to evaluate economic impacts of energy price changes.

Q3.1. What do these terms mean?

Q3.2. Which type of model was used to produce the results you are discussing?

Q3.3. What are the strengths and weaknesses of each of these approaches?

To quote a paper by Mark Jaccard and myself published in the journal EnergyPolicy last year, "Top-down studies have generally been based on more aggregate economic analysis which places energy supply and demand in the context of a model of the entire economy. These studies have generally concluded that emission reductions are costly. Bottom-up studies have been based on engineering studies of specific energy efficiency or renewable energy technologies and have concluded that emission reductions achievable with those technologies are either nearly free or have negative costs.... The differences between these models arise in large part from differences in how market behavior is characterized. In the technology based models, future energy savings are assumed to be inadequately valued compared to initial capital costs for energy efficiency....In market based models, consumers are assumed to make choices that are in their economic interest.... Thus, if a new technology for conserving energy becomes available, and can reduce correctly discounted lifecycle cost, it will be adopted in the baseline."

We have used a combination of top-down and bottom-up analysis. Top-down models take into account the way in which decisions are reached in the marketplace, and utilize the information we can gain from observing past responses to energy price changes. They also include a view of how the entire economy reacts when energy costs and supply and demand change, which is left out of bottom-up models. Bottom-up models represent energy technologies explicitly. Poorly done bottom-up studies, like the Five-Lab study, stop there and limit their analysis to comparisons of costs of various technological possibilities. We have attempted to avoid this error, and used bottom-up analysis of costs of in the transportation sector and the electric utility sector to give us cost-functions that we included in our top-down economic models. Then we proceeded to model how rational consumers and businesses, working in generally competitive and well-functioning markets, will choose among the array of technological possibilities and to estimate the costs of policy interventions that change those choices.

In particular, we included the hidden costs that are ignored in poorly done bottom-up studies. For example, a detailed analysis of fuel economy standards for new cars (CAFE standards) reveals that these standards can impose hidden costs on consumers by changing attributes of vehicles (acceleration, NVH, driveability, towing capacity) that consumers value and forcing consumers into smaller vehicles than they prefer. These hidden costs are

Reasons for Not Reducing CO, Emissions Through Voluntary Efforts

Q4. In 1993, President Clinton announced the Climate Change Action Plan, a voluntary plan which was intended to meet our commitments under the Framework Convention for Climate Change to cap emissions at 1990 levels. Emissions today are now about 10% higher than they were in 1990. Why do you think we did not meet our goals to reduce CO2 emissions through voluntary efforts?

A4.

We did not meet the goals of the climate action plan because it is costly to reduce emissions below baseline levels, and rational consumers and competitive industries will not undertake those costs without adequate incentives. The GCAP did not recognize that its goals would be costly and did not provide adequate incentives.

Energy Prices in the U.S. vs. Other Countries

Q5.

A number of the analyses appear to assume that energy price rises in the U.S. will necessarily result in our energy prices being higher than those of other countries.

Q5.1. How do U.S. energy prices currently compare with those of our economic competitors?

A5.1 U.S. energy prices for industry are generally lower than those in Europe and Japan, and higher than in Canada and Australia. The figure below shows the prices used in our international trade model for key regions of the world (blanks indicate data not available from the International Energy Agency).

« PreviousContinue »