trading. It is also necessary to evaluate the Clean Development Mechanism (CDM) or other approaches to partial extension of emissions trading to developing countries in order to understand the potential impacts of such approaches on different industries. Determining the effects of the CDM or of various possible bilateral arrangements with particular countries might begin to address some of the concerns that are the focus of the Byrd-Hagel Resolution. Finally, the phenomenon known as “leakage” — the increase in emissions in countries not subject to effective carbon limits and the resulting subversion of the goals of the climate treaty—can only be analyzed effectively through the explicit modeling energy-intensive industries.

Restrictions on Trading among Annex | Countries

As in other studies, CRA finds that under an ideal, perfectly competitive market, there are potentially substantial savings to be realized from global trading versus Kyoto-style restrictions without trading. In its current wording, however, the Kyoto Protocol is silent on the topic of full global trading. Even unrestricted Annex I trading is rejected by a number of countries, including the European Union, who have advocated restrictions on the number of permits that a country can purchase and on the sale of low-cost permits by the EITs. In addition, since the EITs will be the only sellers of permits under Annex I trading, they will be in a position to exercise monopoly power and raise permit prices. "Supplementarity" - a term in the Kyoto Protocol meaning that permit trading should be supplementary to domestic action – is cited as a reason for restricting emissions trading. The European Union and several of its member countries have proposed that a concrete ceiling be imposed on what percentage of a country's obligation under the Protocol may be satisfied by purchasing emissions permits; the suggested limit is 50 percent or less. These countries have also demanded that Russia and the other EITs be prevented from selling permits in excess of their projected “baseline” emissions in 2010.

Figure 2 illustrates the impacts of these potential restrictions on emissions trading. The bars titled "Annex I – 10% Limit" and "Annex I – 30% Limit" denote the cases in which each Annex I country could only satisfy (respectively) 10% and 30% of its emissions obligation by purchasing permits. A country's emissions reduction obligation equals the difference between its baseline emissions and its emissions target. Incorporating the kinds of limits on emissions trading that have been proposed by other countries, an international emissions trading system could have very small benefits. GDP losses under scenarios with limited trading would fall between the "Annex I - Unrestricted" and "No Trading" cases, and in the worst cases would be only two-tenths of a percentage point less in 2010 than losses with no international emissions trading. Costs could be much higher still if U.S. regulators were to institute a bureaucratic command-and-control policy to regulate emissions.

'Expanded analysis of the effects on trade and competitiveness will be provided in a study forthcoming from CRA. This analysis will include more industry detail and explicit consideration of the changing competitive picture in different countries. Competitive effects show up much more strongly at the level of individual industries than at the aggregate level because it is primarily the energy intensity of an industry that determines how a carbon abatement policy will affect it.

One other possibility is exercise of market power by sellers of permits. Russia is likely to be the only net seller of permits under a trading system limited to Annex I countries. Therefore, if Russia exercises control over the emissions trading market and restricts its sales, it could charge monopoly prices for its emissions permits (see the "Annex IMonopoly" bar in Figure 2). In 2010, Russia could benefit by restricting sales by as much as 100 million tonnes and raising prices by about 30 percent above competitive levels. Regulators, however, would raise the cost further if they restricted Russia from selling its permits that were in excess of its projected baseline emissions in 2010 (see Figure 2, "Annex I - No Hot Air" bar).

Figure 2. US GDP in 2010 Under Different Annex I Emissions Trading and No
Emissions Trading Regimes

Finally, I have attached to my testimony a comparison, written earlier this year, between CRA's preliminary cost estimates and estimates of costs and permit prices released by the Administration. Briefly, I think we are in agreement that the costs of the Kyoto Protocol would be high without full participation of developing countries in an international emission trading program. For similar emission trading scenarios, CRA's estimates of the costs of the Protocol appear to be about twice as large as estimates made by the Administration. These differences are attributable to differences between the CRA Multi-Region Trade Model and the Second Generation Model relied on by the Administration. The key differences are in two areas: in

How Much Could Kyoto Really Cost?
A Reconstruction and Reconciliation of
Administration Estimates"

1

Paul M. Bernstein and W. David Montgomery

Charles River Associates
Washington, DC

Executive Summary

In December 1997, the Administration and representatives of 160 other countries negotiated the Kyoto Protocol, a climate agreement that commits industrial countries to reduce greenhouse gas emissions to a proportion of their 1990 levels. Implementing the Protocol in the United States would mean capping greenhouse gas emissions during the 5-year period from 2008-2012 at 7% below 1990 levels. The agreement authorized a form of emissions trading among industrial countries, but excluded developing countries from such trading ahd imposed no constraints on these countries' emissions.

The Administration has offered estimates of carbon permit prices and of the cost of the Protocol to the United States economy. These estimates appear to be much lower than those of many other analysts. The underlying analysis has not been released, but Administration officials have identified the assumptions under which results were obtained.

By working with these same assumptions and the economic model that the Administration used, Charles River Associates (CRA) has been able to adjust its climate policy model to correspond to the Administration's model and do three things:

1) Replicate the Administration's analysis and explain how its very low estimates for carbon permit prices and for GDP impacts were derived;

2) Assess the Second Generation Model (SGM) that the Administration is using, and demonstrate that, with minor adjustments to the CRA model, it can reproduce the results of the SGM; and

This paper was prepared under a contract with the American Petroleum Institute. It represents the independent analysis and conclusions of the authors, who are solely responsible for the contents. ·

Charles
River

Associates

3) Use the adjusted CRA model to show what happens if alternative and perhaps more realistic assumptions are used to estimate costs of complying with the Kyoto Protocol.

By undertaking these steps, CRA obtained the following results:

1) The Administration's permit cost estimates, which assume worldwide trading among all countries, show that the U.S. would be purchasing between 82% and 88% of its permits from abroad. The EU and others have objected to any country securing more than 50% of its permits elsewhere.

2) The Administration made a series of assumptions about cost saving from emissions trading. Removing the 40% saving that the Administration assumes from "umbrella trading" increases cost from $7 billion in 2010 to $12 billion. Removing the 55% saving from global trading increases the cost to $27 billion, and removing the 50% saving from Annex I trading gives a starting point, for the cost to the United States with no international emissions trading, of $53 billion. When all the Administration's assumptions about unrestricted emissions trading are removed, permit prices increase from $14 per ton to $193 per ton with no international trading.

3) The Administration's cost numbers take into account only costs in energy markets (direct costs). Most other models also take into account the impacts of higher energy costs in other markets (indirect costs), and derive estimates of GDP loss that are two to four times direct costs.

4) The Administration has assumed extremely rapid replacement of coal-fired powerplants by new natural gas plants by 2008. This is a very optimistic assumption about how rapidly large changes in natural gas infrastructure and power generation can be achieved, and there is some inconsistency between a very low permit price and achievement of these assumed changes.

5) Using perhaps more realistic assumptions about technology, fuel substitution, and the scope of international trading, permit costs of $170 -- or about 10 times Administration estimates – appear plausible even if there is a restricted form of international emissions trading. In such circumstances, GDP losses at least 10 times the costs derived by the Administration could occur, with similarly greater impacts on families, jobs, and businesses – increasing the average household's energy bill by about $850 per year and gasoline prices by almost $.50 per gallon. Such impacts are consistent with the findings of others who have analyzed the likely impact of complying with the Kyoto Protocol.