energy costs to consumers, forcing costly and uncomfortable efforts to reduce purchases of gasoline, home heating oil, natural gas and electricity.

Many of these impacts are unnecessary. Effectively addressing climate change will not be cheap even if all countries agree to institute policies that cause their citizens to reduce carbon dioxide emissions and even if these policies are implemented with maximum efficiency. Unfortunately, the approaches scheduled to be discussed in Kyoto greatly increase the cost of any given emissions reduction. There are two fundamental reasons for this increase.

First, imposing a requirement for carbon dioxide emissions reductions on the industrialized countries and not on the developing countries will be largely self-defeating. Energy costs will be much lower in countries that do not limit emissions that in those that do, causing investment in energy intensive industries to move out of industrial countries into developing countries. Demand for travel and motor vehicles is likely to increase in developing countries, because of lower fuel prices, while it declines in the U. S. and other industrial countries. As a result of the shift of industry and greater fuel consumption, carbon emissions from developing countries will increase, offsetting a significant share of the emissions reductions achieved by the industrial countries.

Paradoxically, many if not most developing countries will face somewhat slower economic growth even if industrial countries unilaterally limit emissions. The entire world is connected by trade. Emission limits in industrial countries will tend to lower the prices that developing countries receive for their exports. It will also increase the cost of the goods that developing countries import from industrialized countries.

Second, even if all countries made commitments to reduce their carbon dioxide emissions, near term emissions limits raise the costs of achieving any eventual global warming objective by causing the countries agreeing to them to forego important cost-saving opportunities. These opportunities include the significant cost reductions achievable through avoiding premature obsolescence of capital equipment. They also include the cost savings resulting from the development and implementation of lower cost technologies for reducing emissions. This is a genuine instance in which “haste makes waste." Since any global warming impacts are driven by concentrations and not emissions of greenhouse gases, countries that incur these extra costs do not necessarily generate any offsetting benefits.

Many of the adverse effects on both industrial and developing countries can be avoided if a mor inclusive and measured approach to emission reduction is taken compared to those contained in the current proposals. This requires full participation of developing countries in commitments to limit emissions and a considerably longer period of time for development and deployment of new energy technologies than is allowed by targets for 2010.

Impacts on the United States economy

I will now turn to a more detailed discussion of our findings about how limits on carbon emissions, adopted solely in the industrial countries, will affect the United States. In developing these estimates, we have assumed that the most efficient, market based policy instruments are

used, but that developing countries do not limit their emissions and that emission limits must be met by 2010.

U.S. economic welfare will necessarily fall because resources will be used less efficiently. Low cost fossil fuels make U.S. workers more productive and provide U.S. consumers with mobility and comfort. Limiting the amount of fossil fuels that can be used will require switching to less costly and satisfactory substitutes, thus lowering productivity, mobility and comfort.

Investment in the U. S. economy will fall, because of the anticipation of lower future productivity and returns to investment and better returns in other regions of the world not subject to as stringent carbon limits. This will lower the long run growth of the U.S. economy.

U.S. energy-producing industries, U.S. energy-intensive industries and the U.S. automobile industry will all face significant losses in output and employment.

U.S. exports of energy-intensive goods will fall and U.S. imports of these same goods will rise as developing countries with energy intensive industries and no carbon limits increase their share of these markets.

All regions of the U. S. will suffer, but those with heavy concentrations of energyproducing and energy-intensive industries will face much larger losses in output and employment than average.

How difficult is it to hold U. S. emissions at 1990 levels?

Holding emissions at 1990 levels from 2010 onwards will require significant efforts, because in the absence of limits on emissions, economic growth will cause rising demand for energy services. Even assuming that there will be substantial future improvements in energy efficiency, the U. S. Energy Information Administration projects that carbon emissions are likely to grow to more than 25% above 1990 levels by 2010. Thus it would be necessary to reduce emissions by 20% or more below the levels they would otherwise reach to meet the 1990 target, and further to comply with more ambitious proposals.'

1

These two numbers are in fact consistent. If emissions grow to an index of 125 by 2010, relative to 100 in 1990, that is a percentage increase relative to 1990 of 25/100, or 25%. To bring emissions down from 125 to 100 is a percentage reduction relative to the 2010 baseline of 25/125, equal to 20%.

Figure 1: Carbon emissions forecast, emissions limit, and the emissions gap.

2

The top line in Figure 1 displays how U.S. carbon emissions are projected to grow in the absence of any greenhouse gas abatement policy. The lower line denotes the carbon emissions limit that would come into effect in 2010. Therefore, the difference between the top and bottom line indicates the amount of carbon emissions that must be reduced from the level that carbon emissions would otherwise reach under no abatement policy.

Reductions in national income from limiting emissions

Figure 2 displays the change in the U.S. GDP, consumption, and investment due to a cap on emissions at 1990 levels from 2010 onward. GDP declines by 1.0% ($94 billion) in 2010 and increases to a loss of 2.7% ($370 billion) in 2030. Consumption, a measure of the standard of living, falls by a similar amount.

Through 2010, this projection follows that of the Energy Information Administration Annual Energy Outlook 1997, Reference Case. Differences in estimates of what measures will be required to cut off the growth in energy consumption and emissions that may occur over the next 13 years are a significant source of uncertainty. So is the forecast of emissions growth, since the proposals under negotiation would put a firm ceiling on emissions, no matter what it costs to keep emissions that low. To use a neutral source, CRA and other recent studies have adopted the Energy Information Administration's reference case projections for emissions growth, but having directed EIA's forecasting activities for many years I can testify that EIA has and still considered such projections to be highly uncertain and not forecasts of what must be. For 2030 the forecast is consistent with the IPCC IS92a

Figure 2: Percentage change in U.S. GDP, consumption, and investment from the no carbon abatement scenario.

Investment also falls dramatically, to 7% below baseline levels by 2030. Emissions limits will reduce the rate of growth in the economy by reducing incentives for investment and diverting investment into projects designed to replace fossil fuels rather than produce goods and services for consumers (see Figure 2). The fall in aggregate investment comes about despite significant efforts to reduce energy use by investing in more energy efficient equipment and processes. Part of the reason for the drop in domestic investment is that savings are invested overseas rather than in the United States because of the enhanced prospects and profitability of energy-intensive industries in developing countries that do not adopt emissions limits.

Costs of reducing emissions

In order to comply as a nation with a cap on emissions, it is necessary to provide an incentive to industries and consumers to induce them to reduce greenhouse gas emissions. As I discuss below, this incentive could appear in many forms: a tax on burning fossil fuels; a subsidy on substitutes for fossil fuels or energy efficient technologies (e.g., solar energy); a permit to emit a given amount of pollution; or command and control regulations that require specific technologies to be used. For any method, there is a cost associated with abating a ton of emissions. Our model assumes that regulators employ the least cost, most economically efficient policy possible. Figure 3 reports our model's result for the cost to eliminate a metric ton of carbon under a cap on emissions at 1990 levels from 2010 onward.

Uncertainties about these costs are all on the upside. Not using carbon taxes or emission trading

Figure 3: Cost per metric ton of carbon abatement needed to stabilize carbon emissions at 1990 levels.

Table 1 reports the corresponding cost increase that could be applied directly to fossil fuels so as to induce the same level of abatement. By applying the cost increase directly to fossil fuels, regulators would induce people to burn less oil, natural gas, and coal.

In percentage terms, an emission trading program or a carbon tax sufficient to return emissions to 1990 levels in 2010 would produce a 23% increase in the cost of electricity, a 45% increase in the cost of heating oil, a 40% increase in the cost of gasoline, and a 46% increase in the cost of natural gas in 2010 to households.

As the U.S. economy grows and it becomes more difficult to hold emissions at this fixed level, increasingly expensive efforts will be required to hold to the cap. Therefore, the carbon tax or permit price would rise to about $400 per tonne by 2030. The taxes on individual fuels would also increase in proportion to the growing carbon tax.

Changes in U.S. energy prices relative to those in developing countries

The competitive position of U.S. industries will be affected by any agreement that exempts some group of countries from obligations to limit emissions. Developing countries will benefit from lower energy costs than they would face in the absence of emission limits, because the drop in