he industrial sector includes agriculture, mining,

represented a relatively small fraction of total production costs in most industries (ranging from less than 2 percent for the food industry to as much as 30 percent of production costs annually for the cement industry.) Consequently, the price of energy has not played a dominant role in recent years in improving industrial energy efficiency.

U.S. energy intensity (energy input per dollar GDP output) has been influenced by two factors in the U.S. economy. First, structural shifts have occurred where the mix of goods and services has shifted away from the production of energy intensive goods (e.g.,iron and steel) and toward services (e.g., telecommunications, entertainment). Second, technological improvements (e.g., continuous casting and electric arc furnaces) have impacted industrial efficiency. For more than a quarter century, the gradual migration of heavy industry and primary metals abroad, growth in the U.S. service sector and technological progress have combined to reduce U.S. energy intensity by about 1.4 percent annually.

The Reference Case projection embodies considerable improvement in energy intensity. Just as lower real prices for oil, gas, and electricity during the late 1980s lowered incentives for energy conservation, the higher prices projected in connection with meeting the Kyoto carbon emission reduction goals result in less energy consumed for each constant dollar of output produced. This analysis projects a replay of the early 1980s, when the cost of both energy and capital inputs were rising. Across the cases, about two-thirds of the projected reduction in industrial I energy intensity is attributable to the structural effect (change in the mix of goods and services); the remaining one-third to increased energy efficiency.

In the carbon reduction cases, relative to the Reference Case in 2010, total carbon emissions from the industrial sector are between 7 and 28 percent lower. Part of this is due to lower U.S. industrial production because higher energy prices lower domestic demand and make U.S. exports relatively more expensive. In addition, industrial consumers are likely to replace existing capacity somewhat faster, invest in more efficient technology, and switch to less carbon-intensive fuels.

order to meet the Kyoto Protocol targets, the proportion of each fuel used in the United States is projected to change from that of the Reference Case. Because of the higher relative carbon content of coal and petroleum products, those two energy sources would be used less-placing more reliance on natural gas and renewable energy, and slowing the decline in nuclear power. Although petroleum use declines relative to the Reference Case in absolute terms, its percentage-share increases slightly because total energy demand is lower in the carbon reduction cases. Most petroleum is used in the transportation sector-where there are limited economic options for fuel substitution. As noted earlier, domestic coal consumption declines substantially, with most of the reduction coming in electricity generation.

Because of lower demand for petroleum in the United States and other developed countries committed to reducing greenhouse gas emissions, world oil prices in 2010 would be about 4 to 16 percent lower than they would have been without the effect of any "carbon price." At the same time, U.S. dependence on imported petroleum by 2010 would be lessened-from about 59 percent of all petroleum consumption in the Reference Case to as much as 53 percent.

More use of natural gas in electric generation is offset only partly by reductions from the end-use sectors. Although carbon emissions from burning natural gas are lower than from coal or petroleum, they are not zero. Later in the forecast period, natural gas starts to face stiffer competition from increasingly economic non-hydro renewables and, particularly in the more stringent carbon reduction cases, from refurbished existing nuclear power units. As a result, by 2020, natural gas use is highest in the 1990+9% Case.

Higher demand forces up the average wellhead price of gas, relative to the Reference Case-moderately by 2010 ($2.78 per thousand cubic feet for the 1990+9% Case, vs. $2.33 for the Reference Case) but considerably by 2020 ($3.71 vs. $2.62). Although the natural gas

54-190 99-6

Even though energy represents only about 7 percent of

our Gross Domestic Product (GDP), it is a crucial factor in virtually all the goods and services we produce and consume. The effects of alternative carbon scenarios can be expressed in terms of their impacts on the economy as a whole (e.g., GDP, capital investment, prevailing interest rates, inflation rates) as well as their impacts on individual families and enterprises (e.g., increased expenditures on energy, disposable income).

When energy costs rise, other factors of productionincluding labor and capital-become relatively less expensive. Energy price increases encourage adjustments in which labor and capital are substituted for more expensive energy to the extent practicable. In the process however, some economic potential is lost. This reduces the "potential" GDP for the Nation. EIA calculates that such losses would range from $13 billion to $72 billion in 2010 (1992 dollars). In an economy today of over $7 trillion, which is expected to grow to over $9.4 trillion (1992 dollars) in 2010, the percentage loss in output ranges from 0.1 percent to 0.8 percent.

In this context, the economy continues to grow, but at a slower rate.

Because it takes time to adjust to a new set of factorcosts, however, there are, in addition to losses in potential GDP, transitional costs-which probably cannot be avoided entirely. Such short-run costs arise whenever price increases disrupt capital or employment markets. The transitional costs are very uncertain, but possibly very significant. They impact the "actual" GDP. Hence, the actual GDP losses are greater than the "potential" losses. The transitional costs can be softened to the extent that price changes are anticipated and appropriate compensatory adjustments can be made to Federal monetary and fiscal policies.

This analysis assumes a carbon-permit trading system is introduced in the form of an auction run by the Federal Government (to focus on the most economically efficient means of reducing carbon emissions). The domestic auction would produce substantial revenue. This study assumes that the revenues would be recycled back into