2. Energy Market Impacts of Alternative Greenhouse Gas Intensity Reduction Goals
Greenhouse Gas Emissions and Permit Prices
Relative to the AEO2006 reference case, projected GHG emissions are reduced starting in 2012 as a result of the allowance program and the GHG intensity targets (Figure 1). With banking of allowances permitted, an incentive exists to over-comply in the first few years of the program, when the emissions targets are relatively easy to meet and allowance prices are low, and to draw down the bank balance later as the targets become more stringent and prices rise. As a result, projected emissions in both the Phased Auction and Full Auction cases are below the intensity-based target for emissions from 2012 to 2020 and above the target thereafter.19 Once the market price for allowances reaches the safety valve level in 2026, the gap between covered emissions and the target continues to widen.
The market price for emissions allowances is assumed to reflect financial incentives to bank emissions allowances over time, limited by the established safety-valve price, which grows over time at a rate of 5 percent per year (after inflation). To represent the proposed bill, a path of allowance prices is calculated based on an assumed rate of return on banked allowances of 8.5 percent per year up until the safety valve price is reached.20
Under these assumptions, the allowances prices derived in the Phased Auction and Full Auction cases are essentially the same (Figure 2). However, the economic impacts of these cases, and some of the energy market impacts, differ due to the distributional impacts from allowance allocation. Table 2 summarizes the emissions and energy market impacts of the Phased and Full Auction cases relative to the AEO2006 reference case.
Lower energy related CO2 emissions and other GHG emissions both contribute to the total reduction in GHG emissions, but their respective shares of total reductions change over time. Abatement cost curves for other GHG based on EPA research suggest that there are a significant amount of emissions reductions that can be made at relatively low costs. As a result, in the early years of the program, when allowance prices are relatively low, other GHG emissions reductions dominate the overall emissions reductions. For example, in 2020 in the Phased Auction case, reductions in other GHG emissions account for nearly 66 percent of the total GHG emissions reductions (Figure 3). By 2030, however, higher allowance prices lead to a significant shift in the fuels used in the energy sector, particularly in the electricity sector, and the reduction in energy related CO2 emissions account for almost 58 percent of the total GHG emissions reductions.
There is also an increase in carbon sequestration in the Phased Auction case. The draft bill calls for allocating up to 5 percent of the allowances each year to create a pilot program to stimulate increased agricultural carbon sequestration. Because it is a pilot program, the increase in sequestration stimulated does not count towards meeting the GHG intensity reduction target. However, based on the supply curves for carbon sequestration opportunities used in EIA’s recent GHG analyses, the allowance allocation incentives would stimulate an increase in carbon sequestration of over 300 million metric tons CO2 equivalent (Table 2). In the Full Auction case, the allowance allocation incentives for carbon sequestration were assumed unavailable, since no allowances would be available to use as incentives, and the draft bill proposed no other mechanism whereby auction revenues might be used directly to promote carbon sequestration.
Because the safety-valve price is binding in these cases, establishing higher or lower safety-valve prices will influence the emission compliance results, including incentives to bank allowances and the extent to which the emissions targets are achieved. Figure 4 compares covered emissions in the Phased Auction case to those in the $5 and $9 Phased Auction cases. In the $5 Phased Auction case, which assumes a lower starting value for the safety valve, the allowance price reaches the safety-valve level in 2022, four years earlier than in the original Phased Auction case, which assumes a $7-per-metric-ton starting price for the safety valve (nominal dollar price). Covered emissions in the $5 Phased Auction case remain higher than the original Phased Auction case throughout the projection. In the $9 Phased Auction case, the safety valve is not triggered until 2029. As a result, on a cumulative basis, the emissions intensity targets are projected to be achieved through 2029 in the $9 Phased Auction case, compared to 2025 in the original Phased Auction case.
In the No Offsets case, the provision to allow compliance with emissions reduction credits from uncovered sources is removed. Without these relatively low-cost emission reduction opportunities, the marginal compliance cost is driven up, reflected in a higher market price for allowances. The projected allowance price in the No Offsets case reaches the safety-valve level in 2022, 4 years earlier than in the Phased Auction case (Figure 2). As a result, the overall reductions in GHG emissions are somewhat less in the No Offsets case (Figure 4).
In previous assessments of similar cap-and-trade proposals for GHGs, EIA has included sensitivity results to highlight some key areas of uncertainty, including the importance of the assumed abatement costs for non-CO2 gases and the effect of alternative assumptions about energy technology cost, performance, and availability. Rather than repeating those sensitivity cases, we will simply point out that these assumptions have had a bearing on the magnitude and cost of compliance. With less optimistic assumptions about abatement opportunities for other GHGs, the projected cost of compliance is driven up, the safety-valve price is likely to be attained several years earlier, and cumulative emissions reductions will be correspondingly less.
With regard to technology assumptions, the technological improvements reflected in the AEO2006 reference case may under- or overpredict future technology trends. For example, the results of the integrated advanced technology case indicated projected CO2 emissions would be 9 percent lower than in the reference case in 2030. Under such assumptions, compliance costs under the proposed bill would be less, and the intensity targets would be achieved over a longer time frame before the safety-valve price is triggered, if at all.
As shown in Figures 1 and 2, the emissions and allowance price paths in the Phased and Full Auction cases are very similar. As a result, with the exception of electricity price and macroeconomic effects, the market impacts in the two cases are essentially the same. The analysis in the remainder of this chapter will focus on the Phased Auction case, only discussing other cases where they are important.
Electricity Sector Emissions, Generation and Prices
Implementing the proposed GHG intensity reduction program could have significant impacts on power sector CO2 emissions, generation by fuel, generating technology selection, electricity sales, and electricity prices. The power sector shifts away from its long-term reliance on coal-fired generation, towards increasing reliance on nuclear, non-hydroelectric renewable, and natural gas generation. These changes lead to lower emissions. However, increased capital expenditures for these technologies, together with higher fossil-fuel prices, result in higher electricity prices. Because a portion of the allowances are allocated to regulated utilities for free in the Phased Auction case and because regulators are expected to pass these savings on to consumers, the impact on electricity prices is slightly smaller than in the Full Auction case.
CO2 Emissions
In the reference case, total power sector CO2 emissions are projected to increase 44.4 percent between 2004 and 2030 as the industry increases its use of fossil fuels, particularly coal (Figure 5). However, in the Phased Auction case, CO2 emissions are forecast to increase by less than half that amount, about 21 percent between 2004 and 2030, because of a greater reliance on nuclear and renewable and a less carbon-intensive fossil fuel mix. Power sector CO2 emissions are expected to be 4.5 percent below the reference case level in 2020 and 15.9 percent below the reference case level in 2030 in the Phased Auction case.
Generation by Fuel
To reduce its CO2 emissions, the power industry, including generators in the industrial and commercial sectors, is expected to shift away from its historical reliance on coal generation (Figure 6). Total coal generation in 2020 is projected to be 135 billion kilowatthours (5.4 percent) below the reference case level in the Phased Auction case. By 2030, coal generation relative to the reference case is 851 billion kilowatthours (25 percent) less in the Phased Auction case. In the reference case, coal accounts for 57 percent of total generation in 2030, but its share falls to 44 percent in the Phased Auction case. While coal generation in 2030 in the Phased Auction case is well below the reference case projection it would still be substantially above the current level, increasing by 28 percent between 2004 and 2030.
The higher coal costs in the Phased Auction case greatly influence the relative economics of new plant alternatives. In the reference case, 174 gigawatts of new coal capacity is projected to be added between 2004 and 2030. In the Phased Auction case, the amount added over the same period is 51 gigawatts. The plant choice results are very sensitive to the allowance price, as indicated by the projections of coal generation across the cases (Figure 7). In the $5 Phased Auction case, projected coal generation in 2030 is 14 percent below the reference case level, compared to 25 percent below in the original Phased Auction case and 31 percent below in the $9 Phased Auction case. Projected coal generation in 2030 grows from its current level in all of the cases analyzed.
While successful development of carbon capture and storage technologies might allow coal-fired plants to remain competitive under a GHG allowance program, the allowance prices in this analysis are not sufficiently high to compensate for the increased capital and operating costs. As a result, power plants using carbon capture and storage are not projected to be commercially viable within the 2030 time frame in the analysis cases.
In contrast to the situation for coal generation, nuclear generation is projected to increase significantly in the Phased Auction case. In the reference case, nuclear generation is projected to increase from 789 billion kilowatthours in 2004 to 871 billion kilowatthours in 2030, as existing plants are upgraded by 3 gigawatts and 6 gigawatts of new capacity, stimulated by incentives in the Energy Policy Act of 2005 (EPACT2005), are added. The 47 gigawatts of nuclear capacity added in the Phased Auction case increases nuclear generation to 1,168 billion kilowatthours. As a result of the additions, the share of generation accounted for by nuclear plants in 2030 increases from 15 percent in the reference case to 20 percent in the Phased Auction case.
Renewable generation is also expected to see significant growth in the Phased Auction case. In the reference case, renewable generation is projected to increase from 358 billion kilowatthours in 2004 to 559 billion kilowatthours in 2030. Part of this growth is stimulated by tax incentives for certain renewable technologies in EPACT2005. In the Phased Auction case, renewable generation is projected to grow to 572 billion kilowatthours by 2020 and to 823 billion kilowatthours by 2030. Most of the increase in renewable generation is expected to be from non-hydroelectric renewable generators, mainly biomass and wind. In the reference case, biomass generation is forecast to rise from 37 to 103 billion kilowatthours between 2004 and 2030. In the Phased Auction case, biomass generation is expected to increase three-fold relative to the reference case to 306 billion kilowatthours. Wind generation, projected to increase from 14 billion kilowatthours in 2004 to 65 billion kilowatthours in the reference case by 2030, is expected to increase to almost twice that amount in the Phased Auction case, where it grows to 108 billion kilowatthours. As a result, the non-hydroelectric renewable share of generation, 2.2 percent in 2004, increases significantly in the Phased Auction case. By 2030, the share grows to 9 percent in the Phased Auction case, more than twice the 4 percent share in the reference case.
Oil and natural gas generation are also impacted by efforts to reduce power sector GHG emissions, but to lesser degrees than coal, nuclear, and renewables. Oil generation, already a very small part of electricity market, falls even further in the Phased Auction case. Relative to the reference case, natural gas generation in 2030 is 20 percent higher in the Phased Auction case, as new combined-cycle plants become more attractive, relative to coal plants, for new baseload capacity.
Electricity Prices
The shift away from coal to increased use of nuclear and renewable fuels, together with the costs of fuel suppliers holding emissions permits, affects electricity prices (Figure 8). The impacts are slightly different in the Phased and Full Auction cases because of the different approaches used to distribute allowances. In the reference case, real electricity prices fall from 7.6 cents per kilowatthour to 7.2 cents per kilowatthour in 2020, and then increase slowly to 7.5 cents per kilowatthour in 2030 as fuel prices rise. In the Phased and Full Auction cases, 2020 electricity prices are, respectively, 4 and 6 percent higher than in the reference case. As the GHG permit price continues to rise between 2020 and 2030 in the Phased and Full Auction cases, the cost of using fossil fuels also continues to grow, contributing to electricity prices that are respectively, 11 and 13 percent above the reference case level in 2030. Electricity prices are slightly lower in the Phased Auction case because a portion of the allowances are given out to power producers for free, lowering the revenue requirements of those producers who are subject to rate regulation. Consumers’ total electricity bills in 2020 in the Phased and Full Auction cases are $10 and $15 billion (2.9 and 4.4 percent), respectively, higher than in the reference case. By 2030, the increase in consumer bills above the reference case level in the Phased and Full Auction cases grows to $34 billion and $41 billion (8.6 and 10.2 percent).
The different regulatory regimes in the various regions of the country do affect the electricity prices in the Phased Auction case. While electricity prices are higher in all regions in both the Phased and Full Auction cases, the price impacts are smaller in the Phased Auction case in regions where prices are set under cost-of-service regulation. For example, in the South Atlantic and Florida regions, 2030 electricity prices are 0.2 cents lower in the Phased Auction case than they are in the Full Auction case. In contrast, in regions where electricity prices are set competitively, the changes relative to the reference case are the same in both the Phased and Full Auction cases.
End-Use Energy Consumption
In response to higher delivered fossil fuel and electricity prices in the Phased and Full Auction cases, consumers and businesses in all sectors of the economy are projected to reduce their energy consumption and, where possible, shift their consumption away from fossil fuels. These changes reduce overall energy consumption, but raise consumers’ energy bills.
Residential and Commercial
Higher fuel prices under the proposed GHG cap and trade program provide an incentive for residential and commercial consumers to use less energy than otherwise. Relative to the reference case, total delivered residential energy consumption in the Phased Auction case is 0.4 percent lower in 2020, and 1.2 percent lower in 2030 (Figure 9). Similarly, for the commercial sector, total delivered energy consumption in the Phased Auction case is 0.8 percent lower in 2020 and 2.2 percent lower in 2030 (Figure 10).
These changes result from consumer responses to higher costs for all fossil fuels and electricity in the Phased Auction case. These costs include the purchase price of the fuels together with the costs of permits needed to cover the GHG emissions associated with their use. For example, relative to the reference case, the average delivered price of natural gas is $0.40 per million Btu (6 percent) higher in 2020 in the Phased Auction case. By 2030, this difference grows to $0.86 per million Btu (11 percent). For distillate fuel oil and electricity, the projected percentage changes in average prices are similar to those for natural gas.
Even with lower energy consumption, households are projected to see higher energy bills because household energy consumption is relatively unresponsive to energy prices. Compared to the reference case, annual per-household energy expenditures in 2020 are 3 percent ($41) higher in the Phased Auction case. By 2030, the difference increases, with annual per household energy expenditures 7 percent ($118) higher in the Phased Auction case.
Where possible, homeowners will increase their use of non-fossil energy. For example, relative to the reference case, the number of homes with solar photovoltaic (PV) systems increases 22 percent in the Phased Auction case by 2030. However, even with a large percentage change, the stock of homes with PV systems remains small. The 22-percent increase results in about 0.1 percent of the homes having PV systems by 2030.
As in the residential sector, the impact of higher energy prices outweighs the impact of reductions in commercial energy consumption, resulting in a $6-billion (3 percent) increase in commercial energy expenditures in the Phased Auction case in 2020, relative to the reference case. The increase in expenditures is greater by 2030, reaching $18 billion (8 percent) higher than commercial sector energy expenditures in the reference case in the Phased Auction case.
Also, as in the residential sector, commercial consumers are expected to increase their use of renewable energy in response to a GHG cap and trade program. In the Phased Auction case, total commercial sector PV capacity is 5 percent higher in 2020 than in the reference case. By 2030, commercial sector PV capacity in the Phased Auction case is 35 percent higher than in the reference case.
The GHG cap and trade program also stimulates commercial users to increase their investments in natural gas-fired combined heat and power plants (CHP). These facilities can be very efficient, and higher fossil fuel prices make investments in them more attractive. Overall, commercial natural gas-fired CHP capacity is 0.6 percent higher in 2020 in the Phased Auction case, when compared to the reference case. By 2030, the increase relative to the reference case increases to 9 percent.
Industrial
Industrial consumers also reduce their energy consumption in response to higher energy prices, particularly their consumption of coal, which includes coal used to produce oil and electricity in coal-to-liquids (CTL) plants. Relative to the reference case, delivered industrial energy consumption in the Phased cases is 2 percent lower in 2020 and 6 percent lower in 2030 in the Phased Auction case (Figure 11). The largest percentage reductions occur in coal used in CTL production and purchased electricity. In the AEO2006 reference case, industrial coal use is projected to grow rapidly in the latter half of the projection as CTL plants are introduced. Under the proposed GHG program policy cases, the cost of coal reduces the economic potential for these plants. Relative to the reference case, total industrial coal use is 15 percent lower in 2020 and 39 percent lower in 2030 in the Phased Auction case. In 2030, the use of coal in CTL plants is lower by nearly 85 percent in the Phased Auction case, and the domestic petroleum supply from CTL plants is about 650 thousand barrels a day lower, compared to the reference case.
Compared to the reference case, purchased electricity consumption in the industrial sector is 1 percent lower in 2020 and 3 percent lower in 2030 in the Phased Auction case.
While energy consumption falls in the industrial sector in the Phased Auction case, total industrial energy expenditures rise. Relative to the reference case, industrial energy expenditures increase by $8 billion (5 percent) in 2020 and $21 billion (10 percent) in 2030, in the Phased Auction case. Industrial output, measured in year 2000 dollars, is also reduced relative to the reference case by $91 billion (1 percent) in 2030 in the Phased Auction case.
Transportation
Responding to higher gasoline, diesel, and jet fuel prices, transportation consumers also reduce their energy consumption under the GHG proposal (Figure 12). Relative to the reference case, the higher prices projected in the Phased Auction case lead to 1 percent lower transportation sector energy consumption in 2020 and 2 percent lower transportation sector energy consumption in 2030.
Lower transportation energy consumption results from a combination of reduced travel and increased purchases of more efficient vehicles. In 2020, the reduction in light-duty vehicle miles traveled from the reference case level is 19 billion miles (1 percent) in the Phased Auction case. By 2030, this difference grows to 46 billion miles (1 percent). Freight truck travel is also slightly lower in the Phased Auction case because of lower industrial output.
Though energy use by railroads accounts for only a small part of overall transportation energy use, projected growth in railroad shipments is expected to be significantly impacted by large reductions in the projected growth of coal use (Figure 13). Relative to the reference case, 2020 rail ton-miles traveled are 50 billion ton-miles (3 percent) lower in the Phased Auction case. With a growing reduction over time in coal use relative to the reference case, by 2030 rail ton-miles are 245 billion ton-miles (10 percent) lower than in the reference case.
Improved fuel economy also contributes to the lower transportation sector energy consumption. The higher fuel prices in the Phased Auction case stimulate consumers to shift away from light trucks and purchase more hybrid and diesel vehicles. However, the increase in gasoline prices in 2030, $0.11 per gallon (2004 dollars), is not large enough to stimulate a significant shift in the mix of vehicles purchased. The changes that do occur are gradual, but by 2030, the percent of new light vehicle sales that are cars increases from 45 percent in the reference case to 46 percent in the Phased Auction case. Sales of hybrid vehicles in 2030 grow from 12 percent of new light vehicle sales in the reference case to 11.5 percent of new light vehicle sales in the Phased Auction case. Because of the shift in vehicle purchases in the Phased Auction case, new light-duty vehicle fuel economy is 0.4 miles per gallon (1 percent) higher in 2030 than in the reference case.
Fuel Supply
Natural Gas
In general, relative to the reference case, total natural gas consumption changes very little in the Phased and Full Auction cases (Figure 14). The change in consumption occurs mainly in the electric power sector, but most other sectors show small changes. Electric power sector natural consumption is projected to be 0.2 quadrillion Btu (2 percent) below the reference case level in the Phased Auction case in 2020 due to reduced total generation requirements. However, by 2030, the pattern reverses itself and electric power sector natural gas consumption is expected to be 0.5 quadrillion Btu (8 percent) higher in the Phased Auction case than in the reference case.
Coal
Because of large reductions in coal use in the electric power sector and in the production of liquid fuels, coal production is much lower in the Phased Auction case (Figure 15). Relative to the reference case, total coal production is 73 million tons (5 percent) lower in 2020 and 383 million tons (22 percent) lower in 2030 in the Phased Auction case. However, even with these changes, coal production in 2030 in the Phased Auction case is projected to be 17 percent (196 million tons) greater than 2004 production. Both eastern and western coal production are lower in the Phased Auction case, but the impact is larger in the west because that is where coal production is projected to grow most rapidly in the reference case.
Petroleum
Relative to the reference case, the consumption of petroleum products is lower in the Phased Auction case, as consumers respond to the higher delivered petroleum product prices that result from cost of allowances under the cap-and-trade program. Petroleum consumption in 2020 is projected to be 0.4 million barrels per day (2 percent) lower in the Phased Auction case than in the reference case. By 2030 the difference grows to 0.7 million barrels per day (3 percent) lower in the Phased Auction case than in the reference case. However, domestic crude oil production is relatively unaffected because the world crude oil prices are unchanged. The reduction in petroleum supply in the Phased Auction case comes from reductions in imports and reductions in domestic CTL production. In the Phased Auction case in 2020, CTL production is 0.2 million barrels per day (74 percent) lower than in the reference case. By 2030, the change is 0.6 million barrels per day (85 percent) lower than in the reference case. The cost of allowances increases the cost of using coal, making CTL production much less competitive with imported and domestic oil.
Economic Impacts
Implementing a GHG emissions cap-and-trade program based on a targeted rate of reduction in emissions intensity in which some emissions permits will be auctioned and others will be sold if the safety valve is triggered will impact the economy through two mechanisms. First, efforts to reduce GHG emissions and the requirement to hold permits for all remaining GHG emissions will raise energy prices, particularly those for fossil fuels. Second, the auctioning of permits and the sale of additional permits if the safety valve is triggered will increase revenues to the government. In turn, higher energy prices and increased government revenues will impact aggregate economic growth.
Government Revenues
Projected government revenue from the allowance program is a function of the market price of the allowances, the number of initial allowances auctioned, and the additional revenue from safety valve fees. The value of allowances allocated for free can be considered a revenue transfer in the sense that recipients will accrue revenue from the resale of these allowances. For simplicity in the following discussion, free allowance allocation from the Federal government to recipients is treated as a revenue transfer.
The auctioning of allowances and their free distribution under the proposal will result in revenue flows to different sectors of the economy. Industry will presumably apply some of the value of allowances allocated to them to implement energy-saving processes and technologies. States will use the funds for a variety of programs, such as addressing economic impacts and promoting technology or energy efficiency. As specified in the proposal, the revenue generated by the Federal government (auction proceeds plus safety-valve fees) is initially used to fund early technology deployment, subject to a $50-billion cap on the maximum cumulative deposits to the trust fund. This limit is projected to be reached in 2017 in the Phased Auction case, after which all remaining revenue flows to the U.S. Treasury and is assumed to be used to retire some part of the Federal debt. Figure 16 shows the flows of these funds among the various sectors for the Phased Auction case. The projected change in the Federal deficit relative to the reference case differs from the amount of allowance revenue allocated to debt reduction due to the impact of the proposed program on the overall economy.
The Full Auction case considers the effects of a policy of letting all of the allowance receipts flow to the Federal government as an alternative to the allocation scheme in the proposal. The Federal expenditure profile for revenue deposited in the Climate Change Trust Fund is assumed to be the same as in the Phased Auction case. The difference in the two cases is the impact on funds flowing to the Federal government and the subsequent rate at which the Federal debt level is lowered (Figure 17). In essence, the Full Auction case draws more money away from the spending stream of the economy and thus lowers aggregate demand to a much larger degree than the Phased Auction case. Some discussion of possible alternative approaches to revenue recycling and their implications is provided in an earlier EIA report requested by Senators Inhofe, McCain, and Lieberman21
Prices
The energy market impacts of the proposed program influence the aggregate economy through the effect on prices and energy expenditures. Figure 18 shows the percentage changes in the consumer price index (CPI) for energy and the All-Urban CPI, a measure of aggregate consumer prices in the economy. The CPI for energy, a summary measure of energy prices facing households at the retail level incorporating the energy price impacts associated with rising petroleum, natural gas, and electricity prices, increases by approximately 8 percent above the reference case level by 2030. Ultimately the consumer sees higher prices directly through final prices paid for energy goods and service, plus higher prices for other goods and services that come about due to changes in the price of other goods and services resulting from energy price changes, as well as changes in interest rates and other prices driven by the flow of revenues to the government and other sectors under the proposal.
In the Phased Auction case, the All-Urban CPI rises steadily and by 2030 is approximately one percent above the reference case. In the Full Auction case, both interest rates and aggregate demand are lower. Interest rates are lower for two main reasons. First, there is less inflationary pressure due to lower aggregate demand and the slightly higher unemployment. Second, with a lower government deficit there is less demand for credit. With less inflationary pressure, the Full Auction variant of the proposal has a lesser effect on the All-Urban CPI than the Phased Auction version. This result implies that energy prices are higher relative to the prices of other goods in the Full Auction case than in the Phased Auction case.
The alternative values of the safety valve price considered in the sensitivity cases will have relatively symmetric impacts on aggregate prices. Figure 19 shows that the All-Urban CPI will rise to approximately 1.2 percent above the reference case by 2030 in the Phased $9 case compared to a 1.0-increase in the Phased Auction case.
Real GDP and Consumption Impacts22
The higher delivered energy prices and the collection of additional government revenues lower real output for the economy in both the Phased and Full Auction cases. They reduce energy consumption, but also indirectly reduce real consumer spending for other goods and services due to lower purchasing power. The lower aggregate demand for goods and services in the both the Phased and Full Auction cases results in lower real GDP relative to the reference case (Figure 20). Relative to the reference case, total discounted GDP over the 2009 to 2030 time period is $232 billion (0.10 percent) lower in the Phased Auction case and $462 billion (0.19 percent) lower in the Full Auction case. Projected GDP impacts generally increase over time, as the cap-and-trade program requires larger changes in the energy system. Relative to the reference case, real GDP in 2030 is $59 billion (0.26 percent) lower in the Phased Auction case and $94 billion (0.41 percent) lower in the Full Auction case. Because the additional impact on economic activity under a Full Auction could be significantly, or even fully, mitigated under alternative revenue recycling assumptions, the results for the Full Auction case presented here should not be construed as suggesting a general conclusion that a Phased Auction will necessarily result in lesser impacts on GDP than a comparable Full Auction.
The alternative values of the safety valve price considered in the sensitivity cases have relatively symmetric impacts on projected GDP losses. The estimated loss in GDP in 2030 is 0.32 percent in the Phased $9 case, compared to 0.26 percent for the Phased Auction case and 0.16 percent for the Phased $5 case. In terms of cumulative GDP losses, the difference between the Phased Auction and Phased $9 cases is much smaller, 0.10 percent compared to 0.11 percent, reflecting the fact that the safety valve does not become binding until after 2025 in the Phased Auction case, so that economic impacts up to that date follow the same path in either case. Under the Phased $5 case, the safety valve comes into play at an earlier date, so there is a longer period of time over which projected economic impacts differ from those in the Phased Auction case.
While real GDP is a measure of what the economy produces, ultimately consumers are interested in their purchases of goods and services. GDP and consumption impacts of a proposed policy can differ if the policy leads to changes the shares of the GDP components, which include consumption, investment, government expenditures, and net exports, as well as the level of GDP. Figure 21 shows two measures of consumption impacts: the change in consumption relative to the reference case and the cumulative discounted loss in consumption over the 2009 to 2030 period. Cumulative discounted consumption losses relative to the reference case are $236 billion (0.14 percent) in the Phased Auction case and $483 billion (0.29 percent) in the Full Auction case. Consumption impacts, like GDP impacts, generally grow over time. In 2030, projected real consumption in the Phased Auction and Full Auction case is, respectively, $55 billion (0.36 percent) and $106 billion (0.69 percent) below the reference case level. Starting from the Phased Auction case in which the safety valve is binding before the end of the modeled time horizon, a lower initial value for the safety valve lowers estimated consumption losses, while a higher safety valve raises them (Figure 21). In terms of cumulative discounted consumption losses over the 2010 to 2030 period, the impact of the $5 Phased Auction and $9 Phased Auction sensitivity cases on consumption losses is somewhat asymmetric relative to the $236 billion estimated cumulative discounted consumption loss in the Phased Auction case. The $5 Phased Auction case lowers the estimated cumulative discounted consumption loss to $161 billion (0.10 percent), while the $9 Phased Auction case raises it to $277 billion (0.17 percent).
Uncertainty
All long-term projections engender considerable uncertainty. It is particularly difficult to foresee how existing technologies might evolve or what new technologies might emerge as market conditions change, particularly when those changes are fairly dramatic. This analysis suggests that, to comply with the GHG emissions growth limits necessary to meet the intensity reduction targets, all energy providers, particularly electricity producers, will increasingly rely on technologies that play a relatively small role today or have not been built in the United States in many years. Sensitivity analyses included in previous EIA studies of cap-and-trade systems for GHG show that estimates of both energy and economic impacts of such programs can change significantly under alternative assumptions regarding the cost and availability of new technologies.
Non-hydroelectric renewable generators currently provide 2.2 percent of the electricity generated. In the reference case, their share is expected to grow to 4.3 percent in 2030. In the Phased and Full Auction cases their share grows to 9 and 10 percent of generation by 2030. While this level of growth is certainly possible, particularly since the GHG emission targets are tightened gradually, it comes with some uncertainty. It is possible that such growth might lead to significant reductions in the costs of these technologies. On the other hand, it is also possible that costly hurdles such as siting resistance, higher than expected transmission interconnection costs, or fuel supply limits could arise that limit their development.
Similarly, this analysis suggests that the power sector would significantly increase its reliance on nuclear power in order to reduce GHG emissions. This is despite the facts that the last nuclear order in the United States was placed in 1978 and the last nuclear plant to enter service began operating in 1996. However, several factors, including rising fossil fuel prices, concern about GHG emissions, tax incentives in the EPACT 2005 and new nuclear plant designs, have recently spurred renewed interest in new nuclear plants. In the reference case, nuclear capacity is projected to increase by 9 gigawatts, including 3 gigawatts of uprates at existing plants and 6 gigawatts of new nuclear plants, about 4 to 6 new plants. In the Phased and Full Auction cases, nuclear capacity is projected to grow by 48 gigawatts and 46 gigawatts, respectively. Such growth in nuclear power might lead to significant cost reductions, encouraging more expansion than projected. On the other hand, costly hurdles, such as unexpectedly high construction costs, public resistance to the siting of facilities, or waste disposal concerns, could arise to limit their development.
If the development of these technologies is limited for one reason or another, power providers will have two choices. First, they can turn to other low-GHG or non-GHG technologies, such as new fossil generators with carbon capture and sequestration equipment, that play a fairly small role in today’s market. Second, they can comply by paying more safety-valve fees to maintain their reliance on current fossil-fired generation. To the extent this occurs, projected reductions in GHGs would be reduced. One way or another, significantly reducing energy-related GHG emissions would require a shift away from fossil energy sources that accounted for 86 percent of U.S. energy consumption in 2004. The costs of such a shift are inherently uncertain.
Particularly uncertain in this analysis is the role that increased research and development (R&D) expenditures might play in spurring the development and deployment of new more efficient, lower emitting technologies. The draft proposal calls for spending significant resources on R&D, but it is impossible to predict the impact of such expenditures.
A final source of uncertainty involves assumptions regarding the availability of reductions in covered GHG emissions outside the energy sector. To the extent that this analysis overstates the availability of such reductions, additional reductions in emissions within the energy sector or additional purchases of allowances at the applicable safety valve price would be required to comply with the proposed program. Previous studies have explored the sensitivity of energy and economic impact estimates to alternative estimates of available emissions reductions outside the energy sector.
Notes
Table 2 
Table 3
|
