Executive Summary

Background

Senator James M. Inhofe requested that the Energy Information Administration (EIA) undertake analysis of S.843, the Clean Air Planning Act of 2003, introduced by Senator Thomas Carper; S.366, the Clean Power Act of 2003, introduced by Senator James Jeffords; and S.1844, the Clear Skies Act of 2003, introduced by Senator James M. Inhofe. The EIA received this request on March 19, 2004. This Service Report responds to his request.

The emissions targets and implementation timetables for the bills are summarized in Table ES1. All three bills implement emissions targets on power sector emissions of nitrogen oxides (NO_x), sulfur dioxide (SO₂), and mercury (Hg). The Clean Air Planning Act and the Clean Power Act also call for limits on power sector carbon dioxide (CO₂) emissions. Under the Clean Air Planning Act, greenhouse gas emission reductions outside of the power sector, referred to as offsets, can be used to meet the emission targets for CO₂.

All three bills cover emissions from larger generators that generate power for sale. This includes central station generators and generators at customer sites that sell power they do not use for their own needs. The Clear Skies and Clean Air Planning Acts cover generating facilities 25 megawatts and larger, while the Clean Power Act covers facilities 15 megawatts and larger. The bills have differing provisions regarding the coverage of combined heat and power facilities that generate some power for sale.

The bills generally rely on emissions cap and trade programs to achieve the required reductions. Under such programs, allowances will be allocated and covered generators will have to submit one allowance for each unit of emissions they produce. However, for mercury, the Clean Air Planning Act combines a minimum removal target for all plants with an emissions cap, and the Clean Power Act specifies a maximum emissions rate for all facilities and allows no trading of mercury allowances. The Clear Skies Act contains a “safety valve” feature that caps the price that power companies would have to pay for Hg ($2,187.50 per ounce or $35,000 per pound), SO₂ ($4,000 per ton), and NO_x ($4,000 per ton) allowances. Should one or more of these “safety valves” be triggered, the corresponding cap on emissions would effectively be relaxed.

Under the Clear Skies Act, emission allowances are to be allocated based on historical fuel consumption, what is often referred to as “grandfathering.” Under the Clean Air Planning Act, a grandfathering approach is used to allocate emission allowances for SO₂, but allowances for NO_x, Hg, and CO ₂, are allocated using an output-based scheme. Under this approach, referred to as a generation performance standard (GPS), generators are given allowances for each unit of electricity they generate. The number of allowances allocated for each unit of generation changes each year as the total generation from covered sources changes. The use of a GPS dampens the electricity price impacts of the bill but raises overall compliance costs.

In addition to the emission caps, the Clean Power Act also requires that all plants have the best available control technology (BACT) beginning in 2014 or when they reach 40 years of age, whichever comes later. This provision, often referred to as a “birthday” provision, requires older plants to add controls even if the total emissions of covered facilities are below the emission caps.

Methodology

This analysis was prepared using EIA’s National Energy Modeling System (NEMS). The reference case used in this report is based on the reference case in the Annual Energy Outlook 2004¹ , and it incorporates final regulatory action under existing laws. However, consistent with standard EIA practice requiring policy neutrality in baseline projections, it does not include pending or proposed actions, such as the maximum achievable control technology (MACT) standards for mercury emissions from power plants or actions that might be taken to comply with the revised National Ambient Air Quality Standards for ozone and fine particulates. The implementation of such actions could affect emissions, generator costs, and electricity prices during the projection period even if there is no additional new legislation. In addition, the potential benefits that might be associated with emissions reductions are not discussed. EIA does not have expertise in the area of health benefits that might be associated with emissions reductions.

The cases prepared in this analysis simulate the response of the economy to changing fuel prices and demands. However, recent information suggests that natural gas intensive industries may be more sensitive to higher natural gas prices than is reflected. Should this be true, the costs to the power sector of complying with the three bills could be lower since reduced industrial sector natural gas use would lower the pressure on natural gas markets, making it more economical for the electricity sector to use natural gas.

Throughout this report the generation and capacity data reported are for all generators, including small generators that are not covered by the emission limits. The emissions data shown are for the electric power sector, which includes all generators whose primary business is to produce and sell electricity.

Analysis of the Three Bills

Clear Skies Act (Inhofe): To comply with the provisions of this bill, power generators are expected to rely primarily on adding emissions control equipment to existing generators. Switching fuels from coal to natural gas and renewables is projected to play a relatively small role. Power generators are expected to reduce their mercury emissions prior to 2010 to take advantage of the early credit program. However, the use of early credits allows them to delay meeting the 2010 34-ton mercury emissions cap until 2013. In the longer term, because of the mercury safety valve, mercury emissions are projected to remain above the 15-ton emission target that takes effect in 2018 throughout the projections. SO₂ emissions are projected to approach the target, but because of allowance banking in the early phases, the 3-million-ton cap is not reached by 2025. The resource cost (the cost to the electric generation industry) and the electricity price impacts are the lowest among the three bills considered.

Clean Air Planning Act Bill (Carper): The addition of emissions control equipment to existing generators is also expected to play an important role in complying with this bill. However, because of the tighter emissions limits on SO₂, NO_x, and Hg and the addition of a CO₂ emissions cap, fuel switching from coal to natural gas and renewables is projected to be much more important than under the Clear Skies Act. The impacts are very sensitive to the availability and cost of greenhouse gas offsets. The Clean Air Planning Act calls for the establishment of an independent review board to evaluate potential greenhouse offsets, but the criteria they might use are uncertain. Because of this uncertainty two separate cases were prepared. One case, Carper Domestic, assumes that only domestic offset programs will be approved, while another, Carper International, assumes both domestic and international offsets will be available. These cases illustrate the sensitivity of the results to the cost and availability of greenhouse gas offsets, but may not span the full range of possible outcomes.

If greenhouse gas offsets are fairly inexpensive, they will be the primary option for meeting the CO₂ emissions limit. The more expensive are greenhouse offsets, the larger will be the role of switching fuels from coal to natural gas and renewables. The output-based allowance allocation scheme used in the Clean Air Planning Act does dampen the electricity price impacts, but it leads to higher resource costs. Overall, the resource cost and electricity price impacts of this bill are projected to be larger than those under the Clear Skies Act.

Clean Power Act (Jeffords): Under this bill the relatively stringent CO₂ emissions cap is projected to make switching from coal to natural gas, renewables, and nuclear especially important. The birthday provision causes many older plants to add emissions control equipment, even though the emissions of SO₂, NO_x, and Hg are projected to fall below their respective targets once power companies switch away from coal. A large number of new small generators and combined heat and power facilities are built because they are not covered by the bill’s emissions caps. However, the bill reduces the emission caps for large generators to offset emissions from these sources. The early timing and stringency of the emissions limits combined with the birthday provision in this bill lead to the largest resource cost and electricity price impacts among the three bills. Because of the higher projected electricity prices, consumers are also expected to reduce their use of electricity.

The differences in the three bills can best be seen by comparing their respective impacts on emissions; coal, natural gas, renewable, and nuclear generation; electricity prices; and resource costs. For NO_x, power sector emissions in the Inhofe and Carper cases are projected to fall to the respective bill targets because the phase 1 and 2 emission targets are so close that there is little opportunity for economical allowance banking. Only in the Jeffords case, where the birthday provision requires all older plants to add emissions controls when they reach 40 years of age, are NO_x emissions projected to fall below the bill’s emission target.

For SO₂, power sector emissions are projected to fall in all the cases, including the reference case (Figure ES1). However, in the Inhofe and Carper cases, SO₂ emissions are projected to remain above the bills’ target levels because of allowances banked from the existing SO₂ reduction program. As with NO_x emissions, in the Jeffords case, SO₂ emissions are projected to fall below the bill’s emission target.

For Hg, power sector emissions are projected to remain above the 2018 target level in the Inhofe case throughout the projection period (Figure ES2). In the early years, this is due to power companies taking advantage of the early credit program to bank allowances prior to the beginning of the required reductions in 2010. The above-target-level emissions in the later years are caused by the mercury allowance price safety valve. In the Carper cases, without a mercury allowance price safety valve, power sector Hg emissions are expected to fall to the required target level. In the Jeffords case, similar to NO_x and SO₂, power sector Hg emissions are expected to fall below the bill’s target level. This is caused by the combination of reduced coal use and the facility-specific mercury emission limit in the Jeffords bill.

For CO ₂, power sector emissions are projected to increase over time in all the cases, except the Jeffords case (Figure ES3). In the Carper cases, power companies are expected to rely heavily on greenhouse gas offsets outside the covered sector to meet the CO ₂ emissions target. The purchase of greenhouse gas offsets accounts for between 46 percent and 64 percent of the overall CO ₂ emission reductions required in the two Carper cases in 2025. In the Jeffords case, power sector CO ₂ emissions are projected to gradually fall below the level because the covered generators’ emissions limit is adjusted for the growing emissions from small generators in the industrial and commercial sectors who sell power to the grid.

All three bills are projected to lead to lower coal generation than in the reference case (Figure ES4). The change in the Inhofe case is relatively modest, 5 percent below the reference case in 2025. In the two Carper cases, the impact on coal generation is larger, falling as much as 24 percent below the reference case level in 2025. The reduction in coal generation is projected to be most pronounced in the Jeffords case where it is 55 percent below the reference case in 2025.

In contrast to coal generation, natural gas and renewable generation are projected to be higher under the three bills (Figures ES5 and ES6). Again, in the Inhofe case, the impact on natural gas and renewable generation is projected to be modest. In the long run, the shift to natural gas is projected to be largest in the Carper cases, due to the less stringent CO₂ emissions cap and the availability of greenhouse gas offsets. In the Jeffords case, the relatively stringent CO₂ emissions cap is expected to result in large increases in the use of non-fossil fuels, especially in the longer term. In the near term, before non-fossil technologies such as advanced nuclear and biomass plants are projected to be available, power generators are projected to turn to natural gas to comply.

The Jeffords case is the only case where new nuclear plants are projected. With CO₂ allowance prices of at least $29 per metric ton ($108 per metric ton carbon equivalent) throughout the projection period, new nuclear plants are projected to be economical in the Jeffords case once they become available. However, because the first new nuclear plant is expected to take 10 years to plan, permit, and construct, their impact is not expected to be large until the later years of the projections. By 2025, nuclear generation is projected to be 53 percent above the reference case level in the Jeffords case.

Relative to the reference case, electricity prices in the Inhofe case are projected to be 3.2 percent higher in 2025 (Figure ES7). In the Carper cases, electricity prices are expected to be as much as 7.8 percent higher in 2025. The electricity price impacts are expected to be much larger in the Jeffords case, 47 and 27 percent above the reference case levels in 2010 and 2025, respectively. The high near-term impact results from the need to rapidly transform the industry from using coal to natural gas, renewables, and nuclear.

The projected change in power sector resource costs, the amount that power companies spend on fuel, capital, and operations and maintenance, tend to follow a pattern similar to that for electricity prices. Over the 2005 to 2025 time period, discounted resource costs are projected to be 1.3 percent higher in the Inhofe case. For the Carper cases, they are projected to be between 2.9 percent and 4.5 percent higher, while in the Jeffords case they are 19.5 percent higher.

Uncertainties

As with any long-term projection, there are considerable uncertainties. It is impossible to predict future fuel prices and how existing generation or emissions control technologies might evolve in cost and performance or what currently unknown technologies might emerge to play unexpectedly important roles in the market. Of particular concern in this analysis are future natural gas prices, the availability and market acceptance of low- or zero-carbon generation technologies, including new nuclear, renewable, and fossil plants with carbon capture and sequestration equipment, the availability and cost of greenhouse gas offsets, and the cost and performance of emerging mercury removal technologies.

One only has to look at the behavior of natural gas prices over the past several years to observe the volatility and uncertainty in the market. Current prices are much higher than they were just a few years ago and, if they remain high, the costs of shifting from coal to natural gas in response to the three bills could be more expensive, particularly in the Carper and Jeffords cases where fuel switching is expected to be a more important compliance option. With higher natural gas prices, other low-carbon generating technologies such as new renewables and nuclear would likely play an increased role. On the other hand, if industrial natural gas users reduce their consumption more aggressively in response to higher natural gas prices, the compliance costs on the power sector could be smaller than estimated.

The potential availability and cost of greenhouse gas offsets is an important area of uncertainty when analyzing the impacts of the Carper bill. There is uncertainty both about what offsets might cost and what sorts of rules and regulations the independent review board called for in the Carper bill would establish for acceptable international trading programs and offset projects. Relatively lenient rules could make offsets less costly, but they could also make it difficult to ensure that emissions reductions or increases in sequestration are occurring.

With regard to mercury control, there have been few full-scale demonstrations of some of the plant configurations that are necessary to meet the requirements of the proposed bills. This is particularly true for the lower-rank coals, subbituminous and lignite. While technologies that remove SO₂ and NO_x have shown great promise in removing mercury from bituminous coals, they have not performed as well with the lower-ranked coals. Supplemental fabric filter systems using activated carbon injection are expected to be a key technology in removing mercury. However, tests of such systems on plants using subbituminous or lignite coals are only now being evaluated.

A key result of the Carper and Jeffords cases is that the power sector is going to increasingly rely on technologies such as wind and biomass, that currently play a relatively small role in the U.S. generation market, or technologies like nuclear that have not been expanded in many years to comply with the CO₂ emission caps. Such a transformation would clearly be challenging, especially in the timeframe called for in the Jeffords case.

Notes and Sources