Testimony of David G. Hawkins
Director, NRDC Climate Center
Natural Resources Defense Council
Hearings on S. 385, “Clear Skies Act of 2003”
U.S. Senate Committee on Environment & Public Works,
Subcommittee on Clean Air, Climate Change, and Nuclear Safety
April 8, 2003
Mr. Chairman and members of the Subcommittee, thank you for inviting me to testify on behalf of NRDC, the Natural Resources Defense Council, and its more than 500,000 members regarding S. 385, the administration’s bill to amend the Clean Air Act. We have examined the administration proposal and we conclude it would harm public health, weaken current pollution fighting programs and worsen global warming.
In my testimony today, I will emphasize three major policy failures in the administration’s bill. S. 385 would do the following to our nation’s clean air program:
NRDC supports good legislation to amend the Clean Air Act. We worked with the first Bush administration in 1989 and supported the cap and trade program that was enacted in the 1990 amendments to the law. But the current administration proposal is not good legislation: it would take two enormous steps backward and fail to take a critical step forward on global warming.
Fortunately, your choices are not limited to accepting the administration plan or taking no action. As Senators you have the right to ask the administration to explain the policy choices in the bill it has sent to you. If you conduct a thorough inquiry into the administration plan and alternatives we believe you will conclude that the administration bill should not become law. Congress can do much better and the public deserves much better. We want to work with you to deliver that better solution to the public.
Air pollution from power plants imposes a staggering toll of death, disease, and environmental contamination on the American people.
• Sulfur dioxide (SO2) and nitrogen oxides (NOx) emissions from power plants create dangerous concentrations of fine particles and ozone (soot and smog) in the air that 175 million people breathe. Soot and smog caused by power plant emissions is causing 30,000 premature deaths, hundreds of thousands of asthma attacks, and millions of days of illness and lost work each year.
• Mercury emissions from power plants fall from the air and wash into lakes, rivers, and coastal waters, where they concentrate in fish. Mercury is a potent brain poison (neurotoxin) even in very small amounts. Forty-four states have issued warnings against eating local fish because of mercury contamination.
Power plant pollution is causing a major, ongoing public health crisis. The Bush administration’s proposed air pollution plan fails to stem this crisis.
By any comparison, the administration plan allows power plant owners to continue an unacceptable and unjustifiable toll of preventable death and illness. Measured against alternative legislative proposals – including an alternative proposal developed within the administration itself in 2001 – the administration plan would result in more than 100,000 additional early deaths and millions of additional asthma attacks and other illnesses between now and 2020. The administration plan also would result in hundreds of tons more mercury released into the atmosphere over this period. The same conclusions emerge when the administration plan is measured against faithful enforcement of the current Clean Air Act.
The administration rejected an alternative proposal, developed by EPA in August 2001, that would have dramatically reduced this toll of death, illness, and environmental contamination. When expressed in monetary terms, the benefits of the EPA proposal dwarf its cost. But the administration has submitted to you the much weaker plan found in S.385. By sending you its weaker plan, the administration is asking you to vote for a program that saves power plant owners $3.5 billion per year in pollution control costs but imposes at least $61 billion per year in additional avoidable health costs on the American people.
In our view, the administration owes you – and the American people – some straight answers to these questions:
• Why should the public accept the enormous toll of preventable death and illness from power plant pollution that will still occur under the administration’s plan?
• Why should Americans suffer tens of billions of dollars each year in health costs that could be avoided at a fraction of that cost?
• Why don’t the American people have a right to expect – whether from the current Clean Air Act or any new legislation – much deeper and quicker reductions in power plant pollution than the administration plan would provide?
The Administration Plan vs. the EPA Proposal
The original “Clear Skies” proposal was developed in 2001 by the Environmental Protection Agency. The EPA proposal was developed with the goal of delivering at least as much clean-up of sulfur, nitrogen, and mercury emissions as required under the current Clean Air Act with the purported advantages of a cap and trade program. Unfortunately, while the administration’s plan before you today keeps the title, the program has been converted from one that speeds clean air to one that shields power plant owners from faster cuts in their pollution.
After intense lobbying by power plant owners, the White House rejected the targets and timetables in the EPA proposal and proposed a plan allowing much greater continuing pollution from this industry. The larger pollution loads allowed by the administration’s CSI plan are summarized in the following chart:
Bush Administration Plan (CSI) vs. EPA Proposal (Aug. 2001)
Sulfur Dioxide (SO2)
Nitrogen Oxides (NOx)
2 million tons in 2010
1.9 million tons in 2008
1.25 million tons in 2012
24 tons in 2008
7.5 tons in 2012, with 70% facility-specific reduction
4.5 million tons in 2010
3 million tons in 2018
2.1 million tons in 2008
1.7 million tons in 2018
26 tons in 2010
15 tons in 2018
The differences in the amount of pollution allowed by these two plans, both year by year and cumulatively out to 2020, are huge. The administration plan would result in 42 million tons more pollution than the EPA proposal: For SO2, 18 million tons more through 2012 and 34 million excess tons out through 2020. For NOx, 3 million tons more through 2012 and 8 million excess tons out through 2020. The administration plan would also allow 58 tons more mercury through 2012 and 163 tons more out through 2020.
• Health Consequences of Administration Plan’s Excess SO2 and NOx . The administration’s plan means large numbers of Americans will continue to die prematurely or suffer illness caused by the excessive pollution power plants would continue releasing under the plan. Figure 7 shows EPA’s estimates of the additional premature death toll and illness in 2020 under the administration plan:
• 7100 additional premature deaths
• 4600 additional chronic bronchitis cases
• 5100 additional hospital stays and ER visits
• 7 million additional days of respiratory illness
While EPA has not presented the cumulative additional premature deaths and illnesses allowed under the administration plan, those numbers are even larger. Using EPA methods, the Clean Air Task Force calculates that between 2008 and 2020, the administration’s plan would allow more than 100,000 additional premature deaths and would allow millions more asthma attacks and other illnesses.
While we have fundamental concerns about attempting to reduce human death, illness, and misery into dollars, it is important to note EPA’s estimate of the costs the administration’s proposal would impose on the public in monetary terms. EPA analyses show that the administration proposal would result in $61 billion more in premature death and disease costs per year by 2020 than the EPA August 2001 proposal. While the EPA August 2001 proposal would cost industry $3.5 billion dollars more per year to implement, it would achieve over $15 in health benefits for every clean-up dollar spent. For some as yet unexplained reason, the administration chose a plan that would inflict an additional $61 billion a year in health damages on the public in order to save power plant owners $3.5 billion in compliance costs. See Figure 8.
The additional pollution from power plants under the administration plan will leave dozens of cities and counties out of attainment of the national ambient air quality standards for fine particles (soot) and ozone (smog) – the Clean Air Act’s bedrock measure of public health protection.
• The administration’s analysis shows that its plan would leave 107 counties – home to 77 million Americans – in violation of these public health standards in 2010. 64 counties with 60 million residents would remain in violation even in 2020 after the plan’s delayed second-phase requirements in 2018.
• The stronger power plant emission curbs in EPA’s 2001 proposal would bring 85 percent of eastern counties with unhealthy soot levels into compliance with the fine particle standard, and 90 percent of eastern counties with unhealthy smog levels into compliance with the ozone standard. Greater power plant pollution reductions would reduce population exposure in the remaining counties and make it substantially easier for them to reach the health standards with reasonable controls on other sources.
Health Consequences of Administration Plan’s Excess Mercury. Coal-burning power plants are the largest industrial source of mercury air pollution, and the only one still not subject to clean air safeguards. Mercury emissions from power plants fall from the air and wash into lakes, rivers, and coastal waters, where they concentrate in fish. Mercury is a potent neurotoxin even in very small amounts. Forty-four states have issued warnings against eating local fish because of mercury contamination.
Methylmercury (the form of mercury that is absorbed in tissue) is highly toxic, interfering with the development and function of the central nervous system. Infants can ingest methylmercury from breast milk when mothers have eaten contaminated fish. Children who eat such fish are exposed that way as well. Children and infants are at higher risk of mercury poisoning because their nervous systems continue to develop until about age 14. Health effects linked to prenatal methylmercury exposure include:
· poor performance on tests of attention and language
· impaired memory
· inability to process and recall information
· impaired visual and motor function
One in 12 women of childbearing age has mercury levels above EPA’s safe health threshold, according to a Centers for Disease Control and Prevention report published in January 2003 and a Journal of the American Medical Association published last week. Nationally, this translates into nearly 4.9 million women of childbearing age with elevated levels of mercury from eating contaminated fish and more than 300,000 newborns at risk of neurological impairment from exposure in utero.
An estimated 60,000 children are born each year at a significantly increased risk of adverse neurological effects from mercury and current exposure levels increase the number of children “who have to struggle to keep up in school and who might require remedial classes of special education,” according to the National Academy of Sciences. Eating mercury-tainted fish also can harm cardiovascular and immune systems in adults.
Every ton of mercury emissions from power plants adds to the cumulative and persistent mercury loadings in our lakes and streams. EPA’s IPM runs show that under the administration plan, power plants would add loadings of 163 tons more mercury through 2020 than under the EPA proposal. Because mercury is an accumulative toxin, these added tons will do their damage for scores of years after they are released.
The cumulative and persistent nature of mercury contamination underscores the need to minimize emissions from all sources. The weak and delayed mercury provisions in the administration plan will result in a failure to apply technology capable of removing power plants as a significant source of mercury. This means more continued mercury pollution not only from U.S. sources but from power plants world wide. As we have demonstrated with another brain poison, lead, when the U.S. shows leadership in advancing policies to cut dangerous pollution, the rest of the world follows. The United Nations has identified global mercury pollution as a priority but the signal by the administration plan is that the U.S. will go slow. This will almost certainly translate into a global go-slow approach, meaning higher mercury emissions from the rest of the world continuing to be deposited in the U.S..
The Administration Plan vs. the Clean Air Act
It is also appropriate to assess whether the administration plan would deliver more pollution reduction than the current Clean Air Act, or less. In fact, the administration plan would result in millions of tons more pollution than faithful enforcement of the current law.
SO2 and NOx: The administration claims that its plan would reduce SO2 and NOx emissions by 35 million tons more than the current Clean Air Act through 2012. In fact, just the opposite is true: compared to enforcing the current law, the administration plan actually would allow major increases in SO2 and NOx in the next ten to fifteen years.
What accounts for these different assessments of the administration plan? The secret is in the administration’s yardstick. The administration is comparing its proposal with a misleading “baseline” that expressly assumes EPA does not enforce the Clean Air Act. EPA Assistant Administrator Holmstead has candidly called this the “Rip Van Winkle scenario.” The Rip Van Winkle scenario includes only the power plant pollution limits that are on the EPA books at this moment – principally the SO2 reductions already required by the Title IV 1990 acid rain program and NOx cuts ordered under the “NOx SIP call” in 1997. The Rip Van Winkle scenario assumes that EPA goes to sleep, doing nothing more for more for a decade.
But the existing Clean Air Act requires much more than that. It requires the states and EPA to bring our cities and counties into compliance with the national ambient air quality standards for fine particles and ozone (soot and smog pollution) before the end of this decade, unless accomplishing that task can be shown to be not possible. EPA concedes that meeting health standards will require steeper and faster reductions in power plant SO2 and NOx emissions than assumed in the Rip Van Winkle scenario or required by the administration plan.
Undr the current Clean Air Act the pathway for meeting public health standards begins with the designation of which cities and counties across the country do not attain the standards, based on several years of pollution measurements. These “nonattainment designations” will take place in early 2004 for ozone and by the end of 2004 for fine particles.
The Clean Air Act then requires the states and EPA to implement the emission reduction measures needed to meet these public health standards within five years of these designations, by 2009, or sooner if feasible – the law says “as expeditiously as practicable”. Deadline extensions are allowed only if a state rigorously demonstrates that pollution control measures to meet health standards on time are not available or feasible. No such demonstrations have been made. While the states have primary responsibility to address local pollution sources, the Clean Air Act gives EPA special responsibility for interstate pollution that interferes with attainment of the health standards in areas downwind. EPA is required to order pollution reductions from upwind power plants where needed to bring areas in downwind states into timely compliance with the health standards.
Administration spokesmen refuse to say how much reduction in power plants’ SO2 and NOx pollution is needed under current law to meet the public health standards on time. But EPA’s analyses of the administration plan show that it will not result in sufficient clean-up to attain the standards on the schedule required by current law.
As I have already noted, the weak emission reductions in the administration’s plan leave 107 counties in nonattainment in 2010 and 64 counties in 2020. As mentioned above, current law requires attainment by 2009 or sooner unless this schedule is not practicable. But the administration has presented no analysis arguing that its 2018 schedule for completing SO2 and NOx reductions is the fastest practicable schedule. Indeed, analyses on EPA’s web site show that while there may be some labor constraints between now and 2005, those constraints disappear well before 2010.
In short, current law requires more reductions sooner than the administration plan and EPA’s analyses show deeper and faster reductions are feasible.
The administration may claim to be making a 35 million ton advance over the Rip Van Winkle scenario, but that is a phony measure. In fact an objective reading of the current law and EPA’s analyses show that the administration plan would result in far more power plant SO2 and NOx air pollution compared with enforcing the existing Clean Air Act
Mercury: The Clean Air Act also requires faster and deeper reductions of mercury than the administration plan. The current Act requires each mercury-emitting power plant to cut its emissions by installing the maximum available control technology (“MACT”). EPA must issue mercury standards in 2004. Compliance with the MACT standard is required three years later, at the end of 2007. Given the extreme toxicity of mercury, the current law does not permit emissions trading between mercury-emitting sources.
Because EPA has not yet issued the MACT standard, the administration argues that no one can say how much mercury it will allow. Mercury MACT controls on other sources, however, provide a good indication of what is feasible. The MACT pollution controls on municipal and medical waste incinerators, for example, eliminate at least 90 percent of these sources’ mercury emissions.
In December 2001, EPA told the Edison Electric Institute, the power sector’s main trade association, that an equivalent MACT standard for power plants would reduce mercury emissions from 48 tons to 5 tons nationwide by the end of 2007. In the regulatory development process now underway, EPA is evaluating performance requirements that would achieve a reduction to 5 tons per year. The weakest option being analyzed by the agency (at the request of the utility industry) is a level only slightly higher than the administration plan’s second-phase target of 15 tons.
No one, including the administration, has contended that a standard as weak as the administration plan’s first-phase target – 26 tons – could pass muster under the MACT requirement of current law. Even a MACT standard that reduced emissions by only 70 percent would cut mercury pollution to 15 tons 10 years earlier than the administration legislation.
The administration plan, however, would repeal the MACT requirement and delay any initial mercury reductions to 2010. After that, the administration plan would allow 26 tons per year from 2010 through 2017, and 15 tons every year thereafter. Compared to the 5-ton level, the administration plan would allow more than five times as much mercury pollution through 2017, and three times as much each year after. From 2008 through 2020, that would be 284 tons more cumulative loading of mercury under the administration plan.
The administration bill takes with one hand while it also takes with the other. In addition to allowing more pollution than needed to protect public health or allowed by current law, the administration bill repeals or weakens each of the specific programs and requirements in the current Clean Air Act that are effectively reducing power plant pollution today and that will reduce it further tomorrow.
The repealers, exemptions and weakening provisions in the administration bill do great damage to fundamental precepts of the Clean Air Act that have helped deliver cleaner air for over thirty years.
· The current law requires clean up of polluted areas as quickly as practicable but the administration plan would grant automatic delays to 2015.
· The current law requires new sources locating in polluted areas to meet state-of-the-art pollution standards and avoid making existing health problems worse but the administration plan would exempt all sources (even those not covered by any cap) from those requirements until 2015, allowing more than a decade’s worth of new pollution sources to make air quality worse.
· The current law gives states victimized by interstate pollution effective rights to remedy that pollution but the administration plan makes those remedies ineffective against power plants and prohibits any reductions from power plants under these provisions until 2012.
· The current law requires new and modified power plants to limit pollution increases to avoid turning clean air areas into polluted areas but the administration bill repeals this safeguard except for a narrow 30 mile circle around certain national parks and wilderness areas.
· The current law requires new and modified power plants to meet up to date emission performance standards to protect areas with clean air but the administration repeals this safeguard for nearly all existing plants and replaces it with a more polluting performance standard for new plants.
· The current law requires EPA to adopt rules to minimize toxic pollution from power plants but the administration bill repeals most of those requirements and replaces them with a weak performance requirement for mercury that is delayed ten years from the current law’s schedule.
The administration defends all of these dismantling provisions as eliminating programs that are not required since its plan establishes national caps for certain power plant pollutants. But the current administration ignores what the first Bush administration recognized—that national caps cannot protect local air quality and must not override the tools that are in the law to protect communities from pollution increases that harm local air quality. Neither the first Bush administration nor congress sought to repeal the tools that protect local air quality when the acid rain cap program was enacted in 1990. Repeal of those tools is no more justified now.
Delaying Attainment of Public Health Standards. Section I of this testimony sets forth EPA’s legal obligations under the current Clean Air Act to assure the attainment of the national ambient air quality standards for fine particles and ozone (soot and smog) by 2009 at the latest, or sooner (“as expeditiously as practicable”).
The administration’s bill would postpone the attainment deadline for the country’s unhealthy air areas by six years or more. As long as states could show that their polluted areas would attain the smog and soot standards by 2015, those areas would be labeled “transitional” rather than “nonattainment” and be granted automatic extensions of the deadlines to meet health standards. Since the requirement to attain the standards “as expeditiously as practicable” applies only to nonattainment areas, states would be under no obligation to bring air quality into line with the health-based standards any earlier than 2015. In other words, the administration’s bill would force as many as 175 million Americans to breathe harmful amounts of air pollution for at least six years longer than current law allows.
By labeling hundreds of polluted counties “transitional” rather than “nonattainment,” the administration’s bill also would allow every major industrial source built or modified in those areas to make health problems worse by evading the lowest achievable emissions rate (“LAER”) and offset requirements of current law. Under current law, anyone wishing to build or modify a major source of air pollution in a “nonattainment” area must ensure that the source employs state-of-the-art methods to minimize its pollution (LAER) and must offset any added emissions so as not to degrade the already poor air quality in the area. This requirement applies not just to power plants, but to all other major air pollution sources (oil refineries, chemical plants, manufacturing facilities, etc.) as well.
Under the administration plan, these health safeguards would no longer apply in areas relabeled as “transitional.” In other words, the administration’s bill would make it easier for the owners of oil refineries, chemical facilities, paper mills, and power plants to churn out additional pollution in hundreds of counties where the air is already unhealthy to breathe. It is important to emphasize that while the administration plan caps only power plant emissions the bill would create this loophole for all major industrial sources. Amazingly, the administration has not offered a word of justification for this remarkable assault on the Act’s public health safeguards.
Section 3(r)(6) of the administration bill amends section 126 of the current Clean Air Act to prohibit EPA from ordering reductions in power plant pollution transported from upwind states unless EPA makes a series of new, onerous findings: EPA must find that every cheaper reduction (in terms of cost per ton of emissions and in terms of cost per microgram of air quality improvement) has already been made from industrial boilers, on-road mobile sources, off-road mobile sources, and any other category identified by EPA.
These provisions would effectively override key court decisions that have upheld EPA’s reasonable interpretation of the Act set forth in two rulemakings addressing interstate transport of NOx pollution. These cases upheld EPA’s determination to require reduction of upwind emissions that “contribute significantly” to downwind pollution. While it was necessary for EPA to show that these reductions are cost-effective, the agency was not required to show that all other more cost-effective ways to reduce emissions and concentrations had been exhaustively required first. In doing so, the courts upheld EPA’s rejection of far more onerous and unmanageable approaches pushed by industry and opposing upwind states.
In particular, EPA found in 1998 that the second approach – demonstrating cost-effectiveness per microgram of air quality improvement – would be utterly impractical. Furthermore, it would be inconsistent with an emissions trading approach, which requires emissions to be treated as equivalent on a ton-for-ton basis, and cannot work if each ton of emissions must be weighted differently depending on its distance from a particular spot where air quality improvement per microgram is assessed.
In addition to imposing new, essentially insurmountable tests, the bill would block EPA from granting downwind states any relief from upwind power plant pollution until after 2012. This stands in stark contrast to the extremely expedited relief structure of the current Clean Air Act. As EPA has noted:
Section 126 provides a tool for downwind states, the entities with most at stake, to force EPA to confront the issue directly. It also sets up an abbreviated, and hence potentially faster, process to achieve emission reductions. . . . . In contrast [to the SIP process] Congress required very expeditious EPA action on a [section 126] petition and from 3 months up to three years for sources to comply.
In an example of “Catch-22,” section 3(r)(6) of the administration bill states that before giving a state relief from transported pollution from power plants, EPA must first determine that the state has achieved all more cost-effective emissions reductions (on both a per-ton and per-microgram basis) from both on-road and off-road mobile sources. But this places states in an impossible situation, since the Clean Air Act elsewhere preempts states from controlling emissions from on-road vehicles and engines, CAA § 209(a), and nonroad vehicles and engines, CAA § 209(e).
As EPA has previously recognized, “Congress provided section 126 to downwind States as a critical remedy to address pollution problems affecting their citizens that are otherwise beyond their control, and EPA has no authority to refuse to act under this section.” But the administration’s legislative response to the problem of transported air pollution is to saddle downwind, polluted states with insurmountable barriers to relief.
Eliminating Safeguards Against Pollution Hotspots. Under the administration’s bill, a power plant can pollute at any level so long as it buys sufficient pollution allowances credits from other plants. The fact that power plant pollution may decline nationwide, however, provides no protection to the communities affected by a plant whose emissions stay the same, or even increase, because of its owner’s reliance on emissions trading. The “new source review” (NSR) provisions in the Clean Air Act provide important protection against the emergence of “pollution havens” or “hotspots” in response to an emissions trading system. NSR requires any person planning to build a new major pollution source, or to change an existing one in a way that will cause an emissions increase, to demonstrate that the source will use the most effective pollution control methods available and that its emissions increase will not degrade air quality either locally or in downwind communities or national parks.
The administration’s bill would eliminate federal new source review provisions for power plants, however. If the bill were enacted, a company would be free to cause even massive pollution increases by building a new plant or expanding an old one without adopting up-to-date pollution controls or determining whether air quality will get worse locally or downwind.
To replace the federal NSR program, the administration bill calls on states to submit state implementation plan (SIP) revisions to create a new and less protective state new source review program. The bill sets no deadlines for these SIP revisions. The bill does not clearly provide that the public must have an opportunity to comment on a permit application. The bill authorizes new loopholes for such state programs that would allow existing power plants to increase emissions by tens of thousands of tons each with no public process. And the bill exempts existing and new power plants everywhere in the country (except within a narrow 30-mile circle around national parks) from the current law’s safeguards for clean air areas.
Replacing Up-To-Date Technology with Obsolete Standards. In place of repealed requirements for case-by-case determination of up-to-date pollution control performance, the administration’s bill would substitute a requirement that EPA establish certain emissions standards that would apply to new power plants. The bill sets these standards at much more polluting levels, however, than the emissions levels of plants being built today. In other words, these standards are already obsolete and behind the curve of current requirements. For example:
• For boilers and integrated gasification combined cycle (“IGCC”) plants, the bill sets a SO2 emissions limit of 2.0 lb/MWh. Three recently issued permits for coal-fired boilers set SO2 emissions limits of 1.0, 1.2, and 1.0 lb/MWh, respectively.
• For boilers and IGCC plants, the bill sets a PM emissions limit of 0.2 lb/MWh. Three recently issued permits for coal-fired boilers set PM emissions limits of 0.12, 0.15, and 0.15 lb/MWh, respectively.
The bill does not place any obligation on EPA to update these already-obsolete emissions standards until eight years after the agency incorporates them into its regulations. Even then, the bill gives the agency discretion to avoid reviewing and updating the standards.
This is a sharp contrast with current law, under which the case-by-case review of LAER and (in areas other than nonattainment areas) “best available control technology” (BACT) assures that emission performance for new and modified plants keeps pace with improvements in pollution control capabilities. Because of BACT and LAER, the state-of-the-art in industrial pollution control has repeatedly graduated to successively higher levels of environmental performance as sources were built or modified over the last two decades.
For example, a review of EPA’s database for BACT and LAER determinations reveals that over just the past ten years, the state-of-the-art in NOx emissions controls for utility boilers and furnaces has advanced from no controls (“good combustion practices”) to low NOx burners to selective catalytic reduction (“SCR”) to selective non-catalytic reduction (“SNCR”) and circulating fluidized bed (“CFB”). Recent determinations by permitting authorities show that further improvements are in the wings.
As EPA and the courts have recognized, Congress intended the Clean Air Act to perform this “technology-forcing” function. The administration’s bill erases that function, leaving in its place static emissions standards that do not even represent the state-of-the-art in pollution control today.
EPA Assistant Administrator Holmstead has acknowledged in testimony delivered before this Committee that the new source review requirements have not adversely impacted construction or investment associated with new power plants. He testified that:
With regard to the energy sector, EPA found that the NSR program has not significantly impeded investment in new power plants or refineries. For the utility industry, this is evidenced by significant recent and future planned investment in new power plants. 
This Committee should recall that in 1990, the first President Bush did not seek to repeal these safeguards when he sought a cap and trade program for SO2 from power plants and congress did not enact such a repeal. Those programs have worked in tandem for the past thirteen years. The Act’s safeguards for local air quality have not interfered with the acid rain cap and trade program and have not prevented the very large economic savings provided by the cap and trade mechanism. Experience proves that both programs can work together and this congress should not ignore that fact.
Eliminating Protections for National Parks. The administration bill would exempt owners of new and modified power plants from the obligation to meet up to date pollution performance standards (BACT) and examine the impacts of any added pollution on national parks or wildernesses --called “Class I areas”-- (except those within 30 miles of the plant). The bill also eliminates the role of the federal land manager (typically the National Parks Service superintendent for a national park) in assuring that the air quality of these treasured lands is protected.
Under current law, if a new or expanded pollution source could affect a Class I area, the federal land manager has an opportunity to review the draft permit and an accompanying air quality analysis to assure that factors relevant to protecting national parks and wilderness areas are taken into consideration, and that harmful effects are mitigated. The federal land manager’s review would be eliminated under the administration bill for all plants outside the 30 mile cordon around each park or wilderness.
The administration bill would also repeal the current Clean Air Act program to lift the haze shrouding the nation’s parks by obligating the states to require the best available retrofit technology (“BART”) on all major sources of air pollution built between 1962 and 1977 that contribute to the haze. The administration’s bill exempts all power plants – the primary contributor to park haze – from the BART requirement. In so doing, the bill lets off the hook those intransigent companies that have not yet installed the best available retrofit technology on their plants.
If the administration elected to enforce the requirement, instead of lifting it, the installation of BART on just the largest power plants would reduce annual SO2 emissions by 4.5 million tons, and annual NOx emissions by 1.9 million tons. Those reductions alone would be equivalent to what the administration’s bill would purportedly achieve in its entire 8-10 year first phase.
Eliminating Protections for Other Areas With Clean Air. The administration bill weakens air quality safeguards across the rest of the country. Except in the 30 mile cordon around our national parks and wildernesses, the bill eliminates any case-by-case review of proposed new power plants anywhere in the country. Under current law, plants locating in the vast majority of the country – areas classified as “Class II PSD areas,” with air quality better than the national health standards – must undergo new source review and demonstrate that they will be equipped with “best available control technology” (BACT) and will not cause excessive degradation of air quality (“pollution increment consumption”) in the surrounding area. And, as already mentioned, plants locating in nonattainment areas must meet the “lowest achievable emission rate” (LAER) and must offset any additional pollution.
The administration bill repeals these requirements, creating a vacuum where there is no federally enforceable obligation for new or modified power plants to provide anyone with notice of their intent to build or expand, no requirement to review air quality impacts, and no requirement to case-by-case review of emission control performance. The only requirement would be to meet the obsolete national standard described above.
Weakening Safeguards Against Hazardous Air Pollution. I have already described how the administration bill would repeal the current Clean Air Act’s requirement for applying “maximum achievable control technology” (MACT) to power plants to curb their mercury emissions. The bill requires no mercury controls until 2010 (a two-year delay over the current law) and substitutes much weaker mercury caps in place of the plant-by-plant MACT requirement. For 2010 through 2017, the bill’s 26 ton cap represents merely the mercury reductions incidental to the bill’s phase-one caps for SO2 and NOx. Mercury cuts beyond these incidental reductions are not achieved until 2018. In other words, the administration’s 3-pollutant bill is effectively a 2-pollutant bill until 2018.
Also repealed with mercury MACT is the current law’s requirement that EPA establish MACT standards for all hazardous air pollutants emitted by power plants, not just mercury. For hazardous pollutants other than mercury, the bill leaves only the authority to set “residual risk” standards through a complex risk-based process, but the earliest that those regulations are permitted to take effect is 2018 – a full 11 years after the MACT compliance deadline of the current Clean Air Act. Moreover, the bill repeals the Clean Air Act’s “residual risk” protections entirely for mercury without regard to any health risks that remain under the bill’s weaker mercury caps.
The administration bill allows unrestricted emissions trading of mercury, something never before allowed under the Clean Air Act for a hazardous air pollutant. The current Clean Air Act requires mercury reductions at each power plant, based on the emissions reductions achievable through advanced technologies applied to individual emissions units. By allowing mercury trading, the bill would allow some power plants not to reduce their emissions at all. Instead, they could buy mercury emission allowances from other power plants and do nothing to stop contamination of local lakes and streams. Some plants could even increase their mercury emissions.
Indeed, EPA’s own analyses of the administration bill acknowledge mercury pollution increases above today’s levels from “specific sources in some states,” due to the trading features of the bill and the bill’s repeal of the 2007 MACT standard. This dirtier outcome would not be allowed if the plant-specific MACT standard were to remain in effect. EPA’s data also show that parts of New England, the Great Lakes, Gulf Coast region and other areas will receive only very small reductions in mercury deposition under the bill.
Because unrestricted trading of mercury emissions could lead to toxic hotspots where mercury contamination increases, the Clean Air Act – as well as other legislative proposals (notably the Clean Power and Clean Smokestacks Acts) -- bar trading in mercury emissions. Hotspot risks under the administration bill are made worse by the fact that the bill does not require continuous emissions monitoring (CEMS) for mercury. EPA itself has identified continuous monitoring and reporting as design features essential to the environmental integrity of the acid rain trading program. Mercury emissions trading is allowed even without continuous monitoring so long as the Administrator determines that “CEMS for mercury with appropriate vendor guarantees are not commercially available.” The responsible approach would be to make any mercury trading (if some carefully limited program were shown to prevent hotspots) contingent on the development of reliable continuous monitoring systems for the pollutant.
As you know, even though electric power plants are responsible for 40 percent of U.S. carbon dioxide (CO2) emissions, the administration bill does not contain any provision to reduce or even limit the growth in these emissions. This feature of the administration bill is not just an omission; it is a serious affirmative mistake that will make it more difficult for the U.S. to take responsible action in the near future to begin addressing the real dangers of global warming.
The administration bill would set in motion major capital expenditures at existing power plants over most of the next two decades. To implement such a program without addressing CO2 emissions is to invite shortsighted investment decisions and promote even greater resistance to proposals to limit CO2 when we do decide to act. A choice to ignore CO2 emissions in a power plant bill is not just a choice to leave this decision open for tomorrow. It is a decision that will raise the cost and difficulty of beginning to address the nation’s largest source of global warming pollution.
Such a choice is not responsible. Delay will turn what is still a manageable threat into a runaway, unmanageable problem. In the national security context, the current administration has no difficulty understanding that waiting until a danger has fully developed runs the risk of foreclosing our ability to avert that danger. This logic applies strongly to the danger posed by global warming. If we wait until this danger has fully developed, it will be too late to prevent.
Global warming is a problem that has enormous built-in inertia. The most important global warming gas, CO2, stays in the atmosphere for hundreds of years. The largest sources of CO2, fossil-fueled power plants, have lifetimes of 50 years and more. Managing the threat of global warming is like navigating a supertanker—to avoid running aground we have to start altering course long before we arrive at the reef. While in the global warming context we may not have identified exactly how close we are to the reef or how severely our ship will be damaged from striking it, it is a fact that if we steam ahead with our current energy systems until we have all the evidence required to satisfy the skeptics, we run very large risks of locking ourselves into very large-scale unavoidable damage.
We are already unalterably committed to a future in which the concentration of global warming gases will be substantially higher than pre-industrial levels. To avoid reaching concentrations that are several times pre-industrial levels, we will need to change the technology we use to generate power and for transportation. In the decision whether to include CO2 in a power plant emission control bill, this congress will either stimulate investors to get serious about developing and using new climate-friendly power technology or it will send a signal to procrastinate.
Advocates of delay argue we should not act until we know exactly how sensitive the climate is to added CO2 and exactly how harmful a given temperature rise will be. Unfortunately, we cannot put the world on “pause” while we do more research. We cannot afford to wait for resolution of these uncertainties before we begin to change energy investments. The CO2 we emit may cause a temperature rise at the high end of published estimates, the low end, or in between; the damage done by a specific temperature rise also may be larger or smaller. But once we know for sure, it will be too late to change course. The fact is that continuing on our current path will commit us to an outcome that we will not be able to undo.
A paper by Ken Caldeira and colleagues published two weeks ago in Science magazine demonstrates the danger of continued procrastination. Using mid-range estimates of climate sensitivity, the authors conclude that we would need to be building the equivalent of about 20 CO2-emission-free power plants a week worldwide, starting now, to keep global temperatures from increasing more than 2° Centigrade. Consider that global temperature in the last ice age was 5° C cooler than today and you can appreciate that 2° would be a very big change. On our current path, however, the world is on track to add the equivalent of less than 2 CO2-emission-free plants per week between now and 2030. The authors go on to demonstrate that even assuming a best-case outcome for these uncertainties, we still need a massive increase in CO2-emission-free energy resources compared to current forecasts.
Today people talk about the need to design climate policy to avoid “premature retirement” of existing capacity. But tomorrow’s “existing” capacity is being designed and financed today and it is not being designed to be climate friendly. Policy procrastination just locks us in to more high-carbon capacity that will either have to be retired “prematurely” or will emit amounts of CO2 that could make it impossible to stabilize concentrations of global warming emissions at safe levels.
The U.S. Energy Information Administration forecasts that the United States will build the equivalent of over 1350 medium-sized fossil energy power plants between now and 2025 (405,000 MW). The path we are on today will result in skyrocketing emissions of CO2 in the U.S. and globally. Figure 9 shows current forecasts for the U.S. and the world over the next 25-30 years: U.S. emissions are projected to increase by 40% and world emissions by nearly 70% over year 2000 levels. These emissions will stay in the air for hundreds of years making the task of protecting the climate that much harder and more expensive.
The problem of delay is particularly intense with respect to the electric power sector. As we know, power plants have extremely long lives. There are plants in the U.S. more than 60 years old that are still operating today. New plants built in the next decade or two will be operating in the third quarter of this century, and their cumulative emissions will determine how much the climate warms. While we procrastinate, energy demand keeps growing and more investments are made in power plants that are no less carbon-emitting than yesterday’s plants.
Which brings us to the choice before you. Including provisions to limit CO2 in a power plant bill can speed the process of bringing advanced technologies to market; leaving CO2 out will keep that activity on the back burner. Analyses discussed in NRDC’s testimony to the full Committee in June 2002 show that it is possible to craft legislation that limits power plant CO2 with modest impacts on the economy.
The administration and many in Congress have resisted including a binding limit on CO2 in power plant legislation out of an apparent belief that any binding cap will have unacceptable impacts on electricity rates and fuel diversity. That is not correct.
For example, even the administration’s own analyses conclude that some versions of binding CO2 caps would have very modest impacts on electricity rates and fuel use, even when using a number of conservative (and we believe, flawed) assumptions. In September and October 2001, both EPA and EIA analyzed a binding carbon cap for the electric sector using a set of requirements specified by Chairman Voinovich, former Senator Smith, and Senator Brownback. Among the scenarios examined by EIA and EPA were requirements to cut SO2 , NOx, and mercury emissions by 75 percent from 1999 levels in two stages (2007 and 2012) and to cap power sector CO2 emissions at forecasted 2008 levels.
EIA’s report calculated this set of requirements would result in an average electricity rate of 7.1 cents per kwh, compared to a 1999 average electricity rate of 6.7 cents per kwh. EIA projected coal consumption in 2020 would be the same as in 1999.
While power sector CO2 emissions need to decline below 2008 levels, the key point is the need to set a schedule now for limiting and then decreasing emissions of CO2. By adopting a schedule now, you can provide the maximum lead-time for the industry and achieve long-term reductions at the most gradual rate of change. By adopting a schedule for limiting carbon emissions you put market forces to work to deliver the clean energy resources we will need to meet economic growth without disrupting the climate that strongly influences the quality of life in our country and others around the globe.
We can do three things to limit carbon emissions from energy use. First, produce and use energy more efficiently. Second, dramatically increase our reliance on renewable energy resources. Third, pursue methods to capture and permanently store CO2 from the fossil energy sources we continue to use. All three of these methods will be stimulated by adopting a program to limit CO2 emissions from the power sector. All three will languish if congress ignores CO2 in a power plant bill.
Members of this Committee and others in congress are concerned about the impact of climate policy on coal. The U.S. and other large countries, including China, India, Russia (to mention just a few) have abundant coal resources. While coal and other fossil fuels have continuing environmental impacts, including global warming emissions, the reality is that large amounts of coal will continue to be used.
Fortunately, technologies in commercial operation today demonstrate it is feasible to capture CO2 from coal-based power plants in a form that can be kept out of the atmosphere provided that suitable geologic repositories are developed. As I mentioned in my June 2002 testimony to the full Committee, in the U.S. today we inject over 30 million tons of CO2 annually into oil fields to recover additional oil. Yet, none of that CO2 is supplied by power plants. Rather it is pulled out of natural CO2 reservoirs and piped hundreds of miles to be stuck back in the ground.
Because industrial CO2 can still be emitted to the air in unlimited amounts for free, there is not an adequate economic incentive to use and optimize existing technology to capture these emissions. Nor is there an adequate incentive to invest to bring down the costs of today’s gasification and CO2 capture systems.
Ironically, the current policy procrastination has made the U.S. coal industry’s posture a very uncertain one. No one believes that action on global warming can be delayed indefinitely and this causes investors to be leery of large new investments in conventional coal-fired power plants. On the other hand, without a policy resolution, setting forth a program to limit CO2 emissions over time, the uncertainty is too great for most investors to develop and plan to deploy advanced coal technologies like gasification and capture systems.
In sum, failure to include CO2 limits in a power plant bill has real costs. It would keep the U.S. and the world on a path of accelerating CO2 emissions—a path that is unacceptably risky given what we already know about the potential of global warming to change our lives for the worse. It would steer investments at the margin to patching up old, existing capacity that should be replaced with modern, efficient systems. And it would continue the policy uncertainty that operates as an obstacle today to business planners considering what energy investments they should pursue.
The good news is that by acting now to adopt a schedule for limiting CO2 emissions we can change behavior and make it easier to address global warming. For example, the International Energy Agency forecasts that nearly the world will build new coal plants equal to nearly five times the current U.S. coal plant capacity between now and 2030. While seemingly a daunting prospect, this projection really means that two out of every three coal power plants forecasted to be operating in 2030 are not yet designed or built. With U.S. leadership, we can design new energy projects to rely on climate-friendly technology. Doing so will expand our options to reconcile aspirations for improved economic well-being around the world while preserving the climate we all depend on to provide us with a hospitable place to live.
In conclusion, let me suggest it is time for all sides to stand down from the posturing of past years on this issue and adopt a more pragmatic approach. There are many sensible policies that can be adopted to start limiting CO2 emissions and there are many compelling reasons to do so. Working together, members of both parties and the administration would be able to identify a path forward that all could embrace and all could point to as a real accomplishment. NRDC will work with you to help make that happen.
 U.S. EPA, “Comprehensive Approach to Clean Power: Straw Proposal and Supporting Analysis for Interagency Discussion,” August 3, 2001. (“EPA August 2001 Analysis”) Available at http://www.catf.us/publications/other/EPA_Straw_Proposal.pdf.
 The results shown in the figures below are based on EPA analyses with the Integrated Planning Model (IPM), the standard modeling tool used by all stakeholders in the power plant debate. They show the pattern of emissions expected under the two plans, including the impact of “banking,” which results in some reductions below the caps in early years in order to emissions at levels above the caps in later years.
 EPA calculated the incidence of premature death and illness for both its August 2001 proposal and the administration CSI plan using the same peer-reviewed methods and summarized the results in its technical analysis documents. The health comparisons in this testimony are taken from the incidence figures presented in those documents. See, EPA August 2001 Analysis, supra, note 1, Appendix A at 3, and EPA CSI Technical Support Package, September 2002 at 29, at www.epa.gov/clearskies.
 Clean Air Task Force, "Health Damages Estimates for Clear Skies Initiative and Straw Proposal," April 2003.
 U.S. EPA, 1997f. Mercury Study Report to Congress, Volume V: Health Effects of Mercury and Mercury Compounds. EPA-452/R-97-007; Toxicological Effects of Methylmercury, National Academy Press, Washington, DC, 2000.
 Centers for Disease Control, January 2003. Second National Report on Human Exposure to Environmental Chemicals; Susan E. Schober, et. Al, “Blood Mercury Levels in U.S. Children and Women of Childbearing Age, 1999-2000,” Journal of the American Medical Association, 289: 1667-74 (2003)
 Derived by the Clean Air Task Force from 2000 census data and fertility data from the National Center for Health Statistics.
 Toxicological Effects of Methylmercury, National Academy Press, Washington, DC, 2000.
 High levels of mercury in seafood linked to infertility. BJOG: an International Journal of Obstetrics and Gynecology. 109:1121-5, 2002; Toxicological Effects of Methylmercury, National Academy Press, Washington, DC, 2000.
 While most mercury comes from coal-fired power plants, the IPM runs reveal that several tons of mercury come from oil-fired plants that would not be covered under the administration plan (or under the EPA 2001 proposal). These tons of mercury are also of concern and should be covered by mercury controls under either new legislation or existing law.
 Mr. Holmstead so characterized the administration’s baseline assumptions in a presentation to the National Association of Regulatory Utility Commissioners in Washington on February 24, 2003.
 The deadline for ozone designations is set forth in a consent decree entered in American Lung Ass’n, et al. v. Whitman, D.D.C. Civil Action No. 02-2239 (March 13, 2003). EPA has committed to an end-of-2004 deadline for fine particle designations as “one of the Agency’s highest priorities, due to the serious health implications of PM2.5 [fine particle] exposure. . . .” Memorandum from Jeffrey R. Holmstead, Assistant Administrator, to Regional Administrators (Nov. 14, 2002), at 3.
 Clean Air Act § 7502(a)(2)(A). See Sierra Club v. EPA, 294 F.3d 155, 162-63 (D.C. Cir. 2002) (state must adopt all reasonably available measures capable of advancing the date on which the polluted area will attain the NAAQS). “In order for the EPA to determine whether an area has provided for implementation as expeditiously as practicable, the State must explain why the selected implementation schedule is the earliest schedule based on the specific circumstances of that area. Such claims cannot be general claims that more time is needed but rather should be specifically grounded in evidence of economic or technologic infeasibility.” Memorandum from John S. Seitz, director of EPA’s Office of Air Quality Planning and Standards, to EPA regional air division directors (November 2, 1999), at 1
 Clean Air Act § 7502(a)(2)(A). EPA also may grant a maximum of two one-year deadline extensions if an area has met all its requirements and experiences “no more than a minimal number” of violations of the health standards in the otherwise applicable deadline year.
 Clean Air Act § 7410(a)(2)(D)(I), § 7426.
 U.S. EPA, Engineering and Economic Factors Affecting the Installation of Control Technologies for Multipollutant Strategies, October 2002.
 Clean Air Act § 7412(n)1)(A) provides that EPA shall issue MACT standards for power plants if the Administrator determines, after a study, that such standards are “necessary and appropriate.” The Administrator made this determination for power plant emissions of mercury in 2000. 65 Fed. Reg. 79,825 (December 20, 2000).
 Puruant to a consent decree in NRDC v. EPA, et al., Case No. 92-1415 (D.C. Circuit), EPA is required to propose a mercury MACT standard by the end of 2003, and to promulgate the standard by the end of 2004.
 Clean Air Act § 7412(i)((3)(A). Under certain circumstances, EPA may allow a specific facility one extra year for compliance. Id. § 7412(i)((3)(B).
 H.R. 999 I.H. (“Sec. 3. Other Amendments”), at 227 lns. 5-22.
 See 42 U.S.C. § 7502(a)(2).
 Current law permits limited postponement of the 2009 deadline only where the EPA makes an appropriateness determination “considering the severity of nonattainment and the availability and feasibility of pollution control measures.” Id. § 7502(a)(2)(B). See also id. § 7502(a)(2)(C), (D). The administration’s bill does not condition the availability of the 2015 postponement on any such determination. See H.R. 999 I.H. at 227 lns. 5-22.
 42 U.S.C. §§ 7502(c)(5), 7503(a)(2), (c).
 Id. § 7502(c)(5).
 See § 3(r)(6)(D) (adding CAA § 126(d)(2)(B)(i) & (ii)).
 63 Fed. Reg. 57356 (Oct. 27, 1998) (NOx SIP Call rulemaking); 64 Fed. Reg. 28250 (May 25, 1999) and 65 Fed. Reg. 2674 (Jan. 18, 2000) (Section 126 rulemakings).
 See Michigan v. EPA, 213 F.3d 663 (D.C. Cir. 2000) (upholding NOx SIP Call approach); Appalachian Power Co. v. EPA, 249 F.3d 1032 (D.C. Cir. 2001) (upholding same approach in section 126 rulemaking).
 See Michigan, 213 F.3d at 675-680; Appalachian Power at 249 F.3d at 1044-1051.
 U.S. EPA, “Rulemaking for Section 126 Petitions -- Responses to Comments Which are Outside the Scope of the June 24, 1999 Notice of Proposed Rulemaking,” Docket A-97-43, XII-A-01, 65 Fed. Reg. 52931 (Aug. 31, 2000).
 65 Fed. Reg. 2674, 2681 (Jan. 18, 2000).
 States may control these mobile sources of emissions only by adopting California standards. CAA §§ 177 & 209(e)(2).
 65 Fed. Reg. at 2681/1 (emphasis added).
 H.R. 999, (“Sec. 403. Allowance System”) at 14 ln. 15 – 15 ln. 18.
 42 U.S.C. §§ 7475, 7501-7503. Current law requires a company to demonstrate that the planned construction or other change will not cause or contribute to pollution in excess of certain maximum allowable increases and maximum allowable concentrations that are separated from the NAAQS by a safety margin. 42 U.S.C. § 7475(a)(3)(A). The administration’s bill simply requires a demonstration that the planned activity will not cause or contribute to a violations of – or inability to achieve – the NAAQS itself. H.R. 999 I.H. (Sec. 483(c)(1), (2)) at 224 ln. 15 – 225 ln. 8.
 Current law requires a company to demonstrate that the planned construction or other change will not degrade visibility or other air quality related values at any national park. 42 U.S.C. § 7475(a)(5), (d). If the administration’s bill were enacted, such a demonstration would not be required unless the plant in question were located within fifty kilometers of a park. H.R. 999 I.H. (Sec. 483(b)) at 224 lns. 8-14. This despite the fact that emissions from major pollution sources have been shown to have a negative impact on parks as far as 700 kilometers away. See Gebhart, K., “Preliminary Particulate Sulfur Source Attributions for BRAVO by Trajectory Mass Balance Regressions” (presentation for BRAVO conference call on November 21, 2002) (analysis on file with the Clean Air Task Force).
 See H.R. 999 (“Sec. 483. Exemption from Major Source Preconstruction Review Requirements and Best Available Retrofit Control Technology Requirements”) at 223 lns. 10-14 (“An affected unit shall not be considered a major emitting facility or major stationary source, or a part of a major emitting facility or major stationary source for purposes of compliance with the requirements of parts C and part D of title I.”).
 Id. (Sec. 481(b)(1), (c)(1), (d)) at 205 lns. 1-9, 207 ln. 9 – 211 ln. 7.
 Id. (Sec. 481(c)(1)((A)) at 207 lns. 17-18.
 Wygen 2 plant in Wyoming; Roundup plant in Montana; IPP plant in Utah.
 H.R. 999 (Sec. 481(c)(1)(B)) at 207 lns. 19-20.
 Wygen 2 plant in Wyoming; Roundup plant in Montana; IPP plant in Utah.
 H.R. 999 (Sec. 481(c)(1)(C)) at 207 lns. 21-22.
 Wygen 2 plant in Wyoming; Roundup plant in Montana; IPP plant in Utah.
 H.R. 999 (Sec. 481(e)(1)) at 211 lns. 8-18.
 Id. (Sec. 481(e)(2)) at 212 lns. 3-8.
 See http://cfpub1.epa.gov/rblc/cfm/basicsearch.cfm.
 See, e.g. Letter from Richard L. Goodyear, permit programs manager, State of New Mexico Air Quality Bureau, to Larry Messinger, Mustang Energy Company (December 23, 2002), at 1-2 (“The analysis must include a discussion of the technical feasibility and availability of IGCC and CFB for the proposed site in McKinley County . . . .”).
 See “Background Statement on the Environmental Protection Agency’s Top-Down Policy” (June 13, 1989) (citing S. Rep. No. 95-252, 95th Cong., 1st Sess. 31 (1977)), reprinted in, 3 A Legislative History of the Clean Air Act Amendments of 1977 at 1405; 123 Cong. Rec. A9171 (remarks of Senator Edmund G. Muskie), reprinted in 3 Legislative History at 729. See also WEPCO v. EPA, 893 F.2d 901, 909 (7th Cir. 1990).
 Testimony delivered Assistant Administrator Jeffrey Holmstead to the United States Senate Committee on Environment and Public Works on July 16, 2002.
 42 U.S.C. § 7491(b)(2)(A).
 H.R. 999 (Sec. 483(a)) at 223 lns. 10-18.
 MSB Associates, analysis using EPA list of BART eligible sources exceeding 750 MW (analysis on file with the Clean Air Task Force).
 § 3(a)(5).
 See U.S. EPA, “Technical Support Package for Clear Skies,” Section B: Human Health and Environmental Benefits, at 44.
 Testimony of Jeffrey Holmstead, Assistant Administrator, Office of Air & Radiation, U.S. EPA, Before the Subcommittee on Public Health of the Committee on Health, Education, Labor and Pensions, United States Senate, at 4-5, September 3, 2002.
 § 405(a)(2)(B)(ii).
 K. Caldeira et al., “Climate Sensitivity Uncertainty and the Need for Energy Without CO2 Emission,” Science 299, 2052 (2003).
 International Energy Agency, World Energy Outlook 2002, October 2002.
 U.S. Energy Information Administration, Annual Energy Outlook 2003.
 See Testimony of David G. Hawkins on S.556, June 12, 2002.
 A number of flaws in the administration’s analyses of “four-pollutant” bills are described in NRDC’s testimony of June 12, 2002 at the full Committee hearing on S.556, the Clean Power Act. Testimony of David G. Hawkins at 12-16.
 Energy Information Administration, “Reducing Emissions of Sulfur Dioxide, Nitrogen Oxides, and Mercury from Electric Power Plants,” September 2001. (“EIA S-V-B report”) and U.S. EPA, “Analysis of Multi-Emissions Proposals for the U.S. Electricity Sector,” October 2001.
 Letter of June 8, 2001 from Senators Smith, Voinovich, and Brownback to John Weiner, EIA, reproduced in EIA S-V-B report at Appendix A. Compliance with the CO2 cap could be achieved with on-system reductions or credits for “sinks” enhancements or reductions from other source categories. EIA’s report calculated costs assuming that only CO2 emission reductions from U.S. energy facilities would be used for compliance.
 While this result represents a decrease in coal consumption from no-control forecasts, EIA’s report assumed no penetration of coal- gasification technology in the electric sector, even by 2020. This is inconsistent with the Department of Energy’s programmatic goals for this technology. EPA’s report on the S-V-B scenario forecasts smaller price and fuel impacts than EIA’s, due to EPA’s broader assumed trading options than EIA assumed.
 IEA, World Energy Outlook 2002, October 2002.
 GW=gigawatt, which is 1000 MW of capacity. Current U.S. coal capacity is just over 300 GW.