In July, the Bush Administration published an astonishing set of documents relating to climate change regulation: a staff-driven EPA AdvanceNotice of Proposed Rulemaking (“ANPR”), accompanied by statements disowning it from OMB, several cabinet agencies, and the Administrator of EPA himself.[1] As a further indication of the Administration’s uncertainty about global warming issues, EPA left the notice open for comment until three weeks after the election.

This article examines how the concepts the ANPR addresses may be treated by the Obama Administration. We see the ANPR principally as a guide to the statutory provisions most likely to be invoked by an Obama EPA that wants both to address underlying concerns and to force new Congressional action.


In October, 1999, nineteen organizations filed a petition asking U.S. Environmental Protection Agency to regulate greenhouse gas emissions from motor vehicles. Section 202 of the Clean Air Act provides for the promulgation of regulations when, in the Administrator’s judgment, emissions “cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare.

Almost four years later, in the summer of 2003, EPA announced — contrary to the publicly stated positions of the two previous EPA General Counsels — that the Clean Air Act did not authorize the Agency to regulate greenhouse gases, because they did not constitute “air pollution.” Judicial challenges followed. In 2007, the Supreme Court addressed the issue and held that the Agency was wrong.

In Massachusetts v. Environmental Protection Agency, the majority of the Justices found that “because greenhouse gases fit well within the Clean Air Act’s capacious definition of ‘air pollutant’ . . . EPA has the statutory authority to regulate the emission of such gases from new motor vehicles.” [2] The court directed that, on remand, EPA make an “endangerment finding.”

In May, 2007, President Bush issued an Executive Order directing EPA and other applicable federal agencies to respond.[3] But the agencies involved could not agree on a uniform position. Instead, EPA published the ANPR.

The ANPR analyzed the potential implications of an endangerment finding as to vehicle emissions and solicited public comment on the implications of such a finding. Extraordinarily, however, the ANPR package was referred to as an “EPA staff draft,” and several other federal agencies submitted detailed comments criticizing various aspects of it.

Among other things, this lack of consensus reflects the potential economic and political impacts of EPA action. In the preface to the ANPR, EPA Administrator Stephen Johnson said “one point is clear: the potential regulation of greenhouse gases under any portion of the Clean Air Act could result in an unprecedented expansion of EPA authority that would have a profound effect on virtually every sector of the economy and touch every household in the land.” [4] If anything, Administrator Johnson understated the implications of invoking Clean Air Act authorities for this purpose.

The Obama Administration is unlikely to continue to duck this issue, although the issue’s potentially enormous economic impacts in an uncertain economic environment will surely influence its approach. President-Elect Obama has expressly endorsed enactment of new “cap and trade” legislation and sees invoking Clean Air Act regulation as a path toward that result. Jason Grumet, a key Obama energy policy aide, has said that “the EPA is obligated to move forward in the absence of Congressional action.” Further, “[i]f there’s no action by Congress in those 18 months [that it would take to promulgate regulations], I think any responsible president would want to have the regulatory approach.” [5] Translation: the forthcoming Obama Administration recognizes the enormous leverage to be gained in legislative negotiations by directing EPA to proceed with the greenhouse gas rulemaking process.

The Problem Identified in the ANPR 

A fundamental message of the ANPR is true: declaring that greenhouse gases are pollutants that endanger human health and therefore subject to regulation under the Clean Air Act’s mobile source (i.e., vehicular) provisions might require that EPA take similar regulatory action with regards to stationary sources. The core concern is that the same words used in the Act’s mobile source provisions to trigger action, or words very close to them, appear in the Act’s stationary source provisions. Thus, a finding of “endangerment” in the mobile source context could be found to require action against emission sources ranging from power plants and factories to home heating systems.

The ANPR invites comment on all aspects and ramifications of such a mobile source endangerment finding. We address below some of them, and suggest possible ways around at least a few. First, however, it is worth flagging the key difference between trying to address climate change through existing Clean Air Act authorities, rather than a statute tailored to that problem: the Clean Air Act doesn’t provide any mechanism to buffer the enormous economic impacts of a climate change regulatory program. And it is the need to do so, we believe — as Mr. Grumet suggested — that will drive the new Administration’s policy.

The Real Clean Air Act Problem: No Subsidies 

The various legislative schemes that have surfaced in the 110th Congress to establish a cap-and-trade system carried a huge price tag: the National Association of Manufacturers, for example, estimated the bill considered by the Senate would cost $151 billion to $210 billion in 2020 and rise to $631 billion to $669 billion annually by 2030. (The estimates vary based on fuel cost and usage assumptions.) All the monies to be paid for emission allowances would create a huge pool of funds. The federal government would then either spend these funds or distribute emission allowances for free in an effort to moderate the effects of the new regulatory scheme. Sure, gasoline prices might go up 20 percent to 69 percent by 2020, and 77 percent to 145 percent by 2030 (again, we cite NAM’s figures). Sure, the price of electricity would go up for homeowners and for industry, perhaps by 28 percent to 33 percent by 2020, and 101 percent to 129 percent by 2030. Sure, a unilateral effort by the United States to regulate greenhouse gas emissions could put domestic industries that face foreign competition in noncompetitive cost positions. But these reserved allowances and/or the accrued federal revenues could be used to offset and overcome these impacts — i.e., to provide subsidies.

Of course, some of those revenues might be diverted elsewhere — a likelihood that only increases with more recent demands on the federal treasury. And potentially affected segments of the population and industry would have to work long and hard to ensure that their interests were protected in the legislative arena, with the almost certain knowledge that climate legislation will yield winners and losers. But at least the legislative process offered some hope for rough justice and some mitigation of the inflationary effects of carbon regulation on energy prices.

The same cannot be said about greenhouse gas regulation undertaken under the existing Clean Air Act. It would impose the costs but does not provide any hope of offsetting them. Indeed, to a large extent, the statute itself prohibits the EPA from considering the cost impacts of its regulatory decisions.

The ANPR does a good job of laying out the inherent limitations of the Clean Air Act provisions, and the critiques offered by the other federal agencies reemphasize them. Listing greenhouse gases as criteria pollutants akin to SO2, NOx, or lead could indeed have disastrous implications: federally mandated permit programs for apartment buildings and private homes, making EPA the super-authority for construction of new industrial capacity across the nation, and on and on. And, absent EPA authority to regulate via a cap-and-trade alternative that offers at least some flexibility to regulators, the parade of horrors grows longer.

The North Carolina v. EPA Monkey Wrench

At the time the ANPR was being prepared, one of EPA’s fundamental assumptions was that sufficient authority existed to authorize some sort of a greenhouse gas cap-and-trade system, even though the Act does not expressly authorize one. As proof, the Agency pointed to, among other precedential clean air programs, the Clean Air Interstate Rule (“CAIR”).[6]

CAIR was a carefully constructed cap-and-trade scheme adopted after years of negotiation among the federal government, industry, states, and environmental advocacy organizations. Issued in March of 2005, CAIR permanently capped emissions of sulfur dioxide (SO2) and nitrogen oxides (NOx) in the eastern United States and would have managed emissions of these pollutants through the use of a regional cap-and-trade program. The Agency estimated that this flexible and market-based approach would reduce SO2 emissions in these states by more than 70 percent and NOx emissions by more than 60 percent from 2003 levels, without “command and control” regulation.

But on July 11, 2008, the same day the EPA Administrator signed and issued the ANPR, the D.C. Circuit Court of Appeals ruled the entire CAIR program unlawful. In North Carolinav. EPA, a unanimous three-judge panel ruled that any trading scheme that allowed some facilities to maintain — indeed, even increase — emissions was inconsistent with the Act, even if it resulted in a regional emissions reduction of air pollutants.[7] Why? Because the CAIR “does not connect states’ emissions reductions to any measure of their own significant contributions” and does not ensure that each state achieves the appropriate clean air standard, but the statute requires just that. That is, the Act does not permit a scenario where one state’s air pollution levels could deteriorate, even when the overall air quality of a region markedly improves. Describing CAIR as “a single, regional program . . . [where] all its components must stand or fall together,” the court vacated the entire rule and said EPA must “redo its analysis from the ground up.”

The North Carolina decision was a body blow to the Agency’s (and, for that matter, almost all of the other involved parties’) understanding of the regulatory flexibility inherent in the Clean Air Act. And it eliminates any chance to create cap-and-trade programs without legislative amendment.

However, the decision does not preclude EPA from regulating greenhouse gas emissions under the Clean Air Act. Many of the Act’s command-and-control tools could be invoked to do so, but — as the ANPR cautions — all the possible approaches have deficiencies of one sort or another.

How Could the Clean Air Act Be Used?

The Act’s stationary source provisions were designed to deal with pollution that came from discrete sources and had local or regional impacts. Never intended to regulate global pollutants that distribute themselves equally throughout the atmosphere, the Act is at best cumbersome when applied to the regulation of greenhouse gases.

A. Criteria Pollutants and National Ambient Air Quality Standards

The broadest and most widely used regulatory regime in the Clean Air Act is the National Ambient Air Quality Standards (“NAAQS”) program. Section 108 of the Act requires that pollutants with the broadest national impacts be identified as criteria pollutants — and the implications of greenhouse gases are certainly national (indeed, international). Section 109 requires EPA then to set NAAQS for every listed criteria pollutant. These are to be set at a level requisite to protect the public health, “with an adequate margin of safety,” and to protect the public welfare (generally understood to mean the environment) without regard to economic considerations. Each state is then charged with developing a statewide plan to either meet those levels or ensure that the levels are maintained.

Inherently, however, a state could never assure that any level for greenhouse gases was met within its borders, because of the universal, global dispersion of these gases. The ANPR suggests a state could attempt to excuse its failure to meet the NAAQS through section 179B of the Clean Air Act, a provision that eliminates sanctions when a state would have complied with the NAAQS “but for” international emissions. Even so, the states and their stationary sources likely would still be required to implement all of the applicable control technology in a useless and wasteful attempt to comply with the NAAQS. While it no doubt will remain to the courts to determine conclusively whether an endangerment finding as to vehicle emissions requires the promulgation of NAAQS for greenhouse gases, the idea of regulating a global pollutant via an unworkable NAAQS program seems unlikely to be the choice of the Obama Administration.

Nonetheless, either voluntarily or in response to a suit from some element of the environmental community, the new Administration will have to move toward promulgation of a standard for greenhouse gases emitted from motor vehicles. And the core concern reviewed in ANPR is that, should EPA make an endangerment finding regarding motor vehicle emissions, that finding could trigger mandatory invocation of the statute’s provisions directed to stationary sources.

While there is some basis for this view, there also is considerable reason to believe it wrong. The Supreme Court has held that while it is generally presumed that “‘identical words used in different parts of the same act are intended to have the same meaning,’ the presumption ‘is not rigid,’ and ‘the meaning [of the same words] well may vary to meet the purposes of the law.’”[8] Therefore, courts can analyze the context of even identical words used within the same act and, where appropriate, give them different interpretations.

Moreover, there are important distinctions in the relevant language and context of different sections of the Clean Air Act. The inclusion of requirements for listing pollutants as “criteria pollutants” that are not in the mobile source sections is a central one, supporting the view that a finding of mobile source CAA jurisdiction does not compel identification of greenhouse gases as criteria pollutants — and thus triggering the establishment of NAAQS and the corresponding State Implementation Plan requirements for stationary sources. Another is the mandate in the mobile source provisions that cost be considered, while consideration of cost is prohibited in establishing NAAQS. In short, EPA would have a colorable legal position if it chose to make an endangerment finding in the mobile source context, yet ducked listing greenhouse gases as a criteria pollutant and establishing a greenhouse gas NAAQS for stationary sources.

B. Hazardous Air Pollutants 

Another alternative explored by the ANPR is regulating greenhouse gases through the National Emission Standards for Hazardous Air Pollutants (NESHAPS), under Section 112. Hazardous air pollutants are those that present (or may present) a “threat of adverse” human health or environmental effects. The Act’s definition of a hazardous air pollutant is notably stricter and more targeted than the broad “endangerment” language applicable to the regulation of pollutants from motor vehicles in Section 202: Section 112 seeks to regulate toxic pollutants that are suspected of causing cancer, birth defects, or other severe health or environmental problems.

Unlike sections 108 and 202, there is no mention of “endangerment” in the definition of what constitutes a hazardous air pollutant. Thus, even if EPA renders an endangerment finding for vehicular greenhouse gas emissions, more analysis would need to be undertaken as to Section 112. Such an examination would likely demonstrate that some greenhouse gases, such as carbon dioxide, are commonplace and hardly threaten unborn children with developmental effects.

That conclusion would be consistent with Congress’ intent in drafting Section 112. Emitters regulated under Section 112 are subject to the Clean Air Act’s strictest and most onerous regime. It is unlikely Congress intended to regulate commonplace greenhouse gases across the nation under such a severe regime. In addition, unlike the broad-based regulation of national pollutants contemplated by Section 108, the hazardous air pollutants language applies to sources that emit as little as ten tons of a hazardous air pollutant per year. This suggests that Congress was trying to regulate an especially small number of highly dangerous chemicals — not commonplace pollutants such as greenhouse gases. It would require a considerable logical leap for EPA or the courts to place greenhouse gases — for example, the carbon dioxide we all exhale — under this provision.

Moreover, even if greenhouse gases could arguably fall within the Section 112 definition and be characterized as a hazardous air pollutant, the low annual emission threshold (ten tons per year) would trigger the regulation of a mind-boggling number of sources. In comparison, the House of Representatives Committee on Energy and Commerce Discussion Draft only covers sources that emit more than 25,000 tons per year[9], and the Senate bill, in general, only covered those emitters that used more than 5,000 tons of coal per year.[10] As a matter of public administration, the addition of the hundreds of thousands of new sources to be stringently regulated under Section 112 would render the whole program inherently unworkable. It thus seems unlikely that the Obama Administration would use this provision to regulate the “downstream” behavior of this many small sources of greenhouse gas emissions.

C. Performance Standards

The ANPR embraces a third section — Section 111 — as the most promising method to regulate greenhouse gases under the Clean Air Act. Section 111 of the Act provides EPA with authority to regulate emissions from new and existing sources if the pollutant is s criteria pollutant is not subject to NAAQS regulation. (This is generally described as the New Source Performance Standards [“NSPS”] program.) While the North Carolina decision likely precludes EPA from developing a cap-and-trade-based regulatory regime under Section 111, the Agency may consider economic and environmental impacts when establishing a NSPS for a pollutant. Consideration of cost and a realistic evaluation of the global nature of global warming would be essential elements in developing a workable regulatory regime.

It must also be kept in mind that, at least when considered on a carbon dioxide equivalent basis, only a handful of categories account for the vast majority of greenhouse gas emissions. Together, transportation and electric power generation make up 61 percent of domestic greenhouse gas emissions. Add industry, and 79 percent of the emissions is covered. Furthermore, a step-by-step implementation of NSPS regulations directed first at these three sectors, and in a manner that takes costs into account, might avoid the horrors detailed by the interagency commentators. On the other hand, it would both constitute action to address the greenhouse gas issue, and ratchet up businesses’ desire to see a better scheme — e.g., new cap-and-trade legislation. The Obama Administration may embrace this approach for these very reasons.

Why Would New Legislation Be Better?

Notwithstanding EPA’s potential ability to manipulate existing Clean Air Act authorities to address greenhouse gas emissions, the potential complexities described in the ANPR and various agency critiques are real. And, ultimately, the fact that EPA may have a legally defensible basis to use at least some portions of the existing Clean Air Act to address climate change doesn’t mean implementing rules will survive judicial review. A new statute, directly focused on the climate change problem, would be a far better mechanism.

This is not just a matter of simplifying management. The key piece that is missing from EPA’s existing authority is a way to address the inevitable result of a climate change regulatory scheme: the increased cost of energy delivered to its consumers. In light of North Carolina v. EPA, and under the existing Clean Air Act scheme, there is no way to maximize the efficiency of expenditures (another way of saying minimize economy-wide costs), provide relief to the individuals, families, or industries that cannot bear the new costs, or even allow for reasonable phase-in of those new burdens. Yet virtually every interest involved in the climate change debate recognizes some subsidization is required.

Regulation as a Tool to Break Legislative Deadlock

Nonetheless, it is not hard to see how — even with increased Democratic majorities — it will take years for cap-and-trade climate change legislation to emerge. We are hardly living in a time when legislators are anxious to impose new economic burdens on the nation. Moreover, the array of conflicting interests is huge, and, even with strong Presidential leadership and one-party control, finding consensus will be extraordinarily difficult.

But President-Elect Obama has committed to do something to address climate change. And, especially if the Congress remains reluctant to come to grips with the issue, proceeding with the greenhouse gas regulatory process is a threefer: it does something; it reverses one of the previous Administration’s policies that were most offensive to the environmental community; and the resulting regulations would threaten to be so costly and intrusive on so many interests that merely proposing action under existing legislative authorities encourages legislative action.

So, what would we expect? A proposal to use Section 111 of the Clean Air Act, where the first stationary source categories targeted will be those with the greatest greenhouse gas emissions: utilities, refineries, and some energy-intensive industries. Even the mere prospect of Section 111 regulation is likely to shake up the legislative debate, with not only companies in these sectors but virtually all others that use substantial quantities of energy clamoring for action. And that clamor will serve directly the Administration’s policy preferences.