A Free-Market Energy Blog

Clean Air Act Regulation of CO2: Rough Road Ahead

By -- June 3, 2009

Even if energy realists and their allies fend off Waxman-Markey, wave after wave of global warming regulation could still sweep across the U.S. economy under the aegis of EPA and the Clean Air Act.  

As explained in a previous post, the carbon dioxide (CO2) litigation campaign that begat the Supreme Court’s Massachusetts v. EPA decision (April 2007) could shut down much of our economy and replace self-government via the people’s elected representatives with the rule of bureaucrats and courts.

Energy realists need to school themselves in this constellation of issues, because the clock is ticking. On April 17, the Environmental Protection Agency, responding to Mass. v. EPA, published a proposed rule concluding that greenhouse gas (GHG) emissions from new motor vehicles cause or contribute to health- and welfare-endangering “air pollution.”  The comment period ends on June 23.Once EPA finalizes the rule, commonly known as an “endangerment finding,” the agency will be required, under Section 202 of the Clean Air Act, to establish first-ever greenhouse gas emission standards for new motor vehicles. But because the Clean Air Act is a highly interconnected statute, with similar endangerment tests dispersed through multiple provisions of the Act, the Sec. 202 endangerment finding and associated emission standards will ignite a regulatory cascade.

Peter Glaser, an energy-realist attorney with Troutman Sanders, was the first to explain how EPA adoption of GHG motor vehicle emission standards would compel the agency to impose Clean Air Act pre-construction permitting requirements known as Prevention of Significant Deterioration (PSD) on vast numbers of previously unregulated buildings and facilities. Glaser also observed that the Sec. 202 endangerment finding prerequisite to establishing motor vehicle GHG emission standards would substantively satisfy the Sec. 108 endangerment test that initiates a National Ambient Air Quality Standards (NAAQS) rulemaking.

EPA’s July 2008 Advanced Notice of Proposed Rulemaking (ANPR), Regulating Greenhouse Gas Emissions under the Clean Air Act, provides a wealth of detailed information on potential EPA regulation of GHG emissions under numerous provisions and programs of the Act including PSD, NAAQS, New Source Performance Standards (NSPS), the Title V operating permits program, and the Hazardous Air Pollutant (HAP) program. We could end up with a Kyoto-on-Steroids regulatory regime without the people’s elected representatives ever voting on it.

The Massachusetts court never addressed these legal ramifications of a Sec. 202 endangerment finding, so the Justices did not see (or pretended not to see) that the case had massive implications for the future of U.S. energy markets, the economy, and the constitutional separation of powers.

Although the ANPR downplayed the economic risks and administrative difficulties of regulating GHG emissions under the Clean Air Act (see here and here), it nonetheless confirmed Glaser’s assessment that promulgating GHG emission standards under Sec. 202 would open a regulatory Pandora’s Box.

An important recent contribution to the discussion is The Road Ahead: EPA’s Options and Obligations For Regulating Greenhouse Gases by Inimai M. Chettiar and Jason A. Schwartz of New York University School of Law’s Institute for Policy Integrity. Weighing in at 181 pages with  777 footnotes, The Road Ahead rivals the ANPR in length and detail. An adequate review of this report would span many blog posts. Here I touch on a few key points.

Mass v. EPA could launch a new era of executive government.

Chettiar and Schwartz are zealous proponents of GHG regulation, and their treatise is clearly intended to guide the Obama administration in the months ahead. Energy realists who want to understand the other team’s playbook would do well to spend time with this report.

What Chettiar and Schwartz propose is nothing less than a new era of executive government in which the President establishes a mega-Kyoto system if Congress drags its feet:  

If Congress fails to act, President Obama has the power under the Clean Air Act to adopt a cap-and-trade system that auctions greenhouse gas allowances. President Obama also has the power under the Clean Air Act to implement an executive agreement at the international level, rendering Senate approval of a climate treaty unnecessary. (p. vii)

Chettiar and Schwartz even argue that EPA has the power to establish the emission allowance auction system that Obama had campaigned on. According to them, “an auction of allowances by EPA does not run afoul of the constitutional designation of the taxation power because an auction of allowances is not a tax–its purpose is not to raise revenue, but instead to affect behavior.” (p. xii)

By this logic, sin taxes on alchohol and tobacco are not taxes, because their purpose is to curb drinking and smoking, so FDA may impose them absent congressional authorization. Similarly, redistributive taxes are not taxes, because their purpose is to foster “social justice,” and carbon taxes are not taxes because their purpose is to “save the planet.” 

It’s like saying that fuel economy regulation is not really such if its purpose is to reduce GHGs. Hey, that’s what the California Air Resources Board said from day one about its fuel-economy/GHG emissions program; and under Obama, CARB is going to dictate federal fuel economy standards, even if it kills what’s left of the U.S. auto industry.

Energy realists take note: America is headed into a new era of legal nominalism in which any constitutional or legal constraint can be evaded just by renaming things. If Obama can’t line up 67 Senators to ratify the Kyoto II agreement negotiated in Copenhagen, no problem. He can “circumvent the legislature” (p. 4) by relabeling whatever is negotiated in Copenhagen as an “executive agreement.” (p. xii). More about this later.

EPA’s endangerment finding will start a regulatory chain reaction.

Chettiar and Schwartz confirm that EPA’s adoption of GHG motor vehicle emission standards will “automatically trigger other provisions in the CAA, generating a cascading effect that may produce some unanticipated or unwelcome consequences.” The standards will start a “chain reaction of regulations.” (p. 4)

EPA, for example, will have to act upon other Title II (mobile source) endangerment petitions similar to the one that culminated in Mass v. EPA. Petitions from California and/or various environmental groups are pending to regulate GHG emissions from marine vessels (October 2007), aircraft engines (December 2007), and nonroad engines and vehicles (January 2008). The authors’ assessment:

[O]nce EPA finalizes its proposed endangerment finding, it will be on an automatic course to make endangerment findings under the remaining sections of Title II . . . EPA will be locked into a path that ends in the regulation of most mobile sources. (32)

Other petitions seek to compel EPA to apply Best Available Control Technology Standards (BACT) to CO2 emissions from coal-fired power plants, and establish GHG performance standards (NSPS) for electric steam generating units. These, too, will likely prevail.

Chettiar and Schwartz summarize:

A chain reaction of regulatory obligations is embedded into the structure of the Clean Air Act: a single spark will set it off. In Massachusetts v. EPA, the Supreme Court handed EPA a match, and soon–once EPA finalizes its endangerment finding–the match will be lit. Other pending petitions and court orders will add fuel to the fire. (p. 19)

Cap-and-Trade Fuel Program

Going beyond the ANPR, Chettiar and Schwartz argue that EPA could establish an “upstream” GHG cap-and-trade program covering all petroleum-based motor fuels. Section 211(c) of the Clean Air Act says that EPA may, “by regulation, control or prohibit the manufacture, introduction into commerce, offering for sale, or sale of any fuel or fuel additive for use in a motor vehicle, motor vehicle engine or nonroad vehicle if, in the judgment of the Administrator, any fuel or fuel additive or any emission product of such fuel or fuel additive causes, or contributes, to air pollution … that may reasonably be anticipated to endanger public health or welfare.”

Chettiar and Schwartz comment:

It [211(c)] allows EPA to do more than just set emission standards; instead, it allows EPA to “control” and “prohibit” the manufacture–including importing or refining–or sale of fuel. (p. 32)

If their reading of this provision is correct, EPA could even ban the manufacture and sale of gasoline!

Like the ANPR, Chettiar and Schwartz understate the risk of a NAAQS rulemaking.

Chettiar and Schwartz agree that (1) a Sec. 202 endangerment finding would satisfy the Sec. 108 endangerment test initiating a NAAQS rulemaking, (2) the NAAQS for GHGs would likely be set below current atmospheric concentrations, turning the entire country into a non-attainment area, and (3) the nation could not come into attainment with a “primary” or health-based NAAQS within the required 10-year period. 

Indeed, even draconian emission reductions by all major emitting countries over many decades would not be sufficient to reduce atmospheric GHG concentrations. Logically, applying NAAQS to GHGs could require complete deindustrialization of the United States.

NRDC v. Train (1976) is the relevant court case. It held that EPA must initiate a NAAQS rulemaking if it determines that “air pollution” from numerous or diverse mobile or stationary sources may reasonably be anticipated to endanger public  health or welfare. 

Although the case remains “good law,” Chettiar and Schwartz argue that  courts now accord Chevron-type deference to an agency’s statutory interpretations of ambiguous language. (p. 34) Consequently, they contend, EPA may be free to read an ambiguity in Sec. 108 to mean that agency does not have to establish NAAQS if it does not “plan” to issue the associated regulatory analyses, known as “air quality criteria” documents. This won’t wash. Congress did not intend to make EPA’s NAAQS obligations discretionary at the whim of the Administrator. 

The authors also argue that the 1977 Clean Air Act Amendments “contemplate that there will be some pollutants that endanger public health or welfare but are not listed as criteria pollutants.” (p. 38) They cite Sec. 111(d), which allows EPA to apply NSPS to existing sources “for any air pollutant … for which air quality criteria have not been issued or which is not included on a list published under 108(a) or 112(b).”

However, the authors note, the only pollutants EPA regulates under NSPS but not under NAAQS are “designated pollutants”: sulfuric acid mist, fluorides, cadmium, and furans, as emitted by aluminum plants, paper mills, fertilizer plants, and solid waste incinerators. They opine that these pollutants  “are not emitted by enough ‘numerous or diverse’ sources to qualify as criteria pollutants under Section 108(a)(1)(B).” However, that consideration clearly does not apply to GHG sources. 

“Nevertheless,” the authors continue, “the potential for alternative options under the Act to regulate the same dangerous air pollutants renders inoperative the argument in Train that NAAQS must be obligatory because there is no alternative.” (p. 38) That, however, was not the argument in Train. On the contrary, the court in Train rejected the argument that EPA did not have to regulate lead under NAAQS because it had a better alternative: regulation under NSPS.

Chettiar and Schwartz also cite Sec. 122, added in 1977, which directs EPA to determine whether radiocative pollutants endanger public health or welfare, and if so, “EPA shall include such substances in the list published under section 108(a)(1) or shall include each category of stationary sources in the listed published under section 111(b)(1)(A), or take any combination of such actions.” (p. 38) The point being that having found endangerment, EPA may choose to regulate under NAAQS or NSPS.

I am dubious. Sec. 122 deals solely and specifically with “radioactive pollutants . . ., cadmium, arsenic and polcyclic organic matter.” Therefore, it does not create a general option for EPA to establish NSPS as an alternative to NAAQS for other health- or welfare-endangering air pollution from numerous or diverse stationary or mobile sources. To their credit, Chettiar and Schwartz admit that “the case for discretion is far from clear-cut.” (p. 39)

Chettiar and Schwartz go on to argue that EPA must consider the climate change impacts of ozone and particulate matter (PM) when the agency revises its NAAQS for those pollutants. “Such standards must be set at levels requisite to protect public health and welfare from the effects of climate change.” (p. 39) But if EPA has to revise NAAQS to take into account the climate change effects of ozone and PM, how could it not have a duty to establish NAAQS for gases with much greater potential to cause climate change?

As the authors acknowledge, “Though ozone can be a potent greenhouse gas, the exact effects of ozone on climate change are complex and somewhat uncertain.” For starters, unlike CO2, the heat effects of ground level ozone are local and short-lived. Ground-level ozone is simply not implicated in global climate change. It is hard to imagine that courts would compel EPA to consider the climate change impacts of ground level ozone in a NAAQS rulemaking and not require EPA to establish NAAQS for the long-lived, well-mixed greenhouse gases.

Like the ANPR, Chettiar and Schwartz downplay the PSD risks to small business.

Chettiar and Schwartz state that once EPA sets GHG emission standards for new motor vehicles, it must regulate “major emitting facilities” under PSD. In fact, the authors argue that EPA’s historical interpretation of “major emitting facility” contradicts the Act’s plain statutory language. Up to now, EPA has applied the PSD program only to “major” sources of pollutants that the agency controls under another part of the Act. However, section 169(l) defines “major emitting facility” as:

any of the following [28 specific] statutory sources of air pollutants which emit, or have the potential to emit, one hundred tons per year or more of any air pollutant . . . [and] any other source with the potential to emit two hundred and fifty tons per year or more of any air pollutant [italics added].

Regardless of whether CO2 is controlled under Sec. 202 or not, Mass v. EPA decided that GHGs are “air pollutants” under the CAA simply by virtue of being emitted. “Thus, acording to strict interpretation of the statute, any major source of such gases should already be a major emitting facility.” (p. 44)

I don’t agree–but then again I think the Court majority in Massachusetts got it wrong when they turned the Clean Air Act defition of “air pollutant” into an empty formalism whereby anything emitted per se qualifies as an “air pollutant” even if it does not pollute the air. If this reading were correct, then all emissions from major stationary sources would have to be controlled even if EPA were to determine that they pose no danger to public health or welfare.

The authors also acknowledge that applying PSD to CO2 would dramatically expand the universe of “major emitting facilities.”

A traditional 500 megawatt coal-fired plant is estimated to emit annually 114 pounds of lead, 720 tons of carbon monoxide, 10,000 tons of sulfur dioxide, but nearly 4 million tons of carbon dioxide. Even relatively small sources–such as indoor malls, many apartment buildings, large houses of worship, some restaurants, and even bakeries–have the potential to emit enough carbon dioxide to meet the threshold for major emitting facilities (44-45)

They estimate that an average-size commercial kitchen has the potential to emit nearly 600 tons of CO2 per year. “Thus, based on its potential greenhouse gas emissions, the restaurant will be considered a major emitting facility, will need a PSD permit, and will have to install BACT for all regulated pollutants (i.e., carbon monoxide, nitrogen oxides, sulfur dioxide, particulate matter, and so forth). (p. 45)

A report sponsored by the U.S. Chamber of Commerce estimates, based on fuel purchase data, that 1.2 million previously unregulated buildings and facilities emit 250 tons of CO2 per year. All would be vulnerable to new controls, monitoring, paperwork, penalties, and litigation.

Chettiar and Schwartz reject the ANPR’s suggestion that EPA could avoid imposing PSD requirements on small entities simply by raising the threshold from 250 tons to 10,000 or even 100,000 tons. That flouts the plain language of the statute. Neither, in their opinion, does the statute allow the ANPR option of defining “major” source based on tons of carbon-equivalent emissions (the hoped-for advantage being that it takes 917 tons of CO2 to reach 250 tons of carbon-equivalent CO2 emissions).

But they believe EPA could use actual emissions instead of potential emissions to identify major emitting facilities. (pp. 103-104) EPA, they suggest, could spare small entities regulatory anguish and expense under a “general permit” whereby the owners or operators would agree not to to run their heating or cooling systems 24 hours a day, 365 days a year.

There are several problems with cure for the PSD nightmare. First, the statute defines major emitting  facility in terms of potential to emit. Second, as noted above, an estimated 1.2 million entities actually emit 250 tons of CO2 annually. Third, PSD permits are based on BACT determinations that Sec. 165 says must be done “case-by-case,” which would appear to preclude general permits applicable to large numbers of entities regardless of their individual circumstances. Fourth, even if courts allow general permits, vast numbers of previously unregulated entities would still incur paperwork burdens unrelated to any measurable environmental benefit. 

The regulatory cascade includes pointless paperwork burdens under the Title V operating permits program.

Sec. 502 (a) makes unlawful the operation of any “major source” without a permit. A source is major under Title V if it has the potential to emit 100 tons per year or more of “any air pollutant.” The Chettiar and Schwartz cite the ANPR’s estimate that 550,000  sources emit at last 100 tons per year of CO2. The actual number is likely much larger since, as noted above, a credible study estimates that 1.2 million entities emit 250 tons per year.

The Title V program typically is not supposed to create new requirements for major sources but to facilitate compliance with other provisions by consolidating all of the source’s Clean Air Act requirements in a single permit. But once Title V extends to CO2, the vast majority of permits will be filed by entities with no other Clean Air Act obligations. The permits will contain no information beyond the fact that somebody filed it to avoid penalty under Title V. A more wasteful form of regulatory make-work is hard to imagine.

Chettiar and Schwartz note that Title V specifically allows EPA to administer a “general permit.” At best, however, this can only limit, not avoid, the imposition of irrational burdens on potentially millions of entities.

We don’t need no stinking treaties.

The authors argue that EPA and the President can use existing CAA authority to implement an international agreement without Senate ratification or congressional approval. (p. 68) President Obama is free to reach an agreement with other heads of state on climate matters, and international agreements “create law for the states parties thereto” regardless of how the agreement is reached. Chettiar and Schwartz’s controlling legal authority in this matter is the Vienna Convention on the Law of Treaties (art. 46.1, 1155, U.N.T.S. 3313)–a treaty, it turns out, that the United States never ratified

That aside, Obama cannot make Kyoto II the law of the land without the Senate’s advice and consent. Nor can Obama and congressional leaders use “executive-legislative” agreements to bypass the super-majority requirement for treaty ratification by the U.S. Senate.

In a typical “executive-legislative agreement,” simple majorities in both houses of Congress legislatively approve whatever agreement the President negotiates. While this may be constitutional with regard to agreements negotiated under Trade Promotion Authority, it is not an option with respect to the Kyoto II agreement being negotiated in Copenhagen.

Here’s why. As my colleague Chris Horner points out, Kyoto II is the successor treaty to Kyoto I, which is a protocol to the United Nations Framework Convention on Climate Change (UNFCCC). When the Senate ratified the UNFCCC, it stipulated that any efforts to bind the United States to further commitments would require its advice and consent to ratification. 

The Senate Foreign Relations Cmmittee, in its Treaties and Other International Agreements (p. 276), is crystal clear on this point:

On the Climate Change Convention, the Foreign Relations Committee also noted that decisions by the parties to adopt targets and timetables for limiting emissions would have to be submitted to the Senate for advice and consent. It noted further: that a decision by the executive branch to reinterpret the Convention to apply legally binding targets and timetables for reducing emissions of greenhouse gases to the United States would alter the “shared understanding”of the Convention between the Senate and the executive branch and would therefore require the Senate’s advice and consent.


Again, the foregoing touches only a few of the issues addressed in this hefty legal analysis. It is must reading for energy realists seeking to anticipate what Obama’s EPA may do if and when the regulatory chain reaction begins.

Use of Sec. 211(c) to establish a cap-and-trade program for petroleum-based fuels is an option I had not considered. Unlike Sec. 202, the provision does not require EPA to consider compliance costs or technological lead times. This suggests that EPA could regulate us “beyond petroleum” before commercially viable alternatives exist. 

But, you protest, Congress never intended for Sec. 211(c) to be used for that purpose! Well, if you think Clean Air Act regulation of CO2 has anything to do with congressional intent, then you have not been paying attention. The central objective of the CO2 litigation campaign is to “enact” regulatory measures that Congress has either repeatedly rejected or declined to approve. The prospect of a gasoline ban under Sec. 211(c) would be truly scary were it not for the fact that the political backlash would be swift and sure.

The politics of CO2 regulation are ultimately more important than the legalities. My hunch is that EPA will not be able to control the cascading effects of CO2 regulation, and that sooner or later the backlash will come. As a ’60s poet once put it, the darkest hour is just before the dawn.


  1. Can EPA ban gasoline? Can Obama enact Kyoto II without Senate ratification? | OpenMarket.org  

    […] school of Law’s Institute for Policy Integrity suggests as much. See my commentary on Masterresource.Org. « NYT: “Buy American” angers allies, loses […]


  2. Cooler Heads Digest 5 June 2009 | GlobalWarming.org  

    […] Clean Air Act Regulation of CO2: Rough Road Ahead Marlo Lewis, MasterResource.org, 3 June 2009 […]


Leave a Reply