Peter Glazer’s Testimony, on the EPA’s Response to Supreme Court’s Decision in Massachusettes v. EPA

By | March 22, 2008

For the Full Report in PDF Form, please click here.

[Illustrations, footnotes and references available in PDF version]


I am Peter Glaser, a partner in the law firm of Troutman Sanders LLP. I have an active Clean Air Act (CAA) practice and have been involved in greenhouse gas (GHG) legal issues for more than a decade. I represented clients in all phases of the Massachusetts v. EPA litigation, including filing comments in the original 1999 rulemaking and amicus briefs before the Court of Appeals and the Supreme Court. I have written and spoken about the decision on a number of prior occasions.

I am not here before the Committee representing or advocating the position of any particular company or industry. I am not receiving remuneration from anyone for my testimony, and the views expressed in my testimony are my own and not necessarily those of any company or group that I currently represent or have represented.

In addition, I am not here to recommend any particular course of action by this Committee or Congress. I have been asked to offer my views as a practicing attorney on issues pertaining to the U.S. Environmental Protection Agency’s (EPA) approach to addressing GHGs in the wake of the Supreme Court’s decision in Massachusetts v. EPA.


Let me begin by saying that I think the Committee has identified two of the most critical issues confronting EPA as it considers its response on remand to Massachusetts v. EPA. First, what would be the impact on stationary sources if EPA regulates motor vehicle GHGs under the CAA. Second, what impact does last December’s Energy Independence and Security Act (EISA) have on EPA’s response to the Court’s decision. These questions raise difficult issues, and the answers counsel caution and deliberation by EPA as it considers how to respond to the Court’s remand.

First, EPA regulation of motor vehicle GHG emissions under the CAA will have a very significant effect on stationary sources, and not just on large stationary sources. If EPA promulgates motor vehicle GHG regulations, CO2 and other GHGs will become regulated CAA pollutants for purposes of the CAA Prevention of Significant Deterioration (PSD) program. As explained in more detail below, if CO2 becomes a regulated CAA pollutant, then (a) no new “major” stationary source of CO2 emissions can be built without first obtaining a PSD permit and complying with CO2 Best Available Control Technology (BACT) requirements and (b) no existing “major” stationary source can undertake a modification that increases its CO2 emissions by any amount without first obtaining a PSD permit and complying with CO2 BACT requirements.

As also explained below, because the emissions threshold for stationary sources to be considered “major” is so low, hundreds of thousands (if not millions) of relatively small GHG-emitters would be swept into the PSD program if GHGs become CAA-regulated pollutants. Buildings of about 100,000 square feet, if they are heated by oil or natural gas, would likely become subject to the program, as would relatively small users of natural gas such as commercial kitchens that use natural gas for cooking, or businesses that use CO2 naturally as a component of its operations. A very large number and variety of buildings and facilities could therefore become subject to the program – including many office and apartment buildings; hotels; enclosed malls; large retail stores and warehouses; college buildings, hospitals and large assisted-living facilities; large houses of worship; product pipelines; food processing facilities; large heated agricultural facilities; indoor sports arenas and other large public assembly buildings; restaurants; soda manufacturers; bakers, breweries and wineries; and many others. None of these types of sources has ever been subject to PSD permitting requirements before because they emit so little traditional air pollution; but they would be if CO2 becomes a regulated CAA pollutant.

The economic consequences of this outcome could be devastating, particularly as the economy slows, because PSD permitting is an incredibly costly, time-consuming and burdensome process. Just the administrative burden alone – putting aside any control technology requirements that would result from the permitting process – would create a significant and unprecedented roadblock to new investment for a host of previously unregulated buildings and facilities.

In light of these significant impacts on stationary sources, the Committee’s second question – the effect of the EISA on potential EPA regulation in response to Massachusetts v. EPA – is particularly relevant. Much attention has been focused on whether or not EPA had planned to issue a motor vehicle GHG proposal at the end of last year. But given enactment of the EISA and given the PSD impacts discussed above, EPA’s decision to pause before doing so made and continues to make perfect sense. The EISA will achieve the goals of the President’s “20 in 10” agenda by dramatically reducing GHG emissions from motor vehicles, and it will do so, in large part, through the CAA and explicitly without triggering PSD impacts. Moreover, the EISA obligates EPA to issue implementing regulations within one year of the statute’s enactment. EPA has appropriately decided to focus its resources on fulfilling this statutory mandate. By giving precedence to implementing EISA, EPA can achieve the purpose for which the Massachusetts case was brought – reduction of motor vehicle GHG emissions – while it continues to consider how best to avoid the negative PSD impacts.

EPA’s pause also makes sense because time is needed to implement the EISA, gauge its impact, and assess what additional regulation may, or may not, be effective. Given EISA’s aggressive requirements, attempting to obtain further GHG reductions from motor vehicles may prove futile. On the other hand, if EPA were to press forward with some type of additional regulation of motor vehicle GHG emissions under the CAA, serious PSD consequences would ensue across the economy for a myriad of small sources. The result would be great pain for potentially little gain.

Moreover, EPA was and is well within its legal authority to pause before formulating a response to Massachusetts v. EPA. The Supreme Court did not establish a deadline for EPA action on remand. To the contrary, the Court stated that “EPA no doubt has significant latitude as to the manner, timing, content, and coordination of its regulations with those of other agencies.”

In my view, EPA cannot be ready to pursue a rulemaking on remand of Massachusetts until and unless it gives detailed consideration to these issues. Indeed, the onus should be on advocates of an immediate response to Massachusetts to explain the risks and rewards of moving forward aggressively. During the briefing in Massachusetts, petitioners maintained that the case was limited to prospective regulation in the automotive sector and that the Court need not consider how a decision in petitioners’ favor would reverberate throughout the CAA to other types of sources. But now that the case is back before EPA, the agency and stakeholders cannot act as if a decision to regulate motor vehicle GHG emissions has no impact on other CAA programs – or that Congress did not separately enact an aggressive program in the EISA to reduce motor vehicle GHG emissions under the CAA. The effect of these developments must be carefully considered.


A few words of background on the PSD program may be helpful. The PSD program was adopted by Congress in 1977 and applies in all areas of the country where existing ambient air quality is better than the National Ambient Air Quality Standards (NAAQS). Although the NAAQS sets a maximum allowable level of a pollutant in the ambient air, Congress decided that in existing clean air areas the air should stay cleaner than the NAAQS, and for that purpose adopted the PSD program.

Under the PSD program, permits must be obtained before construction may begin on “major” new stationary sources of CAA-regulated air pollutants. The CAA lists 28 specific types of stationary sources, such as power plants, refineries, steel mills, chemical plants, etc., that are “major,” and subject to the PSD program, if they can emit at least 100 tons per year (tpy) of any regulated air pollutant. Other, unlisted types of stationary sources do not trigger PSD permitting as “major” sources unless they can emit at least 250 tpy of any air pollutant. The term “stationary source” is very broad. It includes “any building, structure, facility or installation” which emits or may emit a regulated pollutant.

Also, once a facility is “major,” a change to that facility is subject to preconstruction PSD permitting if the change causes a “significant” emissions increase. EPA’s regulations numerically define a “significant” emission increase for a number of pollutants. For instance, an increase of particulate matter emissions of 25 tpy, or of sulfur dioxide or nitrogen oxides emissions of 40 tpy, is considered a “significant” increase. For pollutants for which EPA has not provided a numerical “significance” definition, such as CO2, any emission increase is considered to be a “significant” increase.

For a “major” source, the CAA requires BACT for each pollutant which is “subject to regulation” under the Act. BACT is determined on a case-by-case basis as the maximum emission reduction achievable, taking into account energy, environmental, and economic impacts and other costs.

The PSD program is largely implemented through a state-administered permitting system. Seven states administer the program through “delegated” authority from EPA; they essentially act as EPA’s agent in administering EPA’s PSD permit requirements. On the other hand, forty-three states administer their own PSD programs, for which EPA regulations prescribe the minimum CAA requirements. These states must first promulgate their own revised PSD regulations in their SIPs. Those revised SIPs must then be submitted to EPA for approval. In a few instances, such as a project being located on Native American lands, EPA itself directly administers the PSD permit system.

The 100/250 tpy threshold for PSD applicability was established by Congress based on emission levels of traditional pollutants, such as particulate matter, nitrogen oxides and sulfur dioxide. Emissions above this threshold were considered to be significant enough to trigger a need to regulate these pollutants. The PSD-triggering threshold was not set based on the premise that 100/250 tpy is a significant enough level of CO2 emissions to justify regulation. CO2 is not like traditional pollutants for a number of reasons, one of which is that 100 or 250 tpy are not a great deal of CO2. Although the 100/250 tpy level for traditional pollutants generally limits PSD permit requirements to large stationary sources like coal-fired electric generators, chemical plants, refineries and the like, a 100/250 tpy threshold for CO2 will subject a massive number of small facilities to PSD requirements. Yet that will be the result if EPA regulates motor vehicle GHG emissions and CO2 and other GHGs thereby become regulated CAA air pollutants.

The PSD burden caused by a 100/250 tpy applicability threshold for CO2 could be overwhelming for small and large businesses alike. New sources emitting more than 100/250 tpy of CO2 could not be built without first obtaining a PSD permit after undergoing the BACT process. Existing sources that emit more than 100/250 tpy of CO2 that wish to expand or modify their facilities in a way that would increase CO2 emissions by any amount would likewise first have to obtain a PSD permit after undergoing the BACT process. PSD permitting is complicated, time-consuming and expensive. No small business requiring a moderate-sized building or facility heated with fossil fuel could operate subject to the PSD permit administrative burden.

The requirement that sources emitting more than 100/250 tpy of CO2 apply BACT would also inject considerable, and perhaps fatal, uncertainty for businesses. No one can say at this time what BACT is for CO2 because there is no precedent or guidance. BACT is determined through a case-by-case evaluation of control technology alternatives and involves a complicated weighing of economic, environmental, energy and other factors. BACT can even be no control measure if that weighing process fails to identify a technically and economically feasible technology for controlling the pollutant in question. But since BACT determinations for CO2 have no regulatory history at this time, and can vary by type of facility and from state-to-state, businesses wishing to construct new sources or modify existing ones would have no basis for planning what the regulatory requirements will be.

The consequences of GHGs becoming CAA-regulated pollutants would also be experienced by state PSD-permitting agencies and by EPA. These agencies are wholly unprepared for the flood of PSD permit applications that would ensue. These permitting agencies would either have to reassign scarce resources from other environmental programs to handle the permitting burden, resulting in a decline in environmental regulation in these other areas, or PSD permitting would becomes so backlogged as to effectively create a permitting moratorium.
EPA recognizes this potentially catastrophic PSD implication for small sources if and when it adopts GHG regulations in response to Massachusetts. It may be considering ways to attempt to prevent very small sources of GHG emissions from becoming subject to PSD as a result of whatever motor vehicle CO2 regulations the Agency adopts. Trade press has speculated on several possible alternatives, all of which pose significant legal issues.

Creative solutions by EPA would be welcome. On the other hand, courts have not always been hospitable to creative interpretations of the CAA that do not adhere closely to the statutory text. The 100/250 tpy threshold is statutory; EPA’s ability to get around it seems doubtful. From a business standpoint, because legal uncertainty disincents capital investment, business will have difficulty relying on whatever alternative EPA might formulate (if it does) until the legal issues are resolved in court.

Moreover, putting aside the legal issues, if EPA ultimately adopts a mechanism limiting the effect of a decision to regulate GHGs on small sources, that mechanism may not be immediately effective in most states. As previously discussed, seven states essentially act as EPA’s agents in administering the PSD program, and the mechanism EPA adopts will immediately become effective in these states. However, the forty-three states that independently administer their own PSD programs under EPA supervision must likely undertake their own rulemakings to adopt EPA’s mechanism or possibly a more stringent mechanism (one that subjects a broader range of CO2 stationary sources to PSD regulation) in their SIP. A regulatory gap may therefore exist for sources in these states, after EPA has adopted its new mechanism. In these states, until the state also adopts a mechanism in its SIP and the state’s SIP revision is approved by EPA, sources may continue to be subject to the state’s current PSD regulations.

As can be seen, EPA regulation of motor vehicle GHG emissions will creates serious issues for a multitude of small sources. EPA must carefully consider these issues before it responds to the Massachusetts v. EPA remand.


In Massachusetts v. EPA, the Supreme Court decided that GHGs qualify as CAA “air pollutants,” but that decision in and of itself has no regulatory consequences. As the Court found, the CAA definition of “air pollutant” is “sweeping” and “embraces all airborne compounds of whatever stripe.” The Court held that an EPA obligation to regulate is triggered only if the agency finds that GHGs may reasonably be anticipated to endanger public health or welfare.

As stated in Justice Scalia’s dissent, EPA now has three choices. It can make an endangerment finding and regulate. It can make a non-endangerment finding and not regulate. Or it can decline to do either. As the Court said in the majority opinion, EPA can decline to regulate “if it determines that greenhouse gases do not contribute to climate change or if it provides some reasonable explanation as to why it cannot or will not exercise discretion to determine whether they do.” According to the Court, this discretion must be exercised “within defined statutory limits” and cannot “rest[] on reasoning divorced from statutory text.”

The Court did not establish a timetable for EPA to respond on remand. As noted, the Court stated that EPA has “significant latitude” as to the timing of regulatory action. Although EPA has been criticized for not yet having issued a regulatory proposal on remand, the remand from the Court of Appeals was not issued until September 14, 2007. Given the complexities and overwhelming importance of the issues involved, it should not be surprising that EPA has not yet issued a proposal. I don’t think it can seriously be argued that EPA has violated any kind of legal duty by not yet having responded to the remand.

Evidently, many expected that EPA would issue a proposal last December. That expectation was based on regulatory activity taking place at EPA in response to the President’s expressed desire to implement his “20 in 10” program, first announced in the 2007 State of the Union, by the end of 2008. But the President’s “20 in 10” program was addressed by the EISA. That legislation required EPA to issue renewable fuels regulations in conformity with the statute within one-year of the statute’s enactment. EPA has now turned its attention to meeting Congress’ deadline, and that activity must take precedence over a response to the Massachusetts remand.

Indeed, the EISA represents an aggressive motor vehicle reduction program under the CAA and, as an outgrowth of the President’s “20 in 10” agenda,” likely resembles the program that EPA was considering late last year. The legislation establishes minimum renewable fuels standards for transportation fuels sold or introduced into commerce in the United States, including minimum requirements for advanced biofuels, cellulosic biofuels and biomass-based diesel. All of these renewable fuels must meet standards for lifecycle GHG emission reductions. Under this mandate, the use of renewable fuels will increase by 500 percent, with fuel producers required to supply at least 36 billion gallons of renewable fuel in the year 2022. Significantly, the renewable fuel standard was enacted as an amendment of the CAA and will be implemented by EPA. In addition, the legislation creates a national mandatory fuel economy standard of 35 miles per gallon by 2020, which will save billions of gallons of fuel and increase efficiency by 40 percent. Together, these provisions will significantly reduce GHG emissions from the automotive sector.

In light of EISA, the immediate need to respond to the Massachusetts v. EPA remand has dissipated considerably. While the Administration may previously have believed that including a response to the Massachusetts case in its “20 in 10” implementing regulations made logical and legal sense, enactment of the EISA makes the Massachusetts remand less important both to the “20 in 10” program and to obtaining GHG emission reductions from the automotive sector through the CAA. Just implementing the EISA will be difficult enough – trying to do more at this time may be largely futile.

Of course, the EISA does not render Massachusetts v. EPA a legal nullity. At the appropriate time, EPA will have to respond. But even petitioners in Massachusetts recognize that EPA has considerable discretion to prioritize its own docket and decide what and when to regulate. In fact, as set forth in the following colloquy between Justice Ginsburg and James R. Milkey, Assistant Attorney General for the Commonwealth of Massachusetts, during the Supreme Court oral argument, petitioners took the view that EPA would be subject only to a narrow legal challenge if it decided not to regulate motor vehicle GHG emissions at all because of more pressing docket priorities:

Justice Ginsburg: But if you are right and then it went back and the EPA then said, well, an obvious reason also is constraint on our own resources, we have the authority to say what comes first, Congress – we couldn’t possibly do everything that Congress has authorized us to do; so it’s our decision, even though we have the authority to do this, we think that we should spend our resources on other things.

Suppose they said that? You said they didn’t say it this time around, but how far will you go if all that’s going to happened is it goes back and then EPA says our resources are constrained and we’re not going to spend the money?

Mr. Milkey: Your Honor, while background administrative law principles provide EPA at leas some room to move, we think it’s important that EPA say that. If they – it’s a very different opinion if they say, we are not going to regular [sic: regulate] here because we just don’t want to spend the resources on this problem and we want to look elsewhere.

If they want to say that, they can say that and the, if at all, there’d be a narrow arbitrary and capricious challenge on that.

If, as petitioners’ counsel admitted, at most only a narrow legal standard would govern an EPA decision not to regulate because of more pressing regulatory priorities, then surely EPA cannot legally be forced into an immediate decision on remand of Massachusetts. This is particularly true given that EPA’s reason for deferring immediate action on remand is because it is pursuing a CAA program to reduce motor vehicle GHG emissions, one that was just enacted by Congress and must by statute be implemented within one year. In sum, EPA has good reason to carefully deliberate how to respond to Massachusetts, and it is well within its legal authority in doing so.


EPA has violated no legal duty by not having responded to the Massachusetts remand at this time. Given the EISA and given the PSD impacts of motor vehicle GHG regulation on stationary sources, EPA should take its time in carefully deliberating an appropriate response.
I appreciate the opportunity to provide this testimony.