December 24, 2010

EPA Marches On: Regulating Stationary Source GHG Emissions under the Clean Air Act

The environment received an early Christmas present from the Environmental Protection Agency yesterday, with EPA’s announcement that it would propose New Source Performance Standards (NSPSs) for greenhouse gas (GHG) emissions from power plants and refineries in 2011, and then finalize the regulations in 2012.  The decision resolves a lawsuit brought by states, local governments, and environmental groups. EPA’s initiative will impose cost-effective controls on stationary sources of GHGs, and complement the agency’s existing initiatives for mobile and stationary sources of GHGs.  While the CAA might not be as flexible or comprehensive as recently proposed congressional GHG legislation, EPA is making sorely needed progress to control the nation’s GHG emissions.

Notwithstanding industry’s ongoing criticism of applying the CAA to GHGs, the initiative is hardly a surprise.  The Supreme Court made clear in its 2007 Massachusetts v. EPA decision that GHGs are “air pollutants” subject to the Clean Air Act (CAA).  As such, EPA has unquestionable authority to regulate GHG emissions from stationary sources, and has decided to develop NSPSs for two of the nation’s largest industrial sources.  According to EPA, fossil fuel power plants and petroleum refineries emit 40 percent of U.S. GHG emissions. 

The NSPS requirement will establish industry-specific emission limitations for new and modified facilities in the affected industries.  The emission limitations will take the cost and availability of control options into account.  Although utilities have feared that EPA would require them to install expensive carbon capture and sequestration (CCS) technology, EPA is unlikely to require CCS given its high cost and lack of demonstrated success.  In fact, since there are few, if any, demonstrated technologies that could otherwise remove GHGs from the emissions stream, EPA is most likely to impose measures that would increase the facilities’ energy efficiency and indirectly limit GHG emissions.  While such measures could require initial capital expenditures, they could ultimately provide cost savings to the affected industries, just as installing a new and more efficient furnace in a home would cost money up front, but save money down the line.  It is also conceivable that EPA would require a new or modified facility to switch from a more GHG-intensive fuel (like coal) to a less GHG-intensive fuel (like natural gas).  A NSPS that required fuel-switching would be more controversial, but would likely achieve greater GHG emission reductions.

In developing the GHG NSPSs, EPA has indicated that it will take a comprehensive approach that considers not only GHG emissions, but also a range of other pollutants that are emitted along with GHG emissions.  EPA is currently in the midst of reconsidering the NSPS for these industries’ emissions of “criteria pollutants” (the most common and widespread pollutants, like particulates and sulfur dioxide) and other toxic pollutants, and intends to develop an integrated control strategy that more effectively and efficiently controls the many different pollutants mixed into a given facility’s emissions.

The NSPS effort will come on the heels of EPA’s initiation of GHG controls under the Prevention of Significant Deterioration Program (PSD), slated to begin in January 2011.  That program will require new and modified sources to achieve emission limits based on the “Best Available Control Technology.”  While there are many similarities between the two programs, there are also important differences.  Both programs apply only to new sources or to sources that are planning modifications to their plants that will significantly increase their emissions – for example, a significant expansion.  Both programs are likely to consider the same general types of control strategies.  But the thresholds for determining applicability differ in the two programs.  And the NSPS are determined by the federal government and applied uniformly across the country, while the BACT determinations are made on a case-by-case basis and usually handled by state permitting authorities (with help from federal guidelines).  The NSPS is generally considered a uniform minimum floor.  In contrast, the BACT determination under the PSD program is more case-specific and often (although not necessarily) more stringent than the NSPS.

Significantly, the recently announced settlement agreement goes well beyond NSPSs and the PSD program; it requires EPA to propose emission guidelines for existing sources, not just new and modified sources.  The PSD program, and the NSPSs themselves, do not regulate existing sources; they apply only to new sources and those engaging in modifications that significantly increase their emissions. But the Clean Air Act’s New Source Performance Standard section, in section 111(d), contains an additional requirement: If EPA sets a NSPS for a pollutant that has not been designated as a criteria pollutant or as a hazardous pollutant, then EPA must also create emission guidelines for existing sources of that pollutant.  Since EPA has not designated GHGs as hazardous air pollutants or as criteria pollutants, the duty to promulgate emission guidelines for existing sources has been triggered. Once issued, the states must then develop plans for implementing the guidelines.  While the states have some flexibility in adopting the emissions guidelines, the regulatory effort constitutes a major development in addressing GHG emissions from existing sources.

 Although industry has been strongly opposed to applying the CAA to GHGs, and certain features of the CAA, like the requirement to set ambient air standards, are arguably not a good fit for GHGs, the CAA provisions EPA is using are appropriate and sensible regulatory tools.  Comprehensive GHG legislation that supplemented the CAA would, of course, be preferable.  EPA itself has indicated that it would prefer new legislation to regulating under the CAA, and the Administration arguably hoped that its promise to regulate under the CAA would spur Congress to develop and pass alternative legislation.  But the fossil fuel industry has bitterly opposed comprehensive legislative proposals and that opposition is likely to increase in the new Congress.  Unless and until Congress emerges from its paralysis, the Clean Air Act’s regulatory tools provide the next-best alternative.

Regulating GHGs under the CAA differs considerably from the approach that would have been taken under recent legislative proposals for a federal cap-and-trade program.  Under the CAA, all covered facilities must adopt controls.  Under a cap-and-trade program, facilities would have had greater flexibility: those that could cost-effectively adopt controls would have done so, while those facing higher control costs could have chosen to purchase allowances to cover their emissions rather than reducing emissions.  At least in theory, a cap-and-trade program could thus have lowered the overall costs of control.

That said, the difference between the two approaches – regulating under the CAA and cap-and-trade – should not be overstated.  While the NSPS, PSD requirements, and emission guidelines will impose control requirements broadly and might impose controls on high-cost facilities that would otherwise have purchased allowances, the requirements will take the costs of control into account.  EPA is likely to require greater reductions at facilities that can reduce more cheaply than at facilities facing higher costs of control. 

A cap-and-trade approach would also have brought existing sources more securely into the fold.  Although EPA intends to issue emissions guidelines for existing sources that states will have to incorporate into implementation plans, those guidelines are likely to be weaker, and offer more flexibility, than the standards for new sources.  Under a cap-and-trade program, all sources, both existing and new, would have been subject to the constraints and reduction incentives created by the trading program’s decreasing emissions cap.

Rather than developing a principled alternative (or complement) to CAA regulation, the incoming Republican majority in the House of Representatives will undoubtedly attempt to strip EPA of its authority to regulate GHGs under the CAA, putting a stop to implementation of the PSD program and development of the NSPS programs.  If they succeed, the federal government would be powerless to initiate sensible, cost-effective mechanisms for reducing the stationary source GHG emissions that are slowly but surely distorting the climate.  President Obama appears committed to federal action and is likely to veto any such initiative.  But House Republicans are likely to resort to pressure tactics, like attaching language to an appropriations bill, or stripping funding for the development of the regulations from EPA’s budget.  The Administration will have to stand up to these tactics, and either protect EPA’s authority to address the urgent threats posed by climate change under the CAA or demand comprehensive legislation in return. 

(This blog post has been translated into Romanian, here.)


Alice Kaswan, CPR Member Scholar; Professor, University of San Francisco School of Law. Bio.

  • Read Comments (0)
  • + Add a Comment
Be the first to comment on this entry.

First Name:
Last Name:
We ask for your email address so that we may follow up with you, ask you to clarify your comment in some way, or perhaps alert you to someone else's response. Only the name you supply and your comment will be displayed on the site to the public. Our blog is a forum for the exchange of ideas, and we hope to foster intelligent, interesting and respectful discussion. We do not apply an ideological screen, however, we reserve the right to remove blog posts we deem inappropriate for any reason, but particularly for language that we deem to be in the nature of a personal attack or otherwise offensive. If we remove a comment you've posted, and you want to know why, ask us ( and we will tell you. If you see a post you regard as offensive, please let us know.