The following discussion board post is by Mark Moller, a senior fellow at the Cato Institute in Washington. The Cato Institute filed as an amicus supporting the respondent in Mass v. EPA.

Earlier today, a commentator on NPR suggested that the Court has raised a “high bar” for the EPA, if the EPA wants to decline, again, to regulate global warming after remand. Is that right?

Not really.

“Regulation,” under the Clean Air Act, can take a number of forms: It can take the form of declaring aspirational emission standards. Or it can take more draconian forms, such as looming technology mandates and imminent implementation deadlines, backed by tough civil and criminal penalties.

Even assuming that, after today, the EPA has to “regulate” in the sense of promulgating some GHG emission standards, the Court’s decision leaves the EPA with ample room to argue that it can defer deciding when and how to implement those standards in light of the potentially high and uncertain costs of implementation.

It’s true, of course, that some parts of the Clean Air Act prohibit the EPA from undertaking this sort of cost-benefit analysis. The parts of the CAA governing auto emission standards are, however, different. There, the EPA retains considerable discretion weigh costs and-benefits—particularly when it comes to the “when” and “how” of implementing emission controls. For example, as Justice Stevens notes, section 202(a)(2) of the CAA gives the EPA broad discretion to delay implementation of pollution controls to the extent that “the Administrator finds necessary to permit the development and application of the requisite [pollution control] technology, giving appropriate consideration to the cost of compliance within such period.” Put in plain English, that means that if the “costs” of developing effective pollution-reducing technologies are very large, and the pay off of this R&D is in the far-distant future, the CAA doesn’t require the EPA to implement its standards right away.

The Court’s opinion also reaffirms the great deference owed to the EPA’s decision not to enforce any standards that it might promulgate. In the words of Justice Stevens today, an “agency has broad discretion to choose how best to marshal its limited resources and personnel to carry out its delegated responsibilities.” Given the breadth of discretion granted the agency to defer implementation under provisions like section 202(a)(2), and the costs and uncertainties associated with implementation, that deference may gave the EPA very substantial room to defer—perhaps for a very long time—implementation of a federal GHG enforcement regime, freeing the EPA to deal with more immediate and pressing environmental problems.

Nor is analysis of the EPA’s leeway to delay implementation much different if, as some assume, the Court’s decision means that GHG emissions are also “pollutants” under CAA provisions dealing with “national ambient air quality standards.” True, in Whitman v. American Trucking Association, the Court held that the EPA must set NAAQS without regard to the costs of implementation. But in his concurrence in that case, Justice Breyer suggested that even CAA requirements governing national ambient air quality standards permit some modified cost-benefit analysis. He emphasized, for example, that when setting NAAQS, the EPA doesn’t have to eliminate “any health risk, however slight, at any economic cost, however great.” It is only required to eliminate “unacceptable” risks, defined as those that the public is not willing to tolerate at any cost.

New American car emissions count for only 6% of worldwide carbon dioxide emissions. Eliminating these emissions wouldn’t necessarily reverse global warming or even appreciably slow it—particularly given the dynamic nature of emissions in developing countries. Thus, it’s far from evident that the added global warming risks created by new American car emissions are “unacceptable” in the sense suggested by Justice Breyer. On the face of the record, it’s also far from clear that the risks posed by other GHG-omitting sources in the U.S., such as stationary sources, are any more publicly “unacceptable” in the sense meant by Breyer, given uncertainty about the payoff of unilateral American remediation and given the cost and current feasibility of GHG control technology.

Ultimately, then, the key flaw with the EPA’s decision may not have been the outcome of that decision, or even the overarching reasons given by the EPA for its decision. The fatal flaw may have been only the conclusory nature of the reasons given by the EPA for its decision. For example, the EPA said that it wouldn’t act now because effective GHG-reducing technologies weren’t feasible at present and wouldn’t be feasible in the near future. But the EPA didn’t make any effort to quantify, or otherwise support with evidence, that feasibility assessment. Instead, it offered its conclusions as facts that courts must accept at face value—something five justices weren’t willing to do. But if the EPA can supplement its feasibility conclusions with at least some evidence, it may be able to pull at least one or two justices—most likely Breyer or Kennedy–into the dissenters’ orbit.