The Death of Deference?

Yesterday, the Supreme Court granted cert. in several cases to hear the following question:

“Whether the Environmental Protection Agency unreasonably refused to consider costs in determining whether it is appropriate to regulate hazardous air pollutants emitted by electric utilities.”

The fundamental issue is whether it was unreasonable for EPA to interpret section 112 to preclude consideration of cost at this particular stage of the regulatory process — not only different from what the Court thinks is the best interpretation, but s position that no reasonable person could take.  The Supreme Court and lower courts have rarely found agency interpretations unreasonable in cases where the statute was ambiguous.  This is called the Chevron Step 2 analysis, while deciding whether the statute is ambiguous is called Chevron Step 1.  The rationales for the Chevron doctrine are that Congress meant agencies to work out statutory ambiguities and that it is better for politically accountable members of the executive branch to do that, as opposed to federal judges with lifetime appointments.

By conventional standards, EPA has a very reasonable argument against considering costs when deciding whether to include power plants in the hazardous substance regulations that apply to all other industries.  The applicable provision (section 112(n)(1) calls on EPA to report to Congress on whether power plant emissions threaten public health (which they do) and about alternate control technologies.  EPA then is required to regulate power plants “if the Administrator  finds such regulation is appropriate and necessary after considering the results of the study.”  That’s all very open-ended, and it seems entirely reasonable to interpret it to mean that EPA should regulate if the emissions in fact endanger public health (which they do).  It seems redundant to consider costs at this stage since costs will then be considered anyway at the next stage, when EPA actually determines how much control the industry needs it. Cost is not considered in including other industries in the regulations.  Moreover, as EPA points out, once power plants get included, they would have to stay included even if it later turned out that the regulations were ruinously expensive, because the provision governing delisting clearly does not allow consideration of costs.  EPA’s interpretation may or may not be the best one, but it doesn’t seem patently unreasonable. Courts, including the Supreme Court, have certainly upheld much less plausible interpretations of statutes by agencies.

The grant of cert. in this case suggests that there may be four Justices who are willing to change the approach to reviewing agency interpretations of statutes.  One possibility is that they want to put some teeth into Chevron Step 2 analysis.  “Reasonable” may turn out to mean something like “almost as plausible as the alternative interpretation.”  In that case, Chevron would mean only that the agency wins in cases where there’s a near tie between opposing interpretations of the statute.  Alternatively, there may be four Justice who want to hold that statutes should always be interpreted to allow consideration of costs, except when Congress clearly excludes cost considerations.  That’s a position that Cass Sunstein and others have advocated. In any event, the cert. grant seems to show that the some of the Justices are tired to deferring to agencies and think the executive branch needs to be reined in more.  That also fits with the Court’s decision to hear the Obamacare case this year, another case where the normal standards for granting review seem absent. Maybe some of the Justices have been listening to Mitch McConnell and John Boehner’s complaints about executive initiatives.

Some people have suggested that EPA’s refusal to consider cost in this setting is inconsistent with what the agency did in the EME Homer case, where it did consider cost in applying a provision of the Clean Air Act that was silent on the matter.  But there’s no inconsistency.  First, the point of Chevron is that the agency gets to choose either way when statutes are unclear.  T:he Court upheld the agency’s choice in EME Homer but that doesn’t mean that the agency was compelled to make that choice.  Second, in EME Homer the agency was trying to find the cheapest way of cleaning up the air.  Here, a particular industry is asking for a special exemption from the rules applying to all other industries so that it can continue emitting toxic substances and harming the public.  The two situations just aren’t comparable.

EPA’s analysis showed that implementing the regulation would prevent 11,000 premature deaths a year, not to mention preventing IQ losses among children.  (The lives saved aren’t directly because of mercury reductions; they’re because the control technology would also reduce particulates.)  Putting asie the legal technicalities, it would be tragic if the Court further delayed this important regulation.

, , , , ,

Reader Comments

6 Replies to “The Death of Deference?”

  1. Thank you for this thoughtful and thought-provoking post, Dan!

    One might say that the Supreme Court has already shown its willingness to limit agency deference and put some teeth into Chevron Step 2. In Utility Air v. EPA, the Court found EPA’s interpretation to be unreasonable at Step 2–a rare holding, indeed, as you describe. Time will tell whether Utility Air is an extreme case or a signal of a developing trend in the Court’s application of Chevron. Certainly, Utility Air should be read together with EME Homer.

    In general, I think the Step 2 reasonableness inquiry is distinct in important ways from the common law construct of the reasonable person. Chevron v. NRDC itself used “permissible” and “reasonable” interchangeably to describe a valid interpretation. Courts determining the reasonableness of an agency’s interpretation have historically evaluated whether the interpretation represents a rational means to achieve the broad goals of the statutory program, taking into consideration, e.g., the agency’s technical expertise, whether the interpretation is longstanding, and legislative history. Some lower courts and scholars have even suggested that the reasonableness inquiry does or should resemble arbitrary and capricious review under the Administrative Procedure Act, or more specifically, the “hard look” review of State Farm. Additionally, as Utility Air demonstrates, a Step 2 inquiry can trigger the “major questions canon,” which prohibits courts from extending deference to agency decisions of significant economic or political significance where Congress has not spoken clearly. All of this seems to suggest, I think, that “unreasonableness” has more specific meaning in the context of Chevron Step 2.

    1. Hi Megan. I think you’re right that this isn’t an isolated development, as shown by UARG. In talking about degrees of deference, attitudes may matter more than how the test is articulated. In the past, courts generally gave agencies a lot of leeway when statutes were ambiguous to decide such things as how to balance the purposes of the law. But since the factors used to determine reasonableness are essentially the same factors that a court would consider to interpret the statute independently, the real question is how much room the court will give the agency to make those determinations. It seems to me that a defeat for EPA in this case would bring Chevron much closer to Skidmore than it has normally been in the past. The dissent in the court of appeals seems to have taken the other route by insisting that the word “appropriate” necessarily requires the consideration of costs and that no reasonable person could think otherwise.

      I also wonder whether the way the Court has approached these recent cases supports the views of those who argue that Chevron has only one step, that of reasonableness. The Court seems not to have distinguished very clearly in its cert. grants between the two stages, and in the UARG opinion it also seems to fudge the distinction.

      Thanks for your thoughtful response to the post.


  2. Hi Dan (and Megan),

    One question, one thought:

    Question: I take the 112(n)(1)(a) “necessary and appropriate” determination to be analogous to an endangerment finding. The determination was made in 2000. An attempt to delist was made, unsuccessfully in 2005. Why is this even open to challenge nearly 15 years later? Isn’t it a little late? Is the issue one of ripeness because there was such a long delay in getting final regs setting limits?

    Thought: Reading the petitioners briefs, it’s easy to believe that the court might have been seriously misled about the nature of this rule. In particular, the industry briefs say that costs are $9.6B and benefits are $4-6M. Painting the rule as seriously out of whack when it comes to balancing costs and benefits.

    But they *forgot* to mention the $36-90B in health benefits associated with reduced PM emissions – pretty important co-benefit of the rule. Not to mention the fact that a whole bunch of benefits of Hg pollution are not monetized (cancer risks, lower IQ, etc). The reply brief from the SG focuses on the legal issues but doesn’t really touch the RIA issue.

    That’s covered in a separate brief by Sean Donahue.

    If the facts were as the industry petitioners assert, maybe it would seem a little unreasonable – a factor of 1000 difference in costs and benefits not weighed in a consideration of what is or is not appropriate. It makes you kind of wonder if they took the time to read the amici’s briefs. Or if the real issue here is an attack on the idea that a rule’s cobenefits can be considered in a RIA. But I thought that was settled.


    1. Mike,

      The reason the “appropriate and necessary” finding is still alive as a judicial issue, all these years later, is that there’s an unusual provision in section 112 – specifically, section 112(e)(4) – that says that a source category listing decision is not final agency action subject to judicial review, “except that any such action may be reviewed under such section 7607 of this title [the judicial review provision] when the Administrator issues emission standards for such pollutant or category.” An early suit challenging the 2000 determination was thrown out by the D.C. Circuit for precisely that reason: the “appropriate and necessary” determination, which is tantamount to a listing decision under 112(c), wasn’t final agency action. The issue didn’t get resolved in the 2005 challenge to the Clean Air Mercury Rule because that rule was invalidated on a threshold ground: EPA hadn’t properly delisted power plants. So, only now, with the Mercury and Air Toxics Standards out there, are the conditions for judicial review of the 2000 determination finally satisfied.

      This will be an interesting one to watch.

      Best regards,


  3. Our new leadership in Congress has definitely shown a willingness to limit EPA deference and so have many ordinary citizens who despise this crooked agency’s abuse of its regulatory authority. Reasonable people recognize that the EPA is corrupt
    because it routinely issues false, misleading and inflammatory propaganda. The EPA is not worthy of anyone’s deference. Hopefully it will soon be de-funded and downsized.

Comments are closed.

About Dan

Dan Farber has written and taught on environmental and constitutional law as well as about contracts, jurisprudence and legislation. Currently at Berkeley Law, he has al…

READ more

About Dan

Dan Farber has written and taught on environmental and constitutional law as well as about contracts, jurisprudence and legislation. Currently at Berkeley Law, he has al…

READ more