It’s simple: ‘Chevron’ must go

A few weeks ago, the U.S. Supreme Court held three hours of oral argument about what to do with the statutory interpretation rule arising out of the now 40-year-old Chevron v. NRDC decision, 467 U.S. 837 (1984).

Chevron told courts to defer to an agency’s reasonable interpretation of a silent or ambiguous statutory provision. As was reported in Lawyers Weekly (“Lawyers expect rollback of ‘Chevron’ — one way or another,” Jan. 29), many of those who listened to the argument thought the court was likely to pare back the rule but not eliminate it.

There is no good reason for such a half measure. Chevron is based on the legal fiction that if a statute is ambiguous or silent on a point, then Congress must have intended (clumsily) to give the agency that must implement the statute discretion to adopt any reasonable interpretation of it. There is no evidence whatsoever that this is true.

Moreover, this legal fiction was wholly unnecessary to solve the interpretation problem presented by the Chevron case itself, and it can cause mischief when applied in other situations, including in cases like the two cases on tap before the Supreme Court.

There is a much simpler way to figure out whether an agency has such discretion: just analyze that issue without the Chevron assumption. Take the Chevron case itself. The supposedly difficult interpretation problem presented was whether the Reagan-era Environmental Protection Agency could implement a Clean Air Act provision dealing with “stationary sources” by considering an entire multi-building plant as the stationary source, rather than an individual furnace or boiler or a single factory building with multiple furnaces or boilers.

In the air pollution context, the phrase “stationary source” is not vague at all. It means a non-moving air pollution source. It is silent as to which of the options I just mentioned the EPA can adopt, but the silence is hardly difficult to evaluate. The Clean Air Act differentiates between mobile and stationary sources.

Chevron is based on the legal fiction that if a statute is ambiguous or silent on a point, then Congress must have intended (clumsily) to give the agency that must implement the statute discretion to adopt any reasonable interpretation of it. There is no evidence whatsoever that this is true.

That Congress defined “stationary source” no further shows that it left it up to the EPA to decide whether to go small by focusing on each furnace, go big by looking at an entire plant, or go somewhere in the middle by focusing on each factory building that emitted air pollution. Thus, there was no need for any novel assumption to figure out what stationary source meant.

The Chevron decision went on to say that the EPA’s discretionary decision to regulate by looking at the whole plant was reasonable. On its face that makes little sense because, if this is one of the approaches Congress allowed, why need it be separately evaluated to see if it is reasonable? This evaluation would seem more likely relevant to deciding a substantive challenge to a regulation.
Just a year prior to deciding Chevron, the court had determined that an agency faced with a substantive challenge to a regulation must demonstrate a rational basis for adopting it. What is notable about Chevron is that it simply ducks the issue. The controversial aspect of the EPA’s action was that it allowed a plant owner to install a new air-polluting furnace and offset any additional pollution this caused by an equal reduction in pollution elsewhere in the plant. This was allowed even in states that were not meeting air quality standards.

How can it be rational to allow a manufacturer in such a state to install a new furnace without requiring it to reduce the plant’s overall emissions? The EPA’s answer was that the main way Congress anticipated that states would come into compliance with Clean Air standards would be when manufacturers adopted newer equipment that was less polluting; hence its approach encouraging the installation of new equipment that would more quickly reduce emissions.

Conflating a reasonable statutory interpretation with the substantive reasonableness of the regulation helped the court avoid having to figure out whether the EPA’s approach was substantively reasonable. However, when applied to actual statutory interpretation issues, this approach has been used by federal courts to justify even truly boneheaded agency interpretations so long as the agency could offer them with a straight face.

Take Chemical Manufacturers Association v. NRDC, 470 U.S. 116 (1985). The Clean Water Act limits discharges of pollutants into waterways and allows the EPA to make exceptions in only two narrow instances. Later on, Congress amended the act to provide that these exceptions could not include variances from the toxic substances release standard. Somehow the EPA took this to mean that, in all instances, other than the two exceptions provided in the Clean Water Act, the EPA could allow toxic substances discharge variances.

Although Congress had provided no exceptions other than the two listed in the statute, the court accepted the EPA’s strained interpretation because the statute was too complex for ordinary mortals to understand.

Had the court actually examined whether the EPA had discretion to make this call, it would have seen that this was not an instance in which the agency had at least two viable options available: it either had the authority to issue variances in this situation or it did not.

This is unlike Chevron, in which the agency could pick from multiple options that were consistent with the basic meaning of the statute. One of them must be true and the other not. Had the court recognized this in Chemical Manufacturers Association, it would have had to decide which interpretation was correct, not defer abjectly to the agency’s “expert” opinion.

As for deference, a closer look at Chevron reveals there was no deference involved. It was not as if the EPA was saying that the Clean Air Act compelled it to regulate stationary sources by focusing on entire manufacturing plants and the court was required to defer to this interpretation.

Rather, the EPA made two claims: the phrase “stationary source” gave it options, and the whole-plant approach it chose was one of them. The Supreme Court agreed with it on both points, deciding that the agency’s view was consistent with the statute’s language, and the choice it made was one of the choices Congress gave the agency in implementing the statute.

This is the way it should work. If there is a dispute about whether some vague statutory provision gives an agency some discretion in approaching a problem, the first thing a court should do is determine whether such discretion exists. If it does, then the court must determine whether the agency’s approach is within the limits of that discretion.

If, on the other hand, the court determines that a statutory provision, even if expressed inarticulately by Congress, requires that an agency follow one approach, then the court must determine what singular approach Congress intended.

In none of these situations is an agency’s policy preference relevant. All that matters is what Congress directed or allowed the agency to do.

A court should recognize that in implementing a statute, an agency may have gained extensive experience in discerning the underlying congressional intent. This experience and perspective should be considered seriously, but there is no reason for a court to defer to it completely, which is the kind of deference associated with Chevron.

The two cases before the Supreme Court, Loper Bright Enterprises v. Raimondo and Relentless, Inc. v. Dept. of Commerce, involve a National Marine Fisheries Service rule that required the herring industry to pay the cost of federal observers on fishing boats.

There was once a live dispute about whether the agency had any such authority. The herring fishers maintain that no such authority exists because elsewhere in the statute the agency is allowed to charge for observers on other types of fishing fleets, but not on herring fleet boats.

The agency responded that if herring fleets chose to hire private observers on their boats, the statute requires the fleets to pay those observers, so why not for federal observers on herring boats?

This is the perfect example of a debate in which only one side can be right. It cannot be that on one day Marine Fisheries has a right to charge herring fleets, and on another day it lacks such authority. It must be one or the other. The choice in this instance is not all that difficult, as Marine Fisheries stopped charging herring fleets for on-board observers after two years and refunded the money it had already received.

The fact that the court did not then determine these two cases to be moot shows that it intends to make a significant decision about Chevron deference.

The two cases before it are as good an example as any to demonstrate why it is important for a court to examine first whether any agency has discretion to determine how it will implement a particular statutory provision. Only if it does must a court then go on to determine whether what the agency did fits within the limits of that discretion.

James P. Rooney is first administrative magistrate for the Division of Administrative Law Appeals, an adjunct professor at New England Law Boston, and author of “Rethinking Chevron” (57 New England Law Review 119 (2023)). The opinions expressed above are his alone.