Justices Thomas and Gorsuch Argue for Rejecting Deference to Agency Interpretation of Agency Regulations

An interesting dissent from denial of certiorari this morning.


From today's Garco Construction, Inc. v. Speer, a dissent from denial of certiorari by Justice Thomas, joined by Justice Gorsuch:

Petitioner Garco Construction, Inc. … had a contract with the Army Corps of Engineers to build housing units on Malmstrom Air Force Base. As part of its contract, Garco agreed to comply with all base access policies. After construction began, the base denied access to certain employees of Garco's subcontractor.

Although the text of the base's access policy required only a "wants and warrants" check, the base clarified that the policy also required background checks and excluded many individuals with criminal histories—even if those individuals did not have any wants or warrants. Garco's request for an equitable adjustment of the contract was denied, and the Armed Services Board of Contract Appeals denied Garco's appeal.

The Court of Appeals for the Federal Circuit affirmed. Despite acknowledging "some merit" to Garco's argument that 'wants and warrants'" means only wants and warrants, the Federal Circuit deferred to the base's interpretation of its access policy under Auer v. Robbins, 519 U.S. 452 (1997).

Garco filed a petition for certiorari, asking whether this Court's decisions in Auer and Bowles v. Seminole Rock & Sand Co., 325 U.S. 410 (1945), should be overruled. I would have granted certiorari to address that question.

Seminole Rock and Auer require courts to give "controlling weight" to an agency's interpretation of its own regulations. To qualify, an agency's interpretation need not be "the best" reading of the regulation. Decker v. Northwest Environmental Defense Center, 568 U.S. 597, 613 (2013). It need only be a reading that is not "plainly erroneous or inconsistent with the regulation." Although Seminole Rock deference was initially applied exclusively "in the price control context and only to official agency interpretations," this Court has since expanded it to many contexts and to informal interpretations.

Seminole Rock deference is constitutionally suspect. It transfers "the judge's exercise of interpretive judgment to the agency," which is "not properly constituted to exercise the judicial power." It also undermines "the judicial 'check' on the political branches" by ceding the courts' authority to independently interpret and apply legal texts. And it results in an "accumulation of governmental powers" by allowing the same agency that promulgated a regulation to "change the meaning" of that regulation "at [its] discretion." This Court has never "put forward a persuasive justification" for Seminole Rock deference.

By all accounts, Seminole Rock deference is "on its last gasp." Several Members of this Court have said that it merits reconsideration in an appropriate case [citing opinions by Justices Alito and Thomas, and by Chief Justice Roberts]. Even the author of Auer [Justice Scalia] came to doubt its correctness.

This would have been an ideal case to reconsider Seminole Rock deference, as it illustrates the problems that the doctrine creates. While Garco was performing its obligations under the contract, the base adopted an interpretation of its access policy that read "wants and warrants" to include "wants or warrants, sex offenders, violent offenders, those who are on probation, and those who are in a pre-release program." The Federal Circuit deferred to that textually dubious interpretation.

Thus, an agency was able to unilaterally modify a contract by issuing a new " 'clarification' with retroactive effect." This type of conduct "frustrates the notice and predictability purposes of rulemaking, and promotes arbitrary government."

True, the agency here is part of the military, and the military receives substantial deference on matters of policy. But nothing about the military context of this case affects the legitimacy of Seminole Rock deference. "The proper question faced by courts in interpreting a regulation is … what the regulation means." While the military is far better equipped than the courts to decide matters of tactics and security, it is no better equipped to read legal texts. Pointing to the military's policy expertise "misidentifies the relevant inquiry."

Because this Court has passed up another opportunity to remedy "precisely the accumulation of governmental powers that the Framers warned against," I respectfully dissent from the denial of certiorari.

For more on this question, see this post by Jonathan Adler.

NEXT: D.C. Councilman Warns About "Rothschilds Controlling the Climate" "to Own the Cities"

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Help me out, legal eagles, where does the Chevron decision come into all this?

    1. Auer is the extension of Chevron to agency interpretations of their own internal regulations. Chevron is about agency interpretations of statutes. Parts of the constitutional argument made by Gorsuch and Thomas would seemingly apply with equal force to Auer and Chevron.

      1. So if they overruled the *Auer* decision, the Chevron decision would wither and die?

        1. Not necessarily. It depends on what grounds Auer was overturned.

          1. So this may not be as big as I initially thought.

    2. Chevron is even more constitutionally suspect than Seminole Rock/Auer deference. Chevron refers to judicial deference of an agency’s interpretation of laws, whereas this is only deference to an agency’s interpretation or regulations.

      The issue is they sometimes overlap because the regulations may fit perfectly into the law, but the interpretation of the regulation could be a bad fit for the law. Overturning Chevron would necessarily narror Auer.

      1. No, Chevron is less constitutionally suspect, because it’s not ad hoc and ex post facto. In Chevron the fiction is that Congress delegated to the agency the authority to make these regulations; the agency issues the regulation pursuant to notice and comment and then it is in effect. But with Auer, the agency writes a regulation and then says, after the fact, “When we said X we really meant Y and Z. We never told you that before, but gotcha; that’s what it means.”

        1. Right, there seems to be greater issues with notice and Due Process (although the non-delegation issues remain the same).

    3. They seem to me polar opposites. An administrative agency should be expected to know quite well what its regulations mean, since it wrote them (although I would argue for the rule of lenity, for the same reason that contracts of adhesion are always interpreted to favor the party that didn’t write them). But allowing an agency to decide what powers Congress delegated to it is putting the fox in charge of guarding the henhouse.

  2. Even if there is majority support on the court for overturning Seminole this is a good case to duck. It would be “constitutionally suspicious” (to quote Gorsuch) for the Supreme Court to declare who can and cannot step foot on military bases. If the United States military decides that group [x] shouldn’t get on American military bases, it’s not going to allow [x] on military bases. And if the need to keep [x] off the base is great, they aren’t going to bother with the notice and rule-making procedures, either.

    1. I presume they could adopt a rule that convicted (insert name of crime)ers can’t have access. But it doesn’t seem like they’ve adopted such a rule yet.

      1. It takes time to adopt a rule. There are notice requirements, comment requirements, minimum periods of time for same, etc. If the day after 9/11 military bases decided to have stricter requirements for who could or could not access them, they’re not going to issue a notice and comment rule. They’re going to stop people at the gates. SCOTUS can’t do anything about that.

        1. They can make the agency bear the cost of the disruption of a contract thereby!

    2. I agree with this. The Courts have no business second guessing the military on security concerns.

    3. Overturning Seminole in this case wouldn’t put the Supreme court in charge of “declare(ing) who can and cannot step foot on military bases.” It would simply require them to reimburse the plaintiff for retroactively altering the terms of the contract.

      1. That doesn’t strike me as a good use of the public fisc.

        1. Is the public fisc better served by arbitrary and capricious rule changes or by rule of law?

          I can see both sides but I suspect as the time horizon changes the benefits of the former will dissolve while the benefits of the latter will increase.

          1. I’m not entirely sure. Part of me assumes that since the law for 7 decades has been that the US military can do what it wants, when it wants, the risk is baked into the contractor’s bid. But I also think there’s potentially some value to allowing military bases to make decisions about who to let on the site without having to go through the administrative rule-making process, and also without having to pay government contractors for changing policies.

            1. I find it difficult to believe that military base access policy requires notice and comment rulemaking, and nothing in this case suggests that it does.

              1. The issue here is that the company bid for the job based on one set of base access rules, and then the military tried to hold them to the bid after changing the rules barred a large fraction of their workforce from entering the base.

                And they tried to justify doing it on the basis that the rules hadn’t really been changed, the company had just misunderstood them. Which was total BS.

                1. Exactly. And the contractor’s bid is based in large part on the cost of its subcontracts. Let’s say the military determined that the “wants and warrants” rule now means that the subcontractor’s employees had to have top-secret security clearances. Employees with clearances are far more expensive than employees without, so the change means tat the contractor bid a different contract than the military is enforcing. Not cool.

              2. If they don’t have to follow notice and comment rulemaking to change access to the base, the criticism that the after-the-fact change is unfair goes away, since the military base was always in a position to retroactively change its own regulation without notice. That’s another reason why this case is a poor vehicle for cert.

                1. I don’t know why you think it follows that if N&C isn’t required that they can retroactively change a rule incorporated into a contract.

                  1. Maybe it’s a kind of dynamic incorporation, by reference to a term that can change, rather than to a snapshot of one at the time of the agreement. But who has that much trust?

                  2. This isn’t the harshest result I’ve seen in federal contracting. I’m also not married to Auer. But I maintain that this isn’t the appropriate vehicle for changing Auer because bad facts.

    4. That misses an important point, I think. Garco was not demanding the right of group [x] to set foot on an American military base. The basis of the suit is that “Garco’s request for an equitable adjustment of the contract was denied” despite the company having attempted to comply with the contract as it was originally written.

      We do not know from the above what “equitable adjustment” was requested but it could have been something as simple as a few extra days to find and hire replacement workers. Or perhaps, relief from a late penalty caused by this unannounced change in the rules.

      In other words, the Army can kick whoever they want off the property but that doesn’t (or shouldn’t) automatically excuse them from a breach of contract penalty.

      1. Thomas and Gorsuch were demanding to reverse the rule that the military gets to interpret its own regulations as to who is or isn’t permitted on base. While that may be massively unfair for Garco (who might have simply been seeking time, or going after Eichleay general office overhead delay damages, I don’t know), it’s still a terrible case for the Supremes to take up re: optics. If you want to build consensus for reversing a 73 year old decision, probably don’t pick the case that is fundamentally about how the military is going to regulate invitees.

      2. We do not know from the above what “equitable adjustment” was requested

        Extra money. The plaintiff — or rather, its subcontractor — was getting much of its labor force from “a local prison’s pre-release facility,” whatever that means exactly. (They had been doing this for many years on this base, apparently.) As a result of the base’s new policy, the subcontractor allegedly “incurred nearly half-a-million dollars ($454,266.44) of additional expenses from additional time interviewing and hiring new workers, paying overtime to new workers, and training new and less experienced workers.”

        (They blatantly did not request extra time.)

        1. Okay, they wanted money. Let’s change the scenario a little bit and see how it plays out.

          The Army puts out a bid for hammers of a given spec. The specs define the weight, length, shape and handle material. You put together a bid and offer to sell the Army steel-headed hammers that meet all their specs for $10 each. After you start work, the contracting agency says “nope, we need the heads to be made out of gold”. Maybe there’s a perfectly good reason why they really do need gold-headed hammers. Higher density, non-sparking, plan to use them in an insanely corrosive environment, I don’t know and it doesn’t really matter. That’s the new requirement and it’s news to you as the contract winner. Is it fair for the Army to still hold you to that $10 a hammer price? If not, why is the Garco request different?

          1. The chief difference, is that to make this like the Garco request, you’d have to add that the actual specs specified that the hammer be made of heat treated steel. And then the Army simply interpreted “steel” to mean “solid gold”, after the fact.

            From the OP: ” While Garco was performing its obligations under the contract, the base adopted an interpretation of its access policy that read “wants and warrants” to include “wants or warrants, sex offenders, violent offenders, those who are on probation, and those who are in a pre-release program.”

            They didn’t change the specifications. They ‘interpreted’ the specifications to mean something that they quite clearly didn’t mean, and had never been understood to mean.

            And that’s a power which the Supreme court reserves for itself…

      3. The ASBCA decision held that the sovereign act doctrine shielded the government from liability after the updated text was promulgated in Oct. 2007 and Garco did not dispute that in its appeals (though a dissent at the Federal Circuit suggests they should have). The complaint was that they began enforcing the more restrictive interpretation before that date. This is where the deference comes in, because the Board ruled that the updated text was a clarification of the policy rather than a modification, and that the sovereign act doctrine applied equally to the earlier memoranda and was not waived by the government’s earlier failure to enforce the rules.

    5. Although that interpretation should have been in the original PWS, SOW, etc. so all contractors know the requirement. Garco should have received compensation for the change just as any other change order to a contract.

  3. Regardless of the merits of this particular case, administrative deference should not be lightly cast aside. Supreme court justices are necessarily generalists — they are excellent at articulating broad constitutional principles and very good at construing statutes, but they are more likely to misconstrue an administrative regulation because they lack sufficient familiarity with the arcane system of which the regulation is a part. Deference is not abdication, and the courts must step in when a regulation is unconstitutional or clearly collides with the statutory scheme created by Congress, or when an administrative body uses the guise of “interpretation” to abrogate a regulation created by an earlier set of administrators appointed by a different chief executive. But deference appropriately acknowledges the expertise of administrative agencies, which often are better equipped than the judiciary to determine the meaning of their own rules.

    1. but they are more likely to misconstrue an administrative regulation because they lack sufficient familiarity with the arcane system of which the regulation is a part.

      The public, whether expert or layman, is bound by these regulations. Facing, often, civil and criminal penalties if they act in the wrong way. If the regulation is written in such a way that judges can’t construe it “correctly,” then how can the public?

      1. The public isn’t supposed to be able to construe the regulation correctly.

  4. Seminole Rock deference is constitutionally suspect. It transfers “the judge’s exercise of interpretive judgment to the agency,” which is “not properly constituted to exercise the judicial power.”

    What could that possibly mean? I suggest it could mean that at least some on this Court intend to assert that the political branches are not to be granted scope to be final arbiters of any questions of fact or policy whatever. That would make it an announcement of a virtually unbounded judicial power grab in the making. Stay tuned.

    1. That would make it an announcement of a virtually unbounded judicial power grab in the making.

      “It is emphatically the province and duty of the Judicial Department to say what the law is.”

      1. Yeah. So what?

        I was talking about fact, policy, and the political branches?and speculating about whether the SC is now about to assert something like:

        “It is emphatically the province and duty of the Judicial Department to say what the law, the facts, and the policy are.”

      2. Nieporent, that clarion chestnut of yours is an example of the kind of thing everyone hears over and over, until it wears a groove in their brain, and then they just go, “Of course, I know that,” and never reflect on it again.

        But if someone dissented, and asserted instead that, “It is emphatically the province and duty of the Congress to say what the law is,” but, “It is emphatically the province and duty of the Judicial Department to say what the law was at the time of the case or controversy under consideration,” why wouldn’t that be a more accurate presentation of this nation’s constitutional order?

        1. My point was that claiming that it’s a power grab “in the making” ignores the history of our country.

  5. I think a difficulty here is that the Court has historically been highly deferential to the military on questions like what constitutes a security threat. Thus, under traditional interpretations, the military would have been allowed to control who has access to its bases and deny access to people it deems a security threat independently of anything it might have said in a contract.

    The idea that this issue can even be characterized as being about contract interpretation seems a very strained reading in light of the military’s historic independent authority.

    And that historic independent authority would seem to make this case a very bad one for resolving the issue Gorsach wants resolved here. If the military has authority to exclude people from its bases as security threats independently of what its contracts say, then resolution of the contracts issue would not provide the contractor any relief, rendering any an opinion an advisory one and precluding Article III jurisdiction for want of standing.

    1. Actually, this should have been a simple change order to the contract with appropriate compensation for the change. Although I would be interested in seeing the contract as most contractor’s personnel are either escorted the entire time while on base and/or the personnel are provided temporary access with associated background checks.

  6. Regulations are useless if an agency can unilaterally add terms like this, especially to a contract already in operation. It’s as if there were an “etc.” in invisible ink at the end of every phrase!

Please to post comments