Cooking Up A Mess

Judicial Review of Presidential Action after Trump v. Cook

Trump v. Cook denied the President’s application for interim relief against the reinstatement of Lisa Cook to the Board of Governors of the Federal Reserve System. The more I look at the decision, the stranger it appears. It looks to be one of those decisions that may well be productive of great uncertainty and indeed mischief for years to come, well beyond the immediate context of the case. The majority opinion raises and hastily resolves, or at least gestures at resolving, a number of fundamental questions about judicial review of presidential action, while silently leaving other equally fundamental questions to the lower courts, whose track record of sobriety and impartiality as to review of actions by President Trump leaves so very much to be desired. These questions involve what causes of action may be brought against the President, judicial review of the procedures of presidential decisionmaking, standards of review for presidential determinations, and standards of proof, among many others. Especially striking, as I will discuss at the end, is that Cook very arguably contains the Court’s strongest assertion to date of a loose, unstructured judicial competence to review presidential findings and conclusion for “pretext” - an assertion that is potentially destabilizing in the extreme, and whose dangers the Court probably ought to consider more fully.

At first blush, Cook seems merely a limited coda to Trump v. Slaughter, the decision broadly declaring independent agencies unconstitutional. The Court’s decision in Cook to carve out a continued quasi-independence for the Fed from the sweeping general principle of the unitary executive confirmed in Slaughter was at least entirely predictable, whether or not defensible. The problem in Cook is not so much that central holding, but the congeries of ill-defined arrangements for judicial review of presidential action that the Court threw into the later parts of its opinion. Let me list a few of the issues and problems the Court’s discussion suggests, problems that could well be at issue not only in later litigation in this controversy, but in other controversies and settings involving judicial review of presidential determinations.

Causes of action. As the dissenters noted in Cook, it is by no means obvious what the cause of action was, or even whether one existed. The Court seems remarkably casual about this question, breezily mentioning an “equitable” cause of action in a footnote — even as, in different-but-related contexts, it has taken pains in recent years to restrict so-called “ultra vires” actions and “equitable” actions against executive officers. To be fair, however, this opaque and casual approach to suits against the Presidency has also been on display in other recent cases.

Overall, the impression the Court creates in this body of caselaw is not a happy one. The impression is of a body that sometimes takes great pains over causes of action and worries about the proliferation of nonstandard claims and unstructured equitable actions, yet that also indulges itself by proceeding with ill-specified causes of action when, it seems, the case is one that the Court feels called upon to decide. And the overall pattern seems backwards in a broad sense; one would think that, if anything, causes of action should be more freely available against subordinate executive officers and less freely available against the President, the head of a co-equal branch.

Statutory and constitutional due process. The Court says that its “narrow” holding is that “the President failed to afford Cook the procedural protections to which she was entitled by statute,” and denies that it is even addressing Cook’s argument based on “constitutional due process.” That is on odd thing to say on its face, as the for-cause provision at issue makes no mention of procedure. Hence the Court went on to discern, implicit within the for-cause provision, a settled common-law background such that a statute limiting discharge to “cause” also carries with it required procedures. As the Court puts it: “[T]he rule, we explained in 1901, is that notice and hearing are essential before an officer’s removal where the term of office is for a fixed period” (emphasis added).

The Court, however, immediately pulled back from the full implications of this, adding: “Of course, that is not to say that a Federal Reserve Governor is entitled to an audience with the President or a full-blown judicial trial…. Instead, all that is required is notice to the officer of the charges made against him and an opportunity to be heard in his defense.” And it further added that “[w]e see no reason why that opportunity [to be heard] may not be had on written materials only, with no oral presentation…. All that is required is the right to support [the] allegations by argument however brief, and, if need be, by proof, however informal, before a final decision is made. Londoner v. City and County of Denver, 210 U. S. 373, 386 (1908)” (internal quotation omitted).

If all this sounds suspiciously similar to constitutional review for procedural due process as to administrative officers, despite the Court’s disclaimers, that’s because it is suspiciously similar. Londoner of course is a constitutional due process case, not a statutory case, and the Court also cited to a famous article by Judge Friendly about the adequacy of administrative hearings under procedural due process. Will we soon have judicial review of the adequacy of the opportunity for a hearing afforded by the President? It’s hard to see why not, whether that is put in nominally statutory or constitutional terms; such review will prove irresistible to district judges somewhere or other. Will we even perhaps have Mathews v. Eldridge cost-benefit balancing review of the procedures used in the presidential hearing? Will the Court, or the lower courts, continue to claim with a straight face that judicial elaboration of procedures binding on the Presidency is all part of the common-law background of “for cause” provisions? Will these novelties spill over to other statutory determinations by the President? To my knowledge there is no previous body of caselaw inquiring into the adequacy of the procedures used in determinations made directly and personally by the President himself, as opposed to the adequacy of procedures used in decisionmaking by subordinate executive officials or tribunals, as in Hamdi v. Rumsfeld or Hamdan v. Rumsfeld. We are in terra incognita.

Standards of review. What is or are the standard or standards of review for the President’s determination of cause? After Loper Bright, of course, the abstract definition of “cause” seems to be a strictly legal question decided de novo by the courts, as Cook says. Yet the issue becomes far more difficult as to the application of the statutory standard, however defined, to the facts. Post-Loper decisions have offered broad scope for judicial deference on “mixed questions” of the application of law to fact; would those apply if, say, the President were to reach a determination of “cause” after some kind of hearing? It is hard to see that the substantial evidence standard would apply, as the hearing required by the Court does not seem to be a formal adjudicative hearing on a defined record. (Or is it?) So does arbitrariness review apply? Should a district judge with no legal competence or practical experience in executive management be empowered to say that the President has applied the statutory standard of cause unreasonably or arbitrarily?

Standards of proof. What about the strictly factual components of the decision? Suppose Governor Cook claims during the hearing that the President just has the facts clearly wrong — what standard applies? Preponderance of the evidence (which is the ordinary standard for administrative proceedings)? Or will we perhaps have the spectacle of District Judge NeverTrump deciding that the President of the United States must determine the facts bearing on cause by “clear and convincing evidence,” a standard sometimes used in other high-stakes administrative proceedings? Will we have evidence introduced in the district court to make or dispute such a showing?

Pretext review. To the administrative lawyer, perhaps the most startling passage of them all, and the most pregnant with mischief, ran as follows (with my italics and some citations omitted):

Whether “cause” for removal exists in any given situation will depend, at least in part, on the seriousness of the alleged misconduct, and the extent of any nexus that may exist to the Governor’s professional duties. The key issue is whether “[t]he cause assigned” truly “impl[ies] an unfitness for the place”—or whether it simply represents an effort to secure a “more congenial” replacement…. “Our review is deferential, but we are not required to exhibit a naiveté from which ordinary citizens are free.” Department of Commerce v. New York, 588 U. S. 752, 785 (2019) (internal quotation marks omitted). Without such constraints in place, any perceived or alleged misstep (past or present) could provide a ready pretext for a Governor’s removal.

This is not the first time the Chief Justice has spoken in somber, elliptical but portentous terms about pretextual action by the Presidency, or perhaps by just this President. The tariff case, Learning Resources Inc. v. Trump, is another example. Indeed a hermeneutic of suspicion seems to be a more general feature of the Chief Justice’s jurisprudence. In the passage I have just quoted, the citation to the Department of Commerce case is revealing. There the Chief Justice wrote for the Court to read arbitrariness review under the Administrative Procedure Act as allowing some review for “pretextual” administrative action.¹ Although the discussion in Cook is too elliptical to allow any firm conclusions, it is possible that the Chief thinks that pretext review under the APA transposes easily to review of presidential action, despite the APA’s inapplicability to the President and the serious constitutional questions that would be raised by any attempt to apply the APA to the President. In other words, pretext review under the Department of Commerce case represents an interpretation of a specific statutory standard, the “arbitrary and capricious” standard of review embodied in the APA, which does not bind the President in the first place. It is quite remarkable to casually generalize this to the setting of presidential action.

And how is “pretext” to be determined exactly? Will the district court be allowed to look behind the face of the President’s determination? To conduct discovery as to the internal procedures the President used, or the facts he knew, or his political communications inside the executive branch? Ordinarily, as to subordinate officers, that sort of inquiry requires a “strong preliminary showing of bad faith or improper behavior” and is thus extremely difficult to obtain. Does that same standard apply here? Or a higher standard? The Court says nothing at all about such crucial matters. It is not difficult to imagine the potential Cook creates for district-court shenanigans, intrusive attempts at discovery, interbranch conflict, mutual recriminations, and deepening discord, thanks to the majority’s rather casual endorsement of pretext review against the Presidency, or the President. The Court proceeded in a kind of haste, sketching an unprecedented framework of review in fragmentary terms while leaving critical elements of that framework entirely opaque. It may well have an opportunity to repent at length over the ensuing years.

The presumption of regularity. I will conclude with a point about the so-called presumption of regularity in this setting. Many American legal academics are passionately convinced that no such presumption should attach to this President, to his administration, or both. The Chief Justice, it seems, tends to agree, and sometimes — although by no means always — manages to assemble a coalition of his colleagues to say so. What to make of this stance?

I believe that it would help to distinguish what we might call two different types of justification for the presumption of regularity: a principled rationale and a political rationale. My friend and New Digest colleague Jeremy Christiansen has written powerfully about the principled rationale for the presumption from a classical perspective, grounded in the classical attribution of (quite possibly fictional and idealized) public-spirited aims to those who hold public office. Although I find Christiansen’s analysis entirely convincing, I want to add a less elevated rationale for the presumption, rooted in the spectre of political conflict between branches and in the vulnerable position of the judiciary vis-a-vis the executive.

On this rationale, the presumption represents a kind of Westphalian Peace between the branches, a nonaggression pact, based on the thought that the weapons that the judiciary turns against the Presidency can also be turned back against the judiciary itself. If the Court, or even a district judge, may declare presidential action “pretextual” and thus invalid, what if anything prevents the President from declaring a judicial decision (especially at the lower levels of the court system) “pretextual” and thus invalid? The Court might be wise to consider whether it is destabilizing a fragile peace that the judicial branch ultimately needs more than the President does.

A version of this very point was made long ago by Justice Frankfurter in United States v. Morgan (1941), aka “Morgan IV,” the case that cemented in doctrine the rule that it takes a strong preliminary showing of bad faith faith or improper behavior to obtain direct discovery from executive officials for the purpose of looking behind the face of executive action. For Frankfurter, the key principle was that the judiciary should do as it would wish to be done by, keeping in mind that what it does to the executive may be done to the judiciary in turn, and that if the judiciary wants its official acts to be respected, it must afford the same comity to the executive. Rejecting an attempt to take the deposition of a cabinet secretary, Frankfurter wrote for the Court:

[T]he Secretary should never have been subjected to this examination. The proceeding before the Secretary "has a quality resembling that of a judicial proceeding." Such an examination of a judge would be destructive of judicial responsibility. We have explicitly held in this very litigation that "it was not the function of the court to probe the mental processes of the Secretary." Just as a judge cannot be subjected to such a scrutiny, compare Fayerweather v. Ritch [citation omitted], so the integrity of the administrative process must be equally respected.

Fayerweather v. Ritch had involved an attempt to introduce in-court testimony from a judge about the real basis for his public decision. Frankfurter suggests that if the judicial branch does not want judges to be put on trial, literally or metaphorically speaking, it would do well to observe a certain reciprocity of forbearance vis-a-vis the executive. One hopes that the Court will take this into account before further generalizing its hermeneutic of suspicion as to Presidential action.

1. For the record, I think this is a deeply mistaken understanding of arbitrariness review, which should be understood to speak only in objective terms — that it, to require only that the administrative officer or body proffer an adequate public justification for its action, on whatever record the relevant procedure demands under the APA.