Can the Special Counsel Regulations Be Unilaterally Revoked?

George Conway, writing in Lawfare a few weeks ago, forcefully rejected professor Steven Calabresi’s argument that the special counsel’s appointment was unconstitutional. I agree with his analysis as a general matter, though one of Conway’s particular conclusions strikes me as hasty: that “the special counsel regulations can be unilaterally revoked by the very executive branch that unilaterally created them.”

If the regulations can be rescinded with the stroke of the pen—and everyone from Robert Mueller to the D.C. Circuit agreed—then there is absolutely no problem with respect to the separation of powers. If, however, the regulations cannot be repealed unilaterally—for example, if the rescission may be subject to judicial review pursuant to the Administrative Procedure Act (APA)—then under Justice Antonin Scalia’s Morrison v. Olson dissent, there has been some “removal of Presidential power.” Why? In effect, through the APA, Congress would be allowing one administration to bind its successors. Under a proper application of administrative law, the Trump administration should be able to rescind the Clinton-era regulations—which did not go through the notice-and-comment process in 1999—but, as illustrated by the ongoing Deferred Action for to Childhood Arrivals (DACA) litigation, presidential reversals are not always so simple. This post will consider whether the special counsel regulations can be unilaterally revoked and, if not, whether the rescission could be subject to judicial review. At bottom, Conway is probably correct, but there is enough doubt on the point that courts could sufficiently impede the president’s recission power to raise precisely the sort of separation-of-powers problem his piece argues does not exist.

***

In May 2017, Deputy Attorney General Rod Rosenstein appointed Robert S. Mueller “to serve as a Special Counsel for the United States Department of Justice.” The order specified that “Section 600.4 through 600.10 of Title 28 of the Code of Federal Regulations are applicable to the Special Counsel.” Perhaps the most important provision is § 600.7. Promulgated shortly after the independent counsel statute expired at the end of June 1999, this regulation imposes an important constraint on the executive branch: “The Attorney General may remove a Special Counsel for misconduct, dereliction of duty, incapacity, conflict of interest, or for other good cause, including violation of Departmental policies.”

Former acting solicitor general Neal Katyal, who was involved in drafting Section 600 nearly two decades ago, explained that there were only two paths by which Mueller could be fired. First, President Trump “would have to direct Rosenstein to fire Mueller” for some form of misconduct that rises to the level of “good cause.” Second, Katyal notes, “Trump could order the special-counsel regulations repealed and then fire Mueller himself.” (I suspect that Katyal meant Trump could then order Rosenstein to fire Mueller; whether Trump could directly remove someone he did not appoint is a separate question.) Katyal does not specify exactly how the regulations could be repealed.

As a general matter, the APA affords the public an opportunity to opine on a proposed regulation through what is known as the notice-and-comment rulemaking process. Before an agency can finalize the regulation, it must consider the public comments. If the agency does not follow the procedures required by the APA, parties with standing can challenge the regulation in court. However, not all executive actions must go through this process. In 1999, Attorney General Janet Reno promulgated the special counsel regulation. She cited four reasons why the rules were exempt from the APA’s notice-and-comment process. First, “[t]his rule relates to matters of agency management or personnel, and is therefore exempt from the usual requirements of prior notice and comment and a 30-day delay in the effective date.” Second, “this rule would be exempted from the requirements of prior notice and comment as a rule of agency organization, procedure, or practice.” Third, “the effective date of the rule need not be delayed for 30 days after publication because the rule is not a ‘substantive rule.’” The fourth reason is potentially the most important:

In any event, because the provisions of the Independent Counsel Reauthorization Act of 1994 expire on June 30, 1999, the Attorney General has determined that it is imperative to have these rules governing the appointment and service of a Special Counsel in place as soon as possible. Accordingly, even if the rule were not exempt from the usual requirements of prior notice and comment and a 30-day delay in the effective date, there would be “good cause” for issuing this rule without prior notice and comment and without a 30-day delay in the effective date.

The rulemaking was published in the Federal Register on July 9, 1999, more than a week after the Independent Counsel Reauthorization Act had expired. Had the attorney general submitted the regulations for comment, the process could have been held up for months, or potentially longer. During that gap, there would have been no means to appoint a special counsel.

As a general matter, regulations that were promulgated without going through the notice-and-comment process can be rescinded in the same fashion—that is, unilaterally. Professor Mike Rappaport points out that “[i]f Reno was correct, then this suggests that the President could change the regulation rather quickly. Notice and comment would not be required.” But what if Reno was wrong, and the regulations did not satisfy one or more of the reasons she identified? In that case, Rappaport points out, “the regulation would not have been valid for this period, since it should have gone through notice and comment.” As a result, the regulations would be void ab initio and could in no way constrain Mueller’s removal.

Rappaport posits an alternate argument: What if one of Reno’s reasons justified the regulation’s enactment but not rescission? For example, there may have been good cause to skip the 30-day process in 1999, because the Independent Counsel Reauthorization Act had already expired. However, there is arguably no similar urgency today to rescind the regulations. If the district court accepted that this argument argument is correct, and the Trump administration rescinds the special counsel regulations without going through the notice-and-comment rule-making process, then the court could find that unilateral rescission violates the Administrative Procedure Act.

Even if there is an APA violation, would anyone have standing to challenge the rescission? Adam White writes on Lawfare that Congress should “legislat[e] the process by which the courts would hear a lawsuit challenging the firing of a special counsel.” However, he acknowledges that if Mueller is terminated, even in the absence of new legislation, “there would still be litigation to determine . . . whether that [good cause] standard has been met.” In such a case, Mueller himself might have standing if the executive branch attempts to rescind the regulation, thereby removing one of his tenure protections.

Even if this argument is wrong, and the rescission does not have to go through the notice-and-comment rule-making process, Mueller still might have standing to challenge the repeal. Why? The DACA litigation has illustrated that courts will review agency actions to rescind executive actions, even if that initial policy did not go through the rule-making process. For example, the Clinton-era rule-making stated that “[t]he regulations in this part are not intended to, do not, and may not be relied upon to create any rights, substantive or procedural, enforceable at law or equity, by any person or entity, in any matter, civil, criminal, or administrative.” The 2012 DACA memorandum included very similar language, yet the judiciary has still found the rescission to be subject to judicial review.

Under such an analysis, the executive branch would have to demonstrate to the satisfaction of a district court judge that the rescission is not “arbitrary and capricious.” Traditionally, this standard has been quite deferential, but over the past 18 months some courts have applied it with more exacting precision. I pity the lawyer who has to argue in court why President Trump wants to fire Robert Mueller without good cause. It is not difficult to imagine that a district court enjoins the rescission of the special counsel regulations, much like district courts have enjoined the rescission of DACA. To be clear, this analysis would be wrong. Indeed, while the DACA rescission implicates the complicated statutory framework of immigration law, revoking the special counsel regulations exclusively concerns the president’s authority over prosecutorial discretion, an area where his discretion is at its apex. In any event, if a district court judge takes this approach, a race to the Supreme Court commences.

This process, no doubt, will take time. All the while, the president is not able to remove the special counsel, as he would have wanted to, because of Attorney General Reno’s regulations. At the American Constitution Society blog, professor Steve Vladeck sketches out one possible route for litigation:

To be fair, the Special Counsel regulation has some of the hallmarks of legislative rules (which are harder to repeal). But even if the Executive Branch claims that the Special Counsel regulation is a non-legislative rule (and, indeed, even if it is ultimately correct on that score), it is not hard to imagine that there will be litigation over that issue—litigation that could take time before definitively resolving the matter. All the while, presumably, the investigation could continue apace.

Were this process to unfold during some sort of foreign policy exigency, and the president was disabled from removing the special counsel, the separation-of-powers problems would become apparent. This hypothetical illustrates a point I made last month on Lawfare:

Without question, Robert Mueller has far more constraints on his autonomy then did Ken Starr, yet these two investigations share one important hallmark: to quote Justice Scalia’s Morrison dissent, the President is “deprive[d] ... of substantial control over the prosecutory functions.” Under the extant regulations, the special counsel can only be removed for certain reasons and—in certain cases—can publicize his findings. That these restrictions fall far short of the protections afforded by the independent-counsel statute does not mean they are consistent with Justice Scalia’s dissent.

If the courts were to erroneously hold that the APA controls the rescission of the

special counsel regulations, and are subject to an “arbitrary and capricious” standard of review, then the restrictions are not quite self-imposed by the executive. In Morrison v. Olson, Congress directly imposed a constraint on the president’s removal power through the independent counsel statute. Here, however, the constraint comes from the Clinton administration. Had President Clinton simply adopted these regulations by executive order, or instructed Attorney General Reno to publish them in an internal guidance document, the restrictions would indeed be self-imposed. But the Clinton administration did not take this approach. Reno published them in the Federal Register pursuant to the Administrative Procedure Act (a law enacted by Congress, that can be enforced by the judiciary). This approach was designed to achieve some degree of permanence—as Katyal suggested—and perhaps, justiciability. I’ve referred to this process as presidential insulation.

Furthermore, the mere fact that one president tries to bind another does not render it “self-imposed.” Chief Justice John Roberts’s opinion in Free Enterprise Fund v. PCAOB recognized that one administration cannot shackle another:

Perhaps an individual President might find advantages in tying his own hands. But the separation of powers does not depend on the views of individual Presidents, see Freytag v. Commissioner, nor on whether “the encroached-upon branch approves the encroachment,” New York v. United States. The President can always choose to restrain himself in his dealings with subordinates. He cannot, however, choose to bind his successors by diminishing their powers, nor can he escape responsibility for his choices by pretending that they are not his own. (citations omitted).

In his post, Conway explained that the regulations could be “unilaterally revoked”:

And not only that, the special counsel regulations can be unilaterally revoked by the very executive branch that unilaterally created them. Indeed, according to Attorney General Janet Reno when she issued the special counsel regulations in 1999, those regulations aren’t even subject to notice-and-comment rulemaking, because they relate to “matters of agency management or personnel,” and “agency organization, procedure, or practice”—which suggests they could be dispensed with equally unceremoniously as well. In short, not only does the special counsel regulation not take power away from the executive branch, but the internal rearrangement of authority it brings about within that branch could be relatively easily undone by a presidentially removable principal officer. No matter how you cut it, to borrow Justice Scalia’s words, there has been no “removal of Presidential power”—none—let alone too much. In short, there is no serious argument that Special Counsel Mueller’s appointment violates the Appointments Clause specifically or the separation of powers generally. (emphasis added).

There should be, in short, no “serious argument” that the special counsel regulations intrude on the separation of powers. But there is such a credible argument, to the extent that a district court would find that Mueller has standing to challenge a rescission for not going through the notice-and-comment rulemaking process, and to assert that a rescission is “arbitrary and capricious.”