As one part of a statute officially entitled the “Dodd-Frank Wall Street Reform and Consumer Protection Act,” Congress in 2009 established the Consumer Financial Protection Bureau (CFPB) as “an Executive agency” headed by a Director removable from office by the President “for inefficiency, neglect of duty, or malfeasance in office.” Rumor now has it that the incoming President, Donald Trump, might remove the incumbent CFPB Director, Richard Cordray, shortly after January 20, 2017. That possibility raises the following question: Precisely what legal impediments does the Dodd-Frank Act place on the ability of a new President to remove an incumbent CFPB Director?

That question was addressed five years ago by then-Representative Barney Frank, whose views may be relevant since he is the “Frank” in Dodd-Frank:

[T]his notion that the Director cannot be removed is fanciful. It says in the statute that, yes, the Director is appointed for a 5-year term, but can be removed by the President for insufficiency [sic], neglect of duty, or malfeasance.

No one doubts that if a change in Administration comes, and the new President disagrees with the existing Director, he or she can be removed. And proving that you were not inefficient, the burden of proof being on you, would be overwhelming.

Budget Hearing—Consumer Financial Protection Bureau Before the Subcommittee on Oversight and Investigations of the Committee on Financial Services, 112th Cong. 8 (Feb. 15, 2012) (Serial No. 112-101).

Legislative commentary of this kind is sometimes used by courts to understand the meaning of statutory language, though Representative Frank’s comments in 2012 may have reduced persuasive weight given that they were made three years after the enactment of the statute, during a hearing, by only a single member of Congress. See, e.g., United States v. Price, 361 U.S. 304, 313 (1960) (describing any post-enactment legislative history from “a subsequent Congress” as “a hazardous basis for inferring the intent of an earlier [Congress]”). Yet even if not considered persuasive legislative commentary, Representative Frank’s 2012 remarks—coming from a prominent member of the legal community who is highly informed about the statute that bears his name—provide some evidence of the public understanding of restrictions on the President’s authority to remove officers within the executive branch.

Ever since the Supreme Court’s decisions in Myers v. United States, 272 U.S. 52 (1926), and Humphrey’s Executor v. United States, 295 U.S. 602 (1935), there has been an outpouring of scholarship addressing the constitutionality of removal restrictions. Yet comparatively little academic (or for that matter, judicial) attention has been given to the meaning of those provisions, notwithstanding Congress’s use of language similar to the CFPB statute to restrict the removal of officers within a number of the so-called “independent agencies.” The comments of Representative Frank are a good starting point for an inquiry into statutory meaning (assuming the statute is constitutional), and thus we here ask the question: Was Representative Frank right that after a change in Administration, a new President has a fairly capacious power to remove a CFPB Director? Answering that question raises four significant legal issues.

The first issue is whether the statutory list in the Dodd-Frank Act—which authorizes removal of the Director for three reasons, “inefficiency, neglect of duty, or malfeasance in office”—establishes the exclusive bases to remove the CFPB Director. Supreme Court precedents point in different directions on this question.

In Shurtleff v. United States, 189 U.S. 311 (1903), the Supreme Court confronted statutory provisions quite similar to those authorizing the appointment and removal of the CFPB Director. The officer there, a customs official, was appointed by the President, with the advice and consent of the Senate, and was removable under the statute for “inefficiency, neglect of duty, or malfeasance in office.” The President removed the officer without making any finding required under the statutory removal power, and the removed officer challenged the legality of his removal in a suit for back pay. The Court sustained the President’s removal, holding that the statutory removal power was not exclusive and rejecting as “a mistaken view . . . that the mere specification in the statute of some causes for removal thereby excluded the right of the President to remove for any other reason which he, acting with a due sense of his official responsibility, should think sufficient.” Id. at 317. Constricting the President’s nonstatutory power to remove officers appointed by him would require, the Court held, “very clear and explicit language” and could not be “taken away by mere inference or implication.” Id. at 315.

Three decades later, in Humphrey’s Executor v. United States, 295 U.S. 602 (1935), the Court held that the President’s authority to remove the officer there—a Commissioner of the Federal Trade Commission—was limited to the three listed bases, “inefficiency, neglect of duty, or malfeasance in office.” In its brief arguing to the contrary, the Roosevelt Administration had placed principal weight on the statutory holding of Shurtleff, contending that “nothing in the language or the legislative history of the Federal Trade Commission Act [ ] indicate[s] that Congress intended to depart from the meaning established in the Shurtleff case.” But Humphrey’s Executor distinguished Shurtleff on two grounds: (1) that the FTC’s statute specified a seven-year term for Commissioners, and (2) that the FTC was “neither political nor executive, but predominantly quasi-judicial and quasi-legislative.” 295 U.S. at 624.

The CFPB statute falls somewhere between the statutes in Shurtleff and Humphrey’s Executor. On the one hand (as in Humphrey’s Executor), the statute specifies a specific term of years for the Director’s tenure. On the other hand (unlike Humphrey’s Executor), the statute explicitly creates the CFPB as “an Executive agency.” Other statutes, moreover, explicitly restrict the presidential removal power by limiting the grounds for removal “only” to the three enumerated bases. See, e.g., 5 U.S.C. § 1211 (authorizing presidential removal of a Special Counsel “only” for inefficiency, neglect of duty, or malfeasance in office). Such statutes suggest that Congress knows how to limit the President’s power to remove executive officers when it so desires.

The conventional wisdom is that Humphrey’s Executor fatally undercut Shurtleff’s statutory holding. But the two cases, taken together, illustrate how subtle differences in the statutory framework can have large implications for the President’s authority to remove subordinates. See, e.g., Bowsher v. Synar, 478 U.S. 714, 729 (1986) (describing as merely “arguable” the contention that “the enumeration of certain specified causes of removal excludes the possibility of removal for other causes”); compare Free Enterprise Fund v. Public Company Accounting Oversight Board, 561 U.S. 477, 503 n.7 (2010) (describing as “implausibl[e]” the government’s argument that the removal statute at issue in the litigation “‘does not expressly make its three specified grounds of removal exclusive,’ and that ‘the Act could be construed to permit other grounds’”), with id. at 529 (Breyer, J., dissenting) (relying on Shurtleff and alluding to the possibility that “the removal authority is not always restricted to a removal for the causes set forth by statute”) (internal quotation marks and alteration omitted). Despite the doubt on this point, we will assume for the remainder of this post that Congress intended to permit the President to remove the CFPB Director only on the grounds specified in the statute.

The second major issue involved in any potential removal of a CFPB Director is whether, in the course of exercising his statutory removal authority, the President must give notice and a hearing to the CFPB Director prior to making any finding of “inefficiency, neglect of duty, or malfeasance in office.” The Dodd-Frank Act does not specify any such requirement. Other statutes do. For example, the Judges of the Tax Court “may be removed by the President, after notice and opportunity for public hearing, for inefficiency, neglect of duty, or malfeasance in office but for no other cause.” 26 U.S.C. § 7443(f).

Shurtleff, however, reasoned in dicta that if an officer is removed “for inefficiency, neglect of duty, or malfeasance in office, . . . the officer is entitled to notice and a hearing.” 189 U. S. at 314. This issue is significant because, if the President were to remove the CFPB Director without such notice and a hearing, it is conceivable that a court would later declare the removal unlawful even though the President can meet the statutory grounds for removal. But there is a question whether this aspect of Shurtleff applies to a statute that, unlike the statute in Shurtleff, does constrict presidential power to the three listed grounds for removal. It is possible that Humphrey’s Executor, in undercutting Shurtleff’s holding that the enumerated grounds were not exclusive, also undercut the case’s dicta that a removal based on the enumerated grounds requires notice and a hearing. Restricting presidential removal to enumerated statutory grounds and requiring notice and a hearing further limits presidential power, and courts might balk at inferring such an additional limit without express statutory language.

If a hearing is required, its precise scope and contours would have to be determined. It seems implausible that such a hearing would need to be the type of in-person boardroom meeting made famous by The Apprentice, much as such a dramatic meeting between President Trump and Director Cordray (a former Jeopardy! champion) might make for great television. A “hearing” in the parlance of administrative law need not be an oral, in-person hearing. It seems likely that the President could satisfy any hearing requirement by sending the Director a proposed list of reasons for removal, allowing the Director a set time to respond, and then determining whether the statutory causes for removal exist.

That brings us to the third and perhaps most important issue: What do the terms inefficiency, neglect of duty, and malfeasance in office mean? Representative Frank states that, where “the new President disagrees with the existing Director,” the President could remove the Director. There is not, however, clear legal precedent to support or refute that assertion. In Humphrey’s Executor, President Roosevelt had previously written letters to Commissioner Humphrey expressing the views that the President wanted “personnel of my own selection” on the FTC and that “[Humphrey’s] mind and [his] mind [did not] go along together on either the policies or the administering of the Federal Trade Commission.” 295 U.S. at 618-19. In the ultimate order of removal, however, the President did not purport to satisfy any of the statutory grounds for removal, id. at 619, and thus the meaning of the statutory standards was not an issue presented to the Court.

In Free Enterprise Fund v. Public Company Accounting Oversight Board, 561 U.S. 477 (2010), the Court did comment on the strength of the statutory tenure protections, stating that, if officers in a multi-member commission are protected against removal under a statutory standard similar to that in Humphrey’s Executor, then the President could not remove the officers merely where “the President disagrees with their determination” in a particular matter. Id. at 496; see also PHH Corp. v. CFPB, 839 F.3d 1, 15 n.2 (D.C. Cir. 2016).

On the other hand, the Court in Bowsher v. Synar interpreted a similar protection enjoyed by the Comptroller General—who can be removed by a joint resolution of Congress for inefficiency, neglect of duty, or malfeasance in office—as “very broad” and as permitting removal “for any number of actual or perceived transgressions of the legislative will.” 478 U.S. at 734. Furthermore, dicta in Free Enterprise Fund suggest that the President might be able to remove tenure-protected officials where the President disagrees with a single determination of the officials if “that determination is so unreasonable as to constitute ‘inefficiency, neglect of duty, or malfeasance in office.’” 561 U.S. at 479 (emphasis added). That caveat itself suggests that, even where tenure-protected officials disagree with the President on a single determination, they are protected against removal only within a zone of reasonableness. Free Enterprise Fund does not address a thoroughgoing disagreement over a pattern of administration, but just as margins of error narrow with more data, so too the zone of reasonableness might narrow where a tenure-protected official makes repeated determinations that, in the President’s view, are borderline unreasonable.

In light of the significant uncertainty in Supreme Court precedents, one might try to understand the statutory standards by referring to Executive Branch constructions of the statutory terms or to background principles of administrative law. For example, President Clinton understood the “[p]hrase ‘neglect of duty’ to include, among other things, a failure to comply with the lawful directives or policies of the President.” Statement on Signing the Floyd D. Spence National Defense Authorization Act for Fiscal Year 2001 (Oct. 30, 2000). In making that assertion, President Clinton emphasized that the particular officer there (the Under Secretary for Nuclear Security at the Department of Energy) had “sensitive duties . . . in the area of national security.” Id. That reasoning, however, raises yet another point, which is whether the statutory bases for removal should be given a uniform interpretation across all statutes using similar or identical language or a context-specific interpretation dependent on the structure and duties of the office. If the latter is the correct approach, then the CFPB Director might be afforded weaker tenure protection than FTC Commissioners because the CFPB is expressly denominated by statute as an “Executive agency” whereas the FTC’s statute was interpreted in Humphrey’s Executor as attempting to create a “nonpartisan” commission with duties “neither political nor executive.” 295 U.S. at 624.

Or to take another example, Representative Frank may well be right that “inefficiency” is a particularly easy standard to satisfy, in part because the concept of “efficiency” in administrative law has become nearly synonymous with economic efficiency in general, and cost-benefit analysis in particular. See, e.g., Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208, 218-20 (2009) (holding that a regulatory statute permitted the EPA to use cost-benefit analysis because the statute could be interpreted to impose an efficiency-based standard). Thus, if the President views a Director as managing the agency inefficiently—by, for example, expending agency resources in ways not justified by their public benefits—then the President could remove the officer under the statutory grounds of “inefficiency.” Support for this reading comes from Professor Cass Sunstein, the White House’s chief regulatory official during President Obama’s first term, who has posited that the President’s power to remove tenure-protected officers might be “very broad” and that “[i]f agencies proceed [in regulating] when the benefits do not exceed the costs, they might reasonably be thought to be acting ‘inefficiently.’” Robert W. Hahn & Cass R. Sunstein, A New Executive Order for Improving Federal Regulation? Deeper and Wider Cost-Benefit Analysis, 150 U. Pa. L. Rev. 1489, 1535-36 (2002).

The fourth and last issue raised by any removal is the standard of review that a court must apply to a presidential determination that “inefficiency,” “malfeasance,” or “neglect of duty” has occurred in a particular case. Representative Frank’s perspective in 2012 was that the “burden of proof” would be on the removed officer and that the burden would likely be “overwhelming.”

The case law has not yet established any precise verbal formula to govern the standard for such review of a presidential finding, but Representative Frank’s views do not appear to be far off the mark if we consider background principles of administrative law. Because the President is not considered an “agency” for purposes of the Administrative Procedure Act (APA)—the otherwise generally applicable statute for obtaining judicial review of agency decisions—a presidential determination on a factual issue is not reviewable under the arbitrary-and-capricious, abuse-of-discretion, or other standards of the APA. See Franklin v. Massachusetts, 505 U.S. 788, 800-01 (1992).

Moreover, in any non-APA lawsuit seeking review of the removal (presumably a suit for back pay or possibly a suit for injunctive relief), some or all of any presidential factual findings concerning (for example) the removed officer’s “inefficiency” might be unreviewable in court. In United States v. George S. Bush & Co., 310 U.S. 371 (1940), the Court held that, where Congress gives the President a statutory power contingent upon the President making particular factual findings, the “judgment of the President on those facts” is “determinative” and “not subject to review.” Id. at 379-80. While George S. Bush involved presidential factual determinations related to setting tariffs, the Court relied on precedent applying the same principle in other contexts, including situations where the executive determination concerned quite particularized facts—what administrative law scholars call “adjudicative” facts. See Monongahela Bridge Co. v. United States, 216 U.S. 177, 193-95 (1910) (holding unreviewable an executive factual determination that a company’s bridge was an obstruction to navigation).

Judicial respect for the President’s factual judgments concerning removal also seems consistent with the meaning of the statutory provisions that first introduced into federal law the standard of “inefficiency, neglect of duty, or malfeasance in office” as a restriction on the President’s removal authority. That standard is typically traced back to the statute creating the Interstate Commerce Commission (ICC). See Act to Regulate Commerce, ch. 104, § 11, 24 Stat. 379, 383 (1887). In a 1941 survey of administrative agencies, Professor Robert Cushman claimed that, while “[t]here was no discussion of the President’s power to remove members of the commission or of the restriction in the statute upon that power of removal” when the ICC was created in 1887, some legislative statements show that Congress anticipated a presidential role in overseeing the Commission. Robert A. Cushman, The Independent Regulatory Commissions 60-61 (1941). In particular, Cushman relied on statements made by Senator Wilkinson Call of Florida, who described the ICC as “amenable to the President, the President amenable to the people for their [i.e., the ICC Commissioners’] official and personal responsibility in the administration of this great trust, the greatest and most important possessed by his [i.e., the President’s] office in time of peace.” 18 Cong. Rec. 570 (Jan. 12, 1887). Such statements reflect appreciation that removal was an executive function policed by political forces. Indeed, in the same year that Congress created the Interstate Commerce Commission, it allowed the Tenure of Office Act to lapse. The latter Act (in)famously required the Senate’s consent before an officer could be removed, thereby creating much greater “independence” from the President. It could have been (but was not) used as a template for the statute creating the ICC.

In addition, at least two of the terms that Congress used in the ICC statute—“neglect of duty” and “malfeasance of office”—had been used for decades in a manner suggesting that they were understood to establish quintessentially executive factual judgments. For example, after dismissing the entire cabinet of his predecessor, Zachary Taylor, President Millard Fillmore remarked in his first annual message to Congress that, while he intended to exercise his power to appoint officers carefully, “mistakes will sometimes unavoidably happen and unfortunate appointments be made notwithstanding the greatest care. In such cases the power of removal may be properly exercised; and neglect of duty or malfeasance in office will be no more tolerated in individuals appointed by myself than in those appointed by others.” Millard Fillmore, First Annual Message (Dec. 2, 1850). Fillmore’s explanation of his removal policies to Congress suggests that, even before the concepts of “neglect of duty” and “malfeasance in office” were codified in federal statutory law, they were understood to be a part of executive decisionmaking. State court precedents decided roughly contemporaneously with the passage of the ICC statute likewise recognized that neglect of duty and malfeasance in office were matters for executive determination, with the holder of the removal power being “the sole judge of the sufficiency of the evidence to justify the removal” and the courts “entirely powerless” to review the executive determination of facts. State ex rel. Lamar v. Johnson, 30 Fla. 433, 477 (1892) (collecting cases from other states on the issue).

Yet even if judicial review of the President’s factual determinations is permitted, the standard of review would presumably at least as deferential, and perhaps more deferential, as the standards articulated in the APA for review of agency fact finding. See Universal Camera Corp. v. NLRB, 340 U.S. 474, 490-491 (1951) (suggesting that non-APA standards for reviewing agency findings of fact might be more deferential than APA standards).

In sum, Representative Frank might very well be right that the President possesses a fairly broad power to remove a CFPB Director and that, in any event, a removed Director will likely confront an overwhelming burden in seeking to have the courts overturn the President’s decision. Executive removal of a statutorily tenured officer presents a theoretically interesting separation-of-powers issue precisely because it requires uncovering the balance among the three branches of government struck by constitutional provision and congressional enactment, with each branch having both significant power, and limits on that power.

Aditya Bamzai is an Associate Professor of Law at the University of Virginia School of Law. John F. Duffy is the Samuel H. McCoy II Professor of Law and Elizabeth D. and Richard A. Merrill Professor of Law at the University of Virginia School of Law.