The Supreme Court’s 1935 landmark decision, Humphrey’s Executor, is on the chopping block.  In Trump v. Slaughter, scheduled for argument next month, the Court will consider whether President Trump lawfully removed Rebecca Slaughter from the Federal Trade Commission.  It has directed the parties to brief and argue “whether the statutory removal protections for members of the Federal Trade Commission violate the separation of powers and, if so, whether Humphrey’s Executor v. United States should be overruled.”  Already, the Court seems poised to answer “yes” to both questions, allowing the removal.  But it could reach the same practical outcome by answering “no” to the first and “yes” only to the second.  And there are compelling reasons why it should do so.

Revisiting Humphrey’s

Humphrey’s holds that while the President must be able to remove “executive” officers at will, statutory for-cause removal protections for “quasi-judicial” and “quasi-legislative” offices like the FTC are constitutional.  The Court has cast doubt on these “quasi” categories, tending to see all administrative powers and offices as “executive,” and at least two current justices have accordingly called for the Court to overrule Humphrey’s entirely. 

While the controversy has swirled around Humphrey’s’ constitutional holding, it is rarely observed that this holding rested on, and was made necessary by, a prior statutory holding.  The first question that confronted the Humphrey’s Court was whether the FTC Act of 1913 in fact restricted the President’s power to remove.  The Act provided (and still provides) that commissioners “may be removed by the President for inefficiency, neglect of duty, or malfeasance in office.”  In both 1913 and 1935 it was not obvious that this formulation restricted the President’s power to remove.  In the 1903 case of Shurtleff v. United States, the Court had held that a virtually identical provision did not do so. 

Ferdinand Shurtleff was a member of the Board of General Appraisers (BGA), an administrative tribunal tasked with deciding appeals from tariff decisions of customs officials.  The BGA’s organic statute provided that its members “may be removed from office at any time by the President for inefficiency, neglect of duty, or malfeasance in office.”  When President William McKinley removed Shurtleff without giving cause, Shurtleff sued for backpay, arguing that he had been unlawfully removed.  The Court rejected his argument, holding that the statute did not in fact limit presidential removal to the three specified causes.  Such limitation would have required “clear and explicit” language specifying that the three causes were the only permissible grounds for removal.  The Court acknowledged that the expressio unius canon of interpretation would normally require statutory language like this to be taken to exclude removal for any other cause or for no cause, but it declined to apply the canon because of the problem it would have created for the President’s “constitutional responsibility to see that the laws are faithfully executed.”

Shurtleff was a large part of the reason why leading Progressive-Era jurists like Ernst Freund, Bruce Wyman, and Frank Goodnow all maintained (and, contrary to most historical accounts, celebrated) that the President enjoyed full power of removal even though for-cause removal protections were already on the statute books.  (I explore their rationale further in an article currently in progress.)  It did not matter that the Court had never issued a definitive constitutional holding on the question.  When it finally did so, in Myers v. United States (1926), both Chief Justice Taft’s opinion for the majority (in dicta) and Justice Brandeis’ dissent assumed that under Shurtleff the President could freely remove members of the FTC and other agencies whose organic statutes had similarly worded removal protections.  Shurtleff was also one of the reasons (alongside Myers) why President Roosevelt thought it was lawful to remove Humphrey from the FTC.  Indeed, Roosevelt had been advised to draft his termination order to Humphrey using the same language President McKinley had used to fire Shurtleff, and the two orders are quite similar.

In Humphrey’s Executor, however, Justice Sutherland began his opinion for the Court by rejecting Shurtleff’s application to the FTC Act.  The Shurtleff opinion had stressed that the consequence of applying expressio unius would have been that members of the BGA would have held office for life during good behavior because its organic statute did not establish a definite term of office for its members.  Sutherland reasoned that since the office of FTC Commissioner was for a fixed term of years, Shurtleff did not govern, expressio unius applied, and the statute permitted only for-cause removal.  Therefore, the constitutional question the Court had avoided in Shurtleff could not be avoided, and the opinion proceeded to give the holding for which it is famous (or infamous).

If the Roberts Court is thinking about overruling Humphrey’s Executor in Slaughter, it should consider overruling this statutory holding rather than its constitutional holding.  Doing so would mean, first, recognizing that there are two categories of for-cause removal protections: those that are permissive (like those in Shurtleff and Humphrey’s), and those that are exclusive (for an example see the footnote).  Second, it would mean reviving Shurtleff’s clear statement rule for the former category and holding, as Taft and Brandeis assumed, that the FTC Act therefore does not restrict the President’s power to remove Commissioners at will.  So interpreted, the FTC Act’s removal protection would not raise a constitutional issue, and the Court could therefore avoid the issue and uphold the Act.  Indeed, the traditional canon of constitutional avoidance obliges the Court to prefer such interpretations if reasonably available.

Objections

There are at least three objections to this approach.  The first has to do with stare decisis.  Courts and scholars often maintain that statutory precedents merit stronger stare decisis than constitutional precedents.  While this distinction is often honored more in the breach and scholars have criticized it, it seeks to recognize that statutory holdings can be corrected more easily (by Congress) than constitutional holdings (by amendment).  On this logic, if Congress disapproves a statutory precedent, it will correct it, and if it does not do so it should be taken as “acquiescing” to it.  So, since Congress has not amended the FTC Act to clarify that it means what Taft and Brandeis thought it meant (that for-cause removal is optional), Congress should be taken to have acquiesced to Humphrey’s’ interpretation. 

This fiction about congressional acquiescence has been widely criticized (Justice Scalia called it a “canard”), but it would be especially perverse to infer acquiescence to Humphrey’s’ statutory holding when Humphrey’s itself implicitly rejected such an inference in reaching that very holding.  In Humphrey’s, the Court declined to apply Shurtleff’s clear statement rule to the FTC Act even though the Act had been enacted ten years after Shurtleff, when Congress, according to the fiction, was aware of it.  Indeed, Congress had shown such awareness in 1908, when it amended the BGA’s organic statute to include the clear statement required by Shurtleff.  If the Court in Humphrey’s did not consider that Congress had acquiesced to Shurtleff by omitting the clear statement from the FTC Act, why should it today infer that Congress has acquiesced to Humphrey’s?  This is not to say the Court is justified in overturning any statutory holding, but where doing so enables the Court to avoid a constitutional question it furthers one of the policies of statutory stare decisis (deference to Congress) more than not doing so.

Second, hasn’t the Court already rejected this approach in its past removal cases?  In Seila Law (2020), the Court confronted a for-cause removal restriction of the Shurtleff variety: that is, a permissive one.  The statute provided: “The President may remove the [Consumer Financial Protection Board] Director for inefficiency, neglect of duty, or malfeasance in office.”  The Court rejected the argument that this could be read broadly enough to allow the President sufficient removal discretion to avoid the constitutional issue.  In support, it cited Humphrey’s’ statutory holding and then concluded that the parties had not offered a “workable” alternative.  The alternatives that were offered were too variable and not rooted “in the statutory text.”

As an initial matter, if the Court is now considering whether to overrule Humphrey’s, it does not matter what statutory interpretations Humphrey’s may have rejected.  Moreover, it seems that the Court in Seila Law did not in fact consider the Shurtleff clear-statement rule.  It refers only to the possibility of avoiding the constitutional issue by “broadly construing the statutory grounds” for removal (“inefficiency, neglect of duty, or malfeasance”).  To be sure, this remains a viable way of resolving these cases through statutory interpretation.  But it is distinct from Shurtleff’s clear statement rule, which depends not on the meaning of the causes but on the structure of the whole clause (permissive or exclusive).  This fact also makes it, incidentally, a “workable” alternative, because it is a rule-like binary classification.

Third and finally, wouldn’t the approach I’ve proposed leave the broader issue of for-cause removal restrictions unsettled precisely when the Court wants to settle it?  Since it applies only to permissive for-cause removal restrictions (like that in the FTC Act) but not exclusive ones (like the one protecting members of the National Labor Relations Board[1]), the Shurtleff rule would leave the status of the latter agencies unresolved.  Indeed, President Trump has also removed members of some of these agencies, and they too have sued.  By avoiding the constitutional issue in Slaughter, in other words, the Court would just be kicking the can down the road, and not very far.

It is true, the Court would not gain much by securing full removal power through statutory interpretation and constitutional avoidance in one set of cases if it would promptly turn around and secure the same result in the other cases by striking down the statute.  In such circumstances, the normal “passive virtue” of avoidance would be de minimis.  The approach I have suggested would be of great utility to any judge who is concerned to protect the President’s constitutional duty of faithful execution in the context of independent agencies but who is not convinced by the constitutional case against for-cause removal restrictions per se.  Such a judge could find that the President can remove members of a large number of supposedly independent agencies while ultimately conceding Congress’ constitutional authority to limit removal, up to a point, through clear language.  It seems that most of the justices on today’s Court do not meet that description.  But if any of them harbor doubts, they should know they are not without alternative tools.

Jordan Rudinsky is an Assistant Professor of Law at the Universidad Católica de Chile.

---

[1] 29 U.S.C. 153(a) (“Any member of the Board may be removed by the President, upon notice and hearing, for neglect of duty or malfeasance in office, but for no other cause”).  By my count, of the forty-eight statutes in Appendix A to Justice Breyer’s dissenting opinion in Free Enterprise Fund, twenty-five have such “exclusive” language.  Of the rest, twenty-two have “permissive” language.  (One is ambiguous.)