Removal as a political question.

• Author: Huq, Aziz Z.
• Position: Introduction to II. Removal Authority as a Means of Bureaucratic Control B. The Removal Power in Institutional Context, p. 1-38

INTRODUCTION I. DOCTRINAL FOUNDATIONS A. Removal Before Free Enterprise Fund B. Free Enterprise Fund C. The Political Question Doctrine II. REMOVAL AUTHORITY AS A MEANS OF BUREAUCRATIC CONTROL A. The Plural Technologies of Bureaucratic Control B. The Removal Power in Institutional Context 1. Variable marginal effects 2. Interaction effects C. Theorizing Removal's Limits 1. Information asymmetries 2. Transaction costs D. Empirical Evidence of the Removal Power's Limits 1. Historical and contemporary U.S. evidence 2. Comparative evidence 3. Private contracting III. THE WEAK LINK BETWEEN PRESIDENTIAL CONTROL AND DEMOCRATIC ACCOUNTABILITY A. Interaction Effects of Presidential Control B. Presidential Control as a Democratic Accountability Mechanism C. Unpacking Democratic "Accountability" IV. THE REMOVAL POWER WITHOUT COURTS CONCLUSION INTRODUCTION

Until now, federal courts have played only a small role in elementary design decisions about the regulatory state. Instead, the political branches select policies, while agencies created by the political branches interpret and enforce those policies on the ground. (1) Courts, to be sure, play a supporting part policing the use of delegated authority, (2) but their influence on the administrative state's basic architecture has to date been minimal. (3)

Suddenly, the status quo is in doubt. A recent Supreme Court decision portends a larger judicial role in drawing up blueprints for federal agencies. The holding of Free Enterprise Fund v. Public Company Accounting Oversight Board (4) is modest. But it rests on an underlying principle with wider potential implications. In Free Enterprise Fund, the Court invalidated a single provision in the Sarbanes-Oxley Act of 2002. (5) Among other reforms, the Act created the Public Company Accounting Oversight Board (PCAOB) to protect investors by supervising the audits of public companies. (6) The challenged provisions seriously restricted the President's authority to remove PCAOB members. (7) In the Court's view, the Act permitted removal of PCAOB members only by the Securities and Exchange Commission (SEC) and then only on a showing of good cause; SEC commissioners also could be removed only on a showing of good cause by the White House. (8) This "dual for-cause" regime created a buffer between the PCAOB and the President that, the Court held, conflicted with the promise of democratic accountability immanent in Article II of the Constitution. (9) This specific holding rested on a more general syllogism. First, the Court held that power to remove a bureaucrat was essential to establish control over that official's policy decisions. (10) Second, the Court reasoned that absent presidential control, the democratic accountability demanded by Article II would be wanting. (11) Based on these two premises, the Court concludes that Article II entails a quantum of presidential removal authority respecting agency officials in order to preserve democratic accountability.

If this principle could easily be cabined to the "dual for-cause" regime at issue in Free Enterprise Fund, it would warrant only passing attention. (12) But big things often have small beginnings. The Free Enterprise Fund principle cannot easily be limited to "dual for-cause" regimes. Rather, the decision's fundamental logic "calls into question the constitutionality of hundreds of other governmental positions" buffered from presidential control, (13) even those positions protected by only a single layer of for-cause removal limitations. The opinion thereby invites judges to hunt through the U.S. Code, striking out tenure protection rules. (14) Hence, the limited scope of short-run consequences from Free Enterprise Fund for the PCAOB itself belies a more important long-term ramification: dramatic enlargement of judicial authority to dictate elementary parameters of agency design.

This is no idle hypothesis. In July 2011, Judge Brett Kavanaugh of the District of Columbia Circuit Court of Appeals published a striking concurrence arguing forcefully that Free Enterprise Fund impugned the constitutionality not only of dual for-cause rules, but also of the Nuclear Regulatory Commission and all other "independent agenc[ies] that operate[] free of presidential direction and supervision." (15) The latter are typically insulated from the White House's influence by single for-cause rules. (16) They are "specifically designed not to have the quality ... of being subject to the exercise of political oversight." (17) A glitch in the Article III case-or-controversy requirements in the case in question precluded Judge Kavanaugh from pressing his argument. (18) But it is only a matter of time before an appropriate lawsuit raises the question of how far Free Enterprise Fund goes in shifting agency design authority to federal courts. (19)

In this Article, I focus on the broad reading of Free Enterprise Fund espoused by Judge Kavanaugh in order to investigate how authority to design agencies should be apportioned between the political branches and the judiciary. Assuming that Article II of the Constitution requires democratic accountability over agency actions, I ask, which branch bears responsibility for executing that mandate at the agency design stage through the regulation of removal authority? To investigate the viability of judicial supervision of agency design, the Article draws on the doctrinal framework of the "political question" doctrine. This is a tool for sorting constitutional disputes between the judiciary and the political branches. (20) To that end, the Court has employed a range of tests to sort constitutional questions based on their amenability for judicial resolution. Most relevant here is the Court's development of the political question doctrine, which, among its many applications, identifies the absence of a judicially manageable standard as a reason for treating a legal issue as nonjusticiable. (21) The Supreme Court has not been as clear as might be desirable about the necessary prerequisites of a judicially manageable standard. For the purposes of this Article, I proffer a relatively minimal and parsimonious test for discerning adequate rules of decision: will judicial enforcement of a rule promote the underlying constitutional values or goods that justify the rule in the first instance? If there is no reliable and stable correlation between a rule of decision and those underlying values, and if the results of a rule's application are instead ad hoc and unpredictable, then the Court has failed to identify a judicially manageable standard. In the absence of a plausible alternative doctrinal framework, courts should refrain from acting, with the resolution of a constitutional issue typically redounding to the political branches.

The central claim of this Article is that the rule of decision articulated in Free Enterprise Fund concerning presidential removal authority fails this test. It is not a judicially manageable standard because it does not reliably produce the constitutional good identified by the Court--democratic accountability. Drawing on a range of empirical and theoretical work in political science and institutional psychology, I aim to demonstrate why judicial enforcement of presidential removal authority will not reliably promote presidential control or democratic accountability. To that end, the Article decomposes the logic of Free

Enterprise Fund into its two interlocking causal theses--the link between presidential removal authority and control, and the link between presidential control and democratic accountability--and analyzes them separately. For the Court's proposed rule to be judicially manageable, it must be the case that both premises of the syllogism employed in Free Enterprise Fund hold true.

Neither of the two causal connections necessary to link presidential removal authority with the constitutional good of democratic accountability, however, withstands close scrutiny. Roughly speaking, both fail because both ignore interactions with other design options and the strategic responses of other government actors. With respect to the first link in the Free Enterprise Fund causal claim, I argue that there is no strong correlation between removal authority and political control. (22) Empirical evidence and political science models instead show that the power to remove is sometimes unnecessary and sometimes ineffectual to the goal of political control of the bureaucracy. Worse, presidential removal authority often has perverse and undesirable effects quite apart from democratic accountability goals. As a result, Presidents have tended not to rely too heavily upon the removal power to secure control over bureaucratic subordinates, and have instead looked to other tools. Turning to the second component of the Court's causal argument, there is no stable positive correlation between presidential control and democratic accountability. Instead, judicial promotion of presidential control will sometimes have the paradoxical consequence of diminishing net democratic accountability.

Taken together, these critiques undermine the putative correlation between presidential removal authority and democratic accountability. Judicial interventions in favor of presidential removal authority can therefore either promote or retard, or even leave untouched, net democratic accountability. Courts simply have no way of knowing in advance what effect their intervention will have on the relevant constitutional good. In the argot of the political question doctrine, this means that the promotion of presidential removal authority is not a judicially manageable standard by which a constitutional value can be reliably achieved.

This Article concerns justiciability and takes no position on the underlying questions whether Presidents should have broad removal authority for any given agency, or whether there should be exceptions to allow, say, adjudicative autonomy or...