Emory Law Journal

Administrative Law in an Era of Partisan Volatility
Michael A. Livermore,
Daniel Richardson *Michael A. Livermore is a professor at the University of Virginia School of Law. Daniel Richardson, University of Virginia School of Law, J.D., 2018. The views expressed herein are the authors’ alone. Thanks to Rachel Barkow, Jeffrey Lubbers, Nina Mendelson, Dick Pierce, Teddy Rave, Richard L. Revesz, Mark Tushnet, Christopher Walker, and workshop participants at the University of Houston Law Center for helpful feedback on earlier versions of this Article.

Abstract

Contemporary politics is characterized by a polarized national discourse, weak party organizations, volatile control of government, and an increasingly assertive executive. These political dynamics interact with a system of administrative law inherited from different political times that is ill-suited to addressing the risks of the current moment, which include threats to administrative values such as efficiency, impartiality, and expertise; to democratic values such as accountability, transparency, and participation; and, most generally, to the ability of the law to benefit social well-being through sound policymaking. In recent years, some legal scholars have become attuned to the interaction between administration and these features of the political environment. Unfortunately, administrative law itself has been slow to catch up, with potentially dangerous consequences for the stable functioning of the U.S. administrative state. Drawing from the political science literature that examines the interacting features of organized politics that generate stable party systems, we examine how prior administrative law regimes have responded to political transformations in the past. With this research in view, we find that administrative law as it stands today, which emerged during an earlier party system, is ill-suited to meet the challenges of modern politics. Time will tell whether courts will shape doctrine to better align administrative law with the needs of the times.

 

Introduction

Debates about the relationship between politics and administration are as old as modern bureaucracies, extending at least back to foundational figures such as Woodrow Wilson and Max Weber. 1See Max Weber, Bureaucracy, in From Max Weber: Essays in Sociology 196, 232 (H. H. Gerthand & C. Wright Mills eds., 1946); Woodrow Wilson, The Study of Administration, 2 Pol. Sci. Q. 197, 197 (1887). See generally Dwight Waldo, Politics and Administration: On Thinking about a Complex Relationship, in A Centennial History of the American Administrative State 89 (Ralph C. Chandler ed., 1987). Early notions of a strict dichotomy between politics and administration that helped justify the vast discretion granted to administrative agencies in the early to mid-twentieth century have given way to an understanding of the inseparability of administration and politics in any complex legal regime. 2See Paul H. Appleby, Policy and Administration (1949); Frank J. Goodnow, Politics and Administration: A Study in Government (1900); Dwight Waldo, The Administrative State: A Study of the Political Theory of American Public Administration xvii (2d ed. 1984); Gerald E. Caiden, In Search of an Apolitical Science of American Public Administration, in Politics and Administration: Woodrow Wilson and American Public Administration 51 (Jack Rabin & James S. Bowman eds., 1984) (discussing mid-century heyday and eventual challenges to more traditional notions of a politics/administration dichotomy); Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J. L. Econ & Org. 81 (1985) (introducing “Law of Conservation of Administrative Discretion”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667 (1975). In recent years, some legal scholars have become particularly attuned to the interaction between administration and features of the political environment, such as divided versus unified government and political polarization. 3See Josh Chafetz, Congress’s Constitution 28–35 (2018) (surveying literature on congressional oversight during periods of unified government and finding that Congress uses its power more sparingly during those periods); Jessica Bulman-Pozen, Executive Federalism Comes to America, 102 Va. L. Rev. 953 (2016); Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077 (2014) (arguing that national polarized politics interacts with federalism); Neal Devins, Party Polarization and Congressional Committee Consideration of Constitutional Questions, 105 Nw. U. L. Rev. 737, 758 (2011) (discussing “blood-sport” battles brought on by political incentives); Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245 (2001); Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2311 (2006) (arguing that unified control undermines inter-branch competition essential to oversight); Michael A. Livermore, Political Parties and Presidential Oversight, 67 Ala. L. Rev. 45 (2015); Thomas O. McGarity, Administrative Law as Blood Sport: Policy Erosion in a Highly Partisan Age, 61 Duke L.J. 1671, 1711–12 (2012) (examining consequence of polarization); Anne Joseph O’Connell, Shortening Agency and Judicial Vacancies Through Filibuster Reform? An Examination of Confirmation Rates and Delays from 1981 to 2014, 64 Duke L. J. 1645 (2015). Unfortunately, administrative law itself has been slow to catch up, with potentially dangerous consequences for the stable functioning of the U.S. administrative state.

A recent Supreme Court administrative law intervention illustrates this claim. In Lucia v. SEC, the Supreme Court held that the SEC’s administrative law judges (ALJs) are subject to selection through the Appointments Clause rather than the independent civil service system. 4 Lucia v. SEC, 138 S. Ct. 2044, 2049 (2018). Shortly thereafter, the President issued an executive order removing all ALJs from the competitive civil service process and subjecting them to political appointment by agency heads. 5See Exec. Order No. 13,843, 83 Fed. Reg. 32,755 (July 13, 2018) (excepting ALJs from the competitive service). The administrative law doctrine announced in Lucia, which was limited to the SEC, was transmuted almost immediately through political channels to apply broadly across the government. 6 Opposition to the move has been heated, with the President of the American Bar Association referring to the order as “ill-considered and legally vulnerable.” Letter from Hilarie Bass to Congressman Pete Sessions and Congressman James McGovern (July 16, 2018), https://www.americanbar.org/content/dam/aba/uncategorized/GAO/ALJlettertorulescommitteeaboutEO.authcheckdam.pdf; see also Jeffrey S. Lubbers, The RAA Loses Steam But the Trump Executive Order on ALJ Selection Upturned 71 Years of Practice, 94 Chi.-Kent L. Rev. (forthcoming 2019). This move occurs against a political backdrop that includes the growth of presidential administration, an increasingly polarized electorate, the declining institutional efficacy of political parties, and a pattern of wildly shifting governing coalitions. 7See infra Part III. The seemingly technical constitutional holding in Lucia is given life by this political context, where the consequences of upsetting a longstanding compromise that insulates agency adjudicators from partisan influence will be felt.

This interaction of politics and administrative law is nothing new. Professor Jerry Mashaw’s work in recovering the “lost era” of American administrative law documents how the changing landscape of American politics over the course of the Republic’s first one hundred years were tracked by fundamental transformations in administrative law and practice. 8See generally Jerry Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012). When the Jeffersonian Era of Good Feelings gave way to Jacksonian democracy, for example, that shift was accompanied by a change in administration that moved away from an emphasis on governance by “men of good standing” and toward a spoils system that supported mass political mobilization. 9Id. at 175. Legal changes, including a decline in the importance of qui tam actions and the extension of personal immunity for official acts, followed these political and administrative changes. 10 For other foundational work, see Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920 (1982). Similar linked transformations in political organization, administrative form, and legal doctrine have been a consistent feature of the U.S. system ever since.

 

Indeed, the standard administrative law narrative—which continues to be embodied in the leading treatises and casebooks—essentially tracks doctrinal developments against the backdrop of the waxing and waning of the New Deal coalition. 11See Chloe Breyer et al., Administrative Law and Regulatory Policy 14–29 (6th ed. 2006); see, e.g., Richard J. Pierce, Jr., Administrative Law (5th ed. 2017); Gary Lawson, Federal Administrative Law 99–123 (7th ed. 2016). The 1932 election brought a realigned and ascendant Democratic Party to power with a mandate to combat the Great Depression. New Deal legislation created a massive administrative apparatus that, while building on prior forms, was something new in American political life. Administrative law as it is taught and practiced today—from the non-delegation cases to the Administrative Procedure Act and its subsequent interpretation—remains shaped by doctrine that arose from a legal order in constant conversation with the administrative necessities and political realities of that time. 12 Some recent scholarship fails to account for this evolution, treating the current terms of the debate as if they were universal. See, e.g., Philip Hamburger, Is Administrative Law Unlawful? (2014).

If contemporary politics do not (yet) rise to the level of crisis represented by the Great Depression, they nonetheless require a reckoning. Administrative law could adapt: U.S. history provides an ample storehouse of examples where jurists and other decision makers responded to the needs of the day by altering the legal and organizational structures that translate politics into policy. But success is far from a foregone conclusion. The law could fail to change, or change in ways that are counterproductive, exacerbating the weaknesses of the current system while failing to capitalize on its potential.

Borrowing from recent work in political science, we characterize current politics as an era of “partisan volatility” with several defining characteristics. 13See Byron E. Shafer, The American Political Pattern: Stability and Change, 1932–2016 (2016). These include weak party organizations alongside strong partisan polarization; alternating periods of unified and divided government; the accumulation of policymaking power in the executive; and the weakening of moderating institutions such as the civil service. Together, these political facts have important consequences for how administrative law doctrines—from Lucia’s holding on ALJs to efforts to limit Chevron deference—should be understood.

The core takeaway from our argument is that contemporary administrative law is poorly suited to the political moment and is poised to become even more so. In the post-New Deal period of divided government that lasted from 1969–1992, an almost exclusively Republican executive, aligned around a coalescing coalition of social conservatives, defense hawks, and business interests, negotiated with a Congress that was the remaining holdout of the traditional labor-dominated Democratic Party. It was during this period that administrative law, as recognized today, developed its current form. Core features include notice-and-comment rulemaking as a chief vehicle for policy change and the (somewhat contradictory) judicial doctrines of hard look arbitrary or capricious review and Chevron deference. This period also saw the rise of what Professor Cass Sunstein refers to as the “cost-benefit state,” 14Cass R. Sunstein, The Cost-Benefit State: The Future of Regulatory Protection ix (2002). with its hallmark characteristic of White House regulatory review.

In retrospect, it is relatively clear that the election of Bill Clinton ended the period of divided government in favor of the new period of partisan volatility. Institutionally and doctrinally, change—where it has occurred—has primarily been accomplished via grafting, rather than the development of new legal regimes. Practices from the prior era, such as cost-benefit analysis and doctrines such as Chevron deference took on new meaning with alternating Democratic and Republican administrations, and during periods of united versus divided government. For a time, this process of grafting worked well enough, but the election of Donald Trump at the head of a newly unstable Republican party, and the level of tempest represented by his administration, may serve as the breaking point.

The current Supreme Court and the doctrinal program that it seems inclined to pursue appears extremely ill-suited to the task of responding to the needs created by this new political dynamic. We will review three areas—the failure to increase the stringency of review for policy reversals; the breaking down of politics-insulating barriers within administrative agencies; and the weakening of Chevron deference—where the current Court majority has made, and is poised to continue making, doctrine that does not fit the needs of the times. Even if the Court is set on this unproductive path, understanding the causes and consequences of today’s institutional and doctrinal pathologies may help future policymakers and jurists make better choices. 15See Richard H. Pildes, Institutional Formalism and Realism in Constitutional and Public Law, 2013 S. Ct. Rev. 1, 2 (2014) (“[Institutional] realism would entail constitutional and public law doctrines that penetrate the institutional black box and adapt legal doctrine to take account of how these institutions actually function in, and over, time.”). The current Supreme Court may be disposed toward the alternative “institutional formalism,” tinged with an ideological disposition to rein in the administrative state. See Gillian E. Metzger, 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1, 17–33 (2016) (describing the “current judicial challenges to national administrative government”).

This Article will proceed in three parts. Part I provides a theory of the relationship between politics and administration and explores how this relationship has unfolded in practice. We find that many of the inflection points in administrative law align with changes in the governing party system as jurists and other legal actors respond to their political realities. This alignment is no accident. Systems of public administration are vitally connected to broader “party systems.” 16 Throughout this Article we use the term “party systems” to refer to stable periods of partisan political alignment. The concept is explored in greater detail below. See infra Part II.A.1. Administrative institutions and doctrines adapt to those surroundings and, over time, compliment them as a unified (and stable) system. As Part I demonstrates, many features of our administrative system that emerged in one stable party system evolve under the pressures of later systems, taking on new meaning. This interaction has consequences for how we understand the law. The normative consequences of the answer to any question in administrative law—whether it be about how best to organize the government, review agency action, or structure oversight—can only be understood by referencing the broader political context. And, as this Part demonstrates, that context is shaped by stable party systems.

Part II examines the current party system, which exerts unique pressures on effective administration. Some features of the party system (e.g., ideological polarization and electoral volatility) suggests that policy positions pursued at the national level are likely to change dramatically and frequently. At the same time, a growing body of political science literature suggests that current political actors are not as constrained by forces that could stabilize or moderate more extreme policymaking. As we demonstrate below, the institutions that have historically played a stabilizing role in the policy process, including the Office of Information and Regulatory Affairs (OIRA) in the White House, the D.C. Circuit, and party elites, are increasingly marginalized or subsumed into partisan politics. Despite these challenges, our modern system of administrative law has not developed a responsive set of new doctrines to meet them.

Part III discusses three areas where the law could, at least in theory, mitigate some of the harmful effects of the current party system. First, we consider judicial review of policy reversals. Policy reversals are more common in the era of partisan volatility, as administrations change frequently and alternate between wholly divergent policy views. By constantly looking backward to existing policy, agencies devote considerable resources to examining the same issues over and over again, failing to take on new problems. At the same time, frequent reversals create costly regulatory uncertainty. This tendency could be counterbalanced through a doctrine that holds agencies to a more rigorous standard when they reverse course. Second, we look to current constitutional debates regarding agency independence, where the Court has recently resolved many challenges in ways that conflict with the broader political context. Specifically, the Court’s tendency to consolidate power over executive branch agencies and officials under the President is likely to exacerbate instability and partisanship in the administrative state. Finally, we examine current debates around judicial deference to agency interpretations of law. Again, the Court seems set on an unproductive reform path, targeting its review on “major questions” that tend to generate the most partisan friction. In so doing, it not only further risks threats to judicial independence, but it also misses an opportunity to play a stabilizing role by placing additional judicial scrutiny on agency actions that depart from prior interpretations.

The current era of partisan volatility has placed considerable pressure on existing institutions and raises challenges to a host of normative values. But the modern party system that we describe will not survive forever. As we outline below, stable political dynamics only last so long, and it is likely that the era of partisan volatility will be replaced by a system that poses different challenges to effective administration than the ones we now confront. When that new system emerges, our analysis of the costs and benefits of administrative law doctrine, which is oriented toward contemporary circumstances, will no longer reflect reality. At a more general level, however, our aim is to explore the dynamic and mutually influential connections between law, politics, and administration in U.S. legal institutions. Regardless of the specifics of the day, a better understanding of these relationships will allow for more informed policymaking, regardless of the specific pressures created by a new party system.

I. Politics and Administration

In this Part, we develop a theory of how politics and administration coevolve. Borrowing from the political science literature on party systems, we present a holistic understanding of this relationship that integrates several features of political systems, such as unified and divided government or partisan polarization. Party systems are characterized by several interacting features that include the balance of power between political parties, the substantive issues that dominate the policy debate, and the level of ideological and partisan polarization. No single variable explains any given party system; they all work in tandem to create a stable set of incentives, pressures, practices, and norms during each era. These forces, in turn, shape government administration and administrative law. Next, we demonstrate how this theory operates in practice through several illustrative examples from prior party systems. We conclude with a more thorough discussion of the era of divided government, which stretched from 1968 until 1992 and gave us much of our modern administrative law. As this Part shows, the era that ended in 1992 had unique political features and supported doctrine that may be ill-suited to other times.

A. Taking Party Systems Seriously

Although administrative law scholars have long been interested in the interaction of politics and administration, important features of this dynamics—at least as it has developed in the United States—have been overlooked. In the following discussion, we draw on existing literature from both political science and administrative law to argue that the relationship between politics and administration can be illuminated by understanding politics through the lens of party systems. Organized politics arises from the complex interplay of many social and political forces; the party systems literature focuses on a particular element of the resulting dynamic: periods of stability characterized by consistency in political institutions and behavior, interspersed by periods of transition and change. Party systems are at least partially stable and facilitate (and are facilitated by) corresponding systems of administration and administrative law. But just as stable party systems engender stable administrative systems, political change often brings administrative change as well. Focusing on stability and change at the systems level helps clarify how broader social and political focuses have helped shape U.S. administrative law over the past two centuries. 17 For perspectives on the use of complexity theory to understand system-level dynamics in politics, see generally Political Complexity: Nonlinear Models of Politics (Diana Richards ed., 2000). See also J.B. Ruhl, The Fitness of Law: Using Complexity Theory to Describe the Evolution of Law and Society and Its Practical Meaning for Democracy, 49 Vand. L. Rev. 1407 (1996) (discussing intersection of complexity theory and law).

The idealized notion that there is or can be a sharp distinction between politics and administration has long been abandoned. 18Compare Wilson, supra note 1, at 201 (Wilson advocated for a separate science of administration, apart from politics: “This is why there should be a science of administration which shall seek to straighten the paths of government, to make its business less unbusinesslike, to strengthen and purify its organization, and to crown its duties with dutifulness.”), with Waldo, supra note 2, at 91 (“Nothing is more central in thinking about public administration than the nature and interrelations of politics and administration. Nor are the nature and interrelations of politics and administration matters only for academic theorizing. What is more important in the day-to-day, year-to-year, decade-to-decade operation of government than the ways in which politics and administration are conceptualized, rationalized, and related one to the other.”). Nevertheless, administrative law is still sometimes treated as a set of abstractly derived and transcendentally applicable principles, rather than as part of a consistently shifting political landscape. 19See, e.g., Hamburger, supra note 12 (arguing that modern administrative law violates the common law Anglo-American tradition); Christopher Yoo & Steven Calabresi, The Unitary Executive: Presidential Power from Washington to Bush (2008); Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231 (1994) (arguing that the administrative state was unconstitutional from its creation as an originalist matter). There is also a substantially functionalist understanding of administrative law doctrine that focuses on how doctrines work in practice. See generally Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573 (1984) (arguing in favor of function-based understanding of separation of powers principles). But administrative law doctrines are far more intelligible (both normatively and positively) as components of a much larger complex of legal, political, social, and administrative/bureaucratic systems. Changes in the political environment both alter how doctrine functions and exert the kinds of destabilizing pressure that can result in doctrinal change. Descriptive accounts of the causes or consequences of administrative law that fail to account for political context are incomplete. Normative judgments about the goodness or badness of an administrative law doctrine lack substance unless reference is made to the political circumstances in which they will operate.

There is considerable descriptive literature that examines administrative law and institutions through a political lens. An important thread in this literature examines how administrative processes and structures respond to political pressures, especially during transitional moments. For example, focusing on the legislative process behind core administrative law statutes, political scientists Matthew McCubbins, Roger Noll, and Barry Weingast argue that legislators bargain over structure because they view it as essential to producing desired policy outcomes. 20 Matthew D. McCubbins, Roger G. Noll & Barry R. Weingast, Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 Va. L. Rev. 431, 432 (1989) (“[L]egislators see the choice of administrative structures and processes as important in assuring that agencies produce policy outcomes that legislators deem satisfactory.”). Under their account, the enacting majority uses its position not only to directly affect the law through substantive choices, but also to structure the administrative process. 21Id. In particular, the administrative process is used to empower the interests and groups favored by the enacting majority during the implementation of new statutory regimes, when Congress’s ability to monitor agencies and effect substantive changes to the law has waned. 22See Randall L. Calvert, Mathew D. McCubbins & Barry R. Weingast, A Theory of Political Control and Agency Discretion, 33 Am. J. Pol. Sci. 588, 604 (1989); see also Brian D. Feinstein, Designing Executive Agencies for Congressional Influence, 69 Admin. L. Rev. 259, 266 (2017) (conducting an empirical assessment of how agency design features support ongoing congressional oversight). Congress’s use of procedure to achieve political outcomes conflicts with the President’s desire to enact political priorities through implementation, with the courts often left balancing the two efforts at political influence. 23See Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 Colum. L. Rev. 1749, 1753 (2009) (“Rather, I credit the Court with an awareness of certain political facts in developing administrative law: Congress creates agencies with few constraints on their power and then seeks to control their decisionmaking, just as the President does. In essence, agencies are subject to two political principals. Furthermore, these principals may have divergent policy preferences and may seek agency decisions that move in different, even conflicting, directions. Under these circumstances, the Court might see its role as mediating the needs of both political branches for control of agency decisionmaking, consistent with separation of powers. In sincerely attending to procedural issues in connection with agency decisionmaking, the Court has attended to the needs of the legislative branch. It has rendered agency action more susceptible to ongoing congressional oversight ....”).

Conflict over structure extends beyond interest group-oriented fights in the legislature. Terry Moe and Scott Wilson focus on interactions between the President and Congress, and in particular on efforts by each institution to “engineer the structure of public bureaucracy and exercise control over it ....” 24 Terry M. Moe & Scott A. Wilson, Presidents and the Politics of Structure, 57 Law & Contemp. Probs. 1, 4 (1994). For Moe and Wilson, although the conflict between the legislature and the President often plays out in the context of specific policy debates—and therefore might appear to be about ideological commitments—the “more fundamental struggles” are “battles over structure.” 25Id. at 3–4. Important features of the administrative state, such as the civil service and regulatory review, represent an equilibrium that develops between contesting institutions, each vying for a greater share of influence over administrative decision-making. 26See, e.g., Daphna Renan, The Law Presidents Make, 103 Va. L. Rev. 805, 813–14 (2017) (describing how different models of legal advice in the executive branch empower different actors).

Although given less attention by political scientists, political circumstances may affect administrative law in the courts as well. At the level of individual decisions, there is a longstanding literature that correlates the political affiliation of Justices with their favored outcomes, a literature that has been extended to the federal courts. 27See David W. Rohde & Harold J. Spaeth, Supreme Court Decision Making 140–145 (1976); Jeffrey A. Segal & Harold J. Spaeth, The Supreme Court and the Attitudinal Model 65 (1993); Harold J. Spaeth, The Attitudinal Model, in Contemplating Courts 296 (Lee Epstein ed., 1995); see also Charles Grove Haines, General Observations on the Effects of Personal, Political, and Economic Influences in the Decisions of Judges, 17 Ill. L. Rev. 96, 114 (1922). See generally Lee Esptein, William M. Landes & Richard A. Posner, The Behavior of Federal Judges 27 (2013). However, that literature tends to focus on case outcomes, such as whether employers or employees prevail in a discrimination suit, rather than on structural questions concerning, for example, the standard of review. Even interactions between institutions—for example, whether judges are more willing to overturn the actions of cross-party Presidents—provide only limited insight on how external political circumstances affect the course of the law more generally. And judges themselves are frequently reticent to acknowledge how political circumstances might influence their decisions. 28 Chief Justice John Roberts famously said of his position: “my job is to call balls and strikes ....” See Chief Justice Roberts Statement—Nomination Process, U.S. Courts, http://www.uscourts.gov/educational-resources/educational-activities/chief-justice-roberts-statement-nomination-process [https://perma.cc/E63D-AYTG].

One exception to this reticence came from Justice Scalia. Prior to his initial appointment to the D.C. Circuit, Scalia was a co-editor for Regulation magazine and published an essay on administrative law subjects shortly after Ronald Reagan took office in 1981. In that essay, Scalia was quite explicit about the interaction between general administrative principles and the politics of the moment, especially given what he (correctly) saw as the significance of the Reagan election. In his words:

At a time when the GOP has gained control of the executive branch with an evident mandate for fundamental change in domestic policies ... every curtailment of desirable agency discretion obstructs (principally) the departure from a Democrat-produced, pro-regulatory status quo.
....
Executive-enfeebling measures ... do not specifically deter regulation. What they deter is change. Imposed upon a regulation-prone executive, they will on balance slow the increase of regulation; but imposed upon an executive that is seeking to dissolve the encrusted regulation of past decades, they will impede the dissolution. Regulatory reformers who do not recognize this fact, and who continue to support the unmodified proposals of the past as though the fundamental game had not been altered, will be scoring points for the other team. 29 Antonin Scalia, Regulatory Reform—The Game Has Changed, 4 Regulation 13, 13–14 (1981).

As will be discussed below, then-editor Antonin Scalia was accurately responding to a political reality that had been in place for roughly a decade, in which the Republican Party was able to largely sustain control of the White House, while Democrats held Congress. In this political context, greater discretion granted to the executive had a substantive effect of shifting policy away from a “Democratic-produced, pro-regulatory status quo.” 30Id. at 13. Once taking his position on the judiciary, and especially at the Supreme Court, Justice Scalia implemented his vision of executive empowerment, in part by becoming the chief defender of Chevron deference. 31See United States v. Mead Corp., 553 U.S. 218, 256–57 (2001) (Scalia, J., dissenting) (“To decide the present case, I would adhere to the original formulation of Chevron.... We accordingly presume—and our precedents have made clear to Congress that we presume—that, absent some clear textual indication to the contrary, Congress, when it left ambiguity in a statute meant for implementation by an agency, understood that the ambiguity would be resolved, first and foremost, by the agency, and desired the agency (rather than the courts) to possess whatever degree of discretion the ambiguity allows. Chevron sets forth an across-the-board presumption, which operates as a background rule of law against which Congress legislates: Ambiguity means Congress intended agency discretion. Any resolution of the ambiguity by the administering agency that is authoritative—that represents the official position of the agency—must be accepted by the courts if it is reasonable.” (citations omitted)). Unsurprisingly, once on the Court, the justifications he offered for executive deference changed—abandoning his prior language of raw-knuckled partisan advantage, Justice Scalia favored a justification grounded in Congress’s intent. 32See, e.g., Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke L.J. 511, 516 (1989). Nevertheless, the political reality that deference transferred power to a consistently Republican executive remained a background fact during the time that Justice Scalia was developing and expounding his views. Given that he specifically called attention to these background realities, drawing a causal connection between those political circumstances and his later jurisprudence does not seem entirely unwarranted.

From a normative perspective, law scholars have recently focused on the interaction of administrative law and the facts of the political environment. For example, Daryl Levinson and Rick Pildes argue in a 2006 paper that the party composition of the government closely affects congressional oversight of the executive, with unified party control of government enfeebling Congress’s executive-checking function. 33 Levinson & Pildes, supra note 3, at 2339. Jessica Bulman-Pozen examined the consequences of nationally oriented and ideologically polarized parties for federal-state interactions in the administration of national policy programs. 34 Bulman-Pozen, supra note 3, at 1080 (arguing that national polarized politics interacts with federalism). Bulman-Pozen argues that states have become sites of partisan contestation, with state government officials using their platforms to engage in ideologically charged actions to either antagonize or facilitate national-level policymaking. 35Id.; see also Jessica Bulman-Pozen & Heather Gerken, Uncooperative Federalism, 118 Yale L.J. 1256, 1267 (2009) (citing Larry Kramer, Understanding Federalism, 47 Vand. L. Rev. 1485, 1544 (1994)). Neal Devins and David Lewis have examined the interaction of party polarization and agency independence, finding that insulating factors, such as for-cause removal provisions, have only limited ability to reduce partisan influence over independent agencies. 36 Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. Rev. 459, 462 (2008). Thomas McGarity has examined the consequences of polarization for the rulemaking process and in particular the role of interest in influencing political oversight over agencies. 37 McGarity, supra note 3, at 1678. Eric Posner and Cass Sunstein note the prevalence of “flip-flops” in views over institutional constraints as party control changes hands. 38 Eric A. Posner & Cass R. Sunstein, Institutional Flip-Flops, 94 Tex. L. Rev. 485, 486 (2016). Earlier work by Livermore examines the interaction between the structure of party politics and executive and congressional oversight. 39 Matthew S. Broker & Michael A. Livermore, Centralizing Congressional Oversight, 32 J.L. & Pol. 261, 261 (2017); Livermore supra note 3, at 47. Daniel Farber and Anne Joseph O’Connell argue that the entire legal apparatus of administrative law—with its focus on regularized procedures and judicial review—is out of step with administrative realities such as the importance of White House regulatory review. 40 Daniel A. Farber & Anne Joseph O’Connell, The Lost World of Administrative Law, 92 Tex. L. Rev. 1137, 1160 (2014). With the election of Donald Trump, prominent law scholars have begun to examine the outer boundaries of administrative law that might be tested by an executive bent on disobedience. 41See, e.g., Nicholas R. Parrillo, The Endgame of Administrative Law: Governmental Disobedience and the Judicial Contempt Power, 131 Harv. L. Rev. 685, 687 (2018).

This growing literature grapples in important ways with the connection between administrative law and politics, and each contains an important element of the overall picture. We build on this work by emphasizing the connections between the overall structure of organized politics—the party system—and bureaucratic institutions and related legal rules.

1. A Systems-level Perspective on Organized Politics

Political science work on U.S. party systems investigates periods of stability in the organization of politics as well as the dynamics of change between these systems. 42See, e.g., Shafer, supra note 13, at xii. The transitions between party systems are often accompanied by “realignment elections” that demarcate when an old order has given way to a new. See, e.g., Alan I. Abramowitz & Kyle L. Saunders, Ideological Realignment in the U.S. Electorate, 60 J. Pol. 634, 636 (1998) (applying the theory to the 1994 and 1996 elections and finding that they resulted a realignment of party loyalties). See generally James L. Sundquist, Dynamics of the Party System: Alignment and Realignment of Political Parties in the United States 35–36 (1983); V.O. Key, Jr., A Theory of Critical Elections, 17 J. Pol. 3, 3–4 (1955). Roosevelt’s 1932 victory is the paradigmatic case of a realignment election. When political shifts (and accompanying legal changes) are sufficiently tectonic, they might even be referred to as “constitutional moments.” Bruce Ackerman, We the People: Foundations 58 (1991) (describing Reconstruction and the New Deal as the second and third constitutional moments). The paradigm of party systems and realignment elections has been challenged as failing to capture the nuances of partisan change. See David R. Mayhew, Electoral Realignments: A Critique of an American Genre 12 (2002). Although these critiques may have merit, we stick with the language of party systems and realignments for expositional reasons, understanding that the reality may not be as pat. Periods of relative stability include the years 1800–1828, when the Jeffersonian Democratic-Republicans governed an essentially one-party state, and the years 1898–1932 when a reformed Republican coalition dominated a more regional Democratic Party. These periods were characterized by lasting and consistent patterns in the organization of political life. The patterns retained some coherence over time, until eventually—after a period of transition—giving way to new forms of organization.

A focus on party systems springs from the centrality of parties (broadly understood) to the organization of politics in the United States. Parties serve a variety of functions, including aggregating voter preferences and organizing disparate groups together to compete in elections. 43 Parties can also be understood beyond their impact on the electoral process, such as through their associational features. See, e.g., Tashjian v. Republican Party of Conn., 479 U.S. 208, 225 (1986) (finding that political parties have associational rights under the First Amendment that limit state regulation). Parties can also be understood as purely endogenous institutions that serve as tools for elected officials. See generally John Aldrich, Why Parties? The Origin and Transformation of Political Parties in America 28–29 (1995). For our purposes, however, political parties are important for their electoral consequences and the subsequent substantive conflicts that emerge from their representation in office. For a broader study of the many roles played by the modern American political party, see generally Marjorie Randon Hershey, Party Politics in America 6–10 (17th ed. 2017) (defining parties as including “the party organization,” “the party in government” and “the party in the electorate”). As Justice Reed explained, “... political parties ... were created by necessity, by the need to organize the rapidly increasing population, scattered over our Land, so as to coordinate efforts to secure needed legislation and oppose that deemed undesirable.” Ray v. Blair, 343 U.S. 214, 220–21 (1952). The structure of American elections favors an equilibrium of two dominant political parties. 44See also Kenneth Benoit, Duverger’s Law and the Study of Electoral Systems, 4 French Pol. 69, 70 (2006) (stating Duverger’s Law as: “The simple-majority single-ballot system favours the two-party system”). See generally Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State 207–08 (Barbara North & Robert North trans., 1954). The interaction between these parties helps define the politics of that era. These eras are referred to within the political science literature as party systems. 45 William Nisbet Chambers, Party Development and the American Mainstream, in The American Party Systems: Stages of Political Development, 3-4 ( William Nisbet Chambers & Walter Dean Burnham eds., 2d ed. 1967).

These systems can be compactly understood through three organizing concepts: party balance, ideological polarization, and substantive conflict. 46See Shafer, supra note 13, at xii. The first concept captures the relative strength of the parties in electoral politics. The latter two concepts capture the breadth of the ideological differences between the parties and the substantive issues that make up their policy agendas, respectively. 47Id. From the Founding generation’s feuds between Hamilton and Jefferson to the polarization of the current political environment, these organizing concepts help to demarcate different periods of American life and to understand the resulting policy outcomes. Politics in the United States have progressed through many distinct phases, including the Jacksonian Era, New Deal Democratic dominance, and the entrenched divided government of the latter half of the twentieth century. 48 Chambers, supra note 45, at 3.

The pattern by which these systems emerge and collapse is familiar. 49Id. A stable party system exists when the two major parties compete for voters and influence, each winning over a relatively consistent coalition of voters. This coalition may lock in a majority of the electorate, leading to a period of sustained dominance. The coalitions may also be evenly matched, such that competitive elections are common. In either event, the coalitions themselves remain stable, even if the outcomes are less predictable. Moreover, the Constitution’s division of power may allow one party to lock in control of a single branch, while the other branch or state governments continue to support their opposition. 50See, e.g., Bulman-Pozen, supra note 3, at 1080, 1108 (arguing that both the state and federal governments are “sites of partisan affiliation,” such that voters are likely to identify more with a state government when the federal government is controlled by an opposing party). As a result, a stable party system may nonetheless include predictable periods of divided government. 51David R. Mayhew, Divided We Govern: Party Control, Lawmaking, and Investigations, 1946-2002, 4-5 (2005). Periods of relative stability can be upset by major political change. These inflection points often involve “realignment elections” that destabilize the status quo. 52 The concept of realignment elections has played a prominent role in political science since the publication of V.O. Key, A Theory of Critical Elections, 17 J. Pol. 3 (1955). See also Walter Dean Burnham, Critical Elections and the Mainsprings of American Politics 2–3 (1970) (providing an early account of critical realignment elections). See generally Mayhew, supra note 42, at 7–12 (summarizing the subsequent literature in this area). A similar dynamic is also observed in the policy process, where “political processes are characterized by stability and incrementalism, but occasionally ... produce large-scale departures from the past.” James L. True, Bryan D. Jones & Frank R. Baumgartner, Punctuated Equilibrium Theory: Explaining Stability and Change in Policymaking, in Theories of the Policy Process 155 (Paul A. Sabatier ed., 2007). Mayhew is critical of the concepts of realigning elections and critical elections as oversimplified and inaccurate. To Mayhew, these ideas fail because specific empirical claims that result from the theory concerning how long party systems last, whether critical elections look different from other elections, or whether specific policy outcomes result from critical elections, sometimes lack support. Mayhew, supra note 42, at 142. Although interesting, Mayhew’s critiques are largely orthogonal to how party systems are understood in this Article, as sustained periods of stable electoral coalitions, which in turn coincide with stable periods of administration. The empirical claims rejected by Mayhew are not required for this more modest use of the theory. Once the dust settles, new political affiliations take hold, with different coalitions and political dynamics. 53 The cyclical patterns of stability and realignment that characterize party systems can exist alongside other patterns of change, cyclical and otherwise. Political scientist Stephen Skowronek, for example, has argued for a cyclical notion of “political time” that is tied to the presidential leadership. Stephen Skowronek, Presidential Leadership in Political Time: Reprise and Reappraisal x (2011). There are also well-recognized secular trends that have endured across party systems, such as the growing reach and complexity of the administrative state. American politics is also, of course, embedded within broader historical realities involving, for example, race, gender, culture, globalization, and technological development. See, e.g., Theda Skocpol, Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (1992) (examining intersection of gender and politics during origin period of U.S. welfare state). Our examination of the relationships between party systems and administrative law is not meant to endorse any particular cleavage between law and society, but rather to emphasize an underappreciated set of interactions within the broader interpenetration of law and society. See William J. Novak, Response: The People’s Welfare Redux, 57 Am. J. Legal Hist. 248, 249 (2017) (“[T]he separation of law from politics and both from society and/or economy [is] an obstacle to historical understanding.”). Although there is no uniform agreement on the boundaries of each party system, 54 The election of 1968, for instance, has led to scholarly disagreement. See generally Arthur C. Paulson, Electoral Realignment and the Outlook for American Democracy 6–9 (2007) (surveying the academic disagreement on whether or not the 1968 election constituted a realignment). many inflection points are widely recognized as producing systemic changes in American politics.

2. Administrative Law through the Party System Lens

We draw from the literature on party systems to derive two basic insights concerning the interaction between administrative law and the organization of politics. The first and most general is that the organization of politics can be understood as periodized, meaning that over certain periods, there is a relatively stable and understandable structure to politics. 55 Chambers, supra note 45, at 7. During these periods, politics is not just “one damn thing after another,” but instead involves stable patterns over time. 56 Periodization raises its own set of issues concerning where to draw lines, and more generally, the criteria to use when engage in the line-drawing exercise. For an alternative to the party system approach, see Joel H. Silbey, The American Political Nation, 1838–1893, 6 (1991) (offering alternative periodization of political eras based on “other constituents of the political nation besides electoral behavior—the nature of political leadership, the importance of particular political institutions, the strength of government, popular attitudes toward politics, and the like”). We maintain the party system approach in part because of its familiarity and, more so, because it tracks features of the political environment that are particularly important for administrative law. That said, we also recognize that the party system framework might miss elements of political context—such as changes popular attitudes toward the role of experts—that are important to administrative law and may not be reflected in the structure of organized politics. For the origins of the “one damn thing after another” quote, see Quote Investigator, https://quoteinvestigator.com/2015/09/02/life-one/ (last visited July 6, 2019). Because administrative law interacts with politics, the stable dynamics of a party system will help shape a corresponding stable administrative system. The party systems literature shows that the organization of politics is not a matter of a single variable that might change over time, but the interaction of several variables (i.e., party balance; polarization; substantive dimensions) that tend to form stable regimes. 57See Shafer, supra note 13, at xii. Administrative law responds to regime-level characteristics, and therefore is best understood by looking to the system as a whole, rather than any single political reality (e.g., unified or divided government). Periods of transition between party systems demonstrate the dynamic and versatile nature of administrative law and administrative institutions.

 

From a normative perspective, the fit between an administrative law regime and its political period has important consequences. A mismatch between administrative law and the broader political context can have a wide range of undesirable effects across a variety of normative dimensions that include administrative values such as efficiency, impartiality, legality, and expertise; 58 For an account of the demand of the rule of law on administrative agencies, see Kevin M. Stack, An Administrative Jurisprudence, 115 Colum. L. Rev. 1985 (2015). democratic values such as accountability, transparency, and participation; and, most generally, in the ability of the law to benefit social well-being through sound policymaking. Consideration of these kinds of social consequences may affect (and under some jurisprudential theories, should affect) judges’ choices when shaping the law. 59See Charles L. Barzun, Three Forms of Legal Pragmatism, 95 Wash. U. L. Rev. 1003, 1004 (2018). Accounting for the organization of politics when shaping administrative law doctrine is not the same, and should not be understood as, judges using their position to construct law that favors a particular party program (Justice Scalia’s example notwithstanding). Rather, judges can take a pragmatic perspective that encompasses both the deep values embedded in the law as well as the political and administrative landscape in which their decisions will be given effect. Even if these considerations do not ultimately inform changes to legal doctrine or public policy, however, they are still a necessary part of understanding the effects of new developments.

Our second insight draws from the party systems literature to identify a specific break in the organization of politics in 1992, when a basic transformation occurred that only became apparent after some time. 60Shafer, supra note 13, at 122–23 (“Neither of these intendedly major changes, a new kind of party balance fostered by growing partisan independence or revised partisan majorities built on moving against ideological polarization, ever came close to realization. Someone had to win, and Bill Clinton did. Yet none of these putative strategic theories were of any use in predicting the veritable kaleidoscope of electoral outcomes that would follow.... The presence of all possible combinations of election outcomes and, even more strikingly, their rapid circulation and replacement, was what distinguished this period electorally from its predecessors.”). This new system, discussed extensively in Part II, is a meaningful break from the past. The differences between the system we now find ourselves in and the one that preceded it are vital to understanding the challenges of modern administrative law. 61See Mark Tushnet, Politics as Rational Deliberation or Theater: A Response to Institutional Flip-Flops, 94 Tex. L. Rev. 82, 83 (2016) (arguing that “charges of flip-flopping [over inconsistent positions on questions such as presidential power] result from the specific circumstances of contemporary U.S. politics, with a hyperpartisan and ideologically polarized party system”). It is often noted that current doctrine seems ill-suited to these challenges; although the reasons for the mismatch are often unexplained. A theory of administration that accounts for party systems explains why such a mismatch exists. Our current doctrines largely emerged during a time quite unlike the present. The party system approach helps clarify the causes and consequences of the current shortfalls in administrative law.

B. The Versatility of Administrative Law

There is a rich body of political science literature documenting the party systems that have existed in this country since its founding. 62See, generally Richard Hofstadter. The Idea of a Party System: The Rise of Legitimate Opposition in the United States, 1780–1840 (1969). Administrative law scholars have also examined legal regimes as they have existed at various time in the past, demonstrating how the law responded to the dominant political and social dynamics. 63See generally Jerry Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012). The aim of this section is to bring these two projects together. In doing so, we find that party systems track with stable periods of administrative law. When party systems change, the law tends to change with it. In some cases, existing parts of administrative law evolve to take on new meaning, such as judicial review doctrines following the New Deal. 64See infra notes 110–115 and accompanying text. At other times, new institutions or doctrines emerge to respond to political forces, such as the rise of civil service administration in the Gilded Age. 65See generally Silbey, supra note 56. In the end, this history suggests that administrative law is versatile and dynamic, responding to its political context.

This perspective can help destabilize current debates in administrative law—for example, over the removability of independent agency heads, the vitality of Chevron deference, or the proper approach to arbitrary and capricious review—which have taken on an aura of permanence. In reality, these debates are relatively recent and contextual. Remembering the versatility of U.S. administrative law—how it has been shaped to meet different demands at different times—undermines the sense that administrative law responds to some set of transcendent good-for-all-time principles rather than needs of the moment.

The role of the New Deal realignment in shaping U.S. administrative law is well-known, 66 The post-New Deal period generated a considerable amount of administrative law doctrine that is sufficiently vital that it continues to be taught in introductory administrative law courses. See Lawson, supra note 11 (categorizing the development of law as “Before the New Deal,” “The New Deal,” and “After the New Deal”). but these are hardly the only examples of innovation in administrative law or practice in response to a transition between party system. One early inflection point occurred in 1828, with the decline and dissolution of the Jeffersonian Democratic-Republican Party in favor of the first version of the contemporary Democratic Party, with Andrew Jackson at its head. This election is widely viewed as marking a transition from the first U.S. party system to the second, 67See generally Richard P. McCormick, The Second American Party System: Party Formation in the Jacksonian Era 3 (1966). but it also serves as a transition between two very different systems of administration.

Administration during the Jeffersonian period had a variety of characteristics that interacted with the politics of the time. During the Jeffersonian period, a patrician approach to governance coincided with a very broad political coalition that covered, more or less, the entirety of the American elite political class. 68See Mashaw, supra note 8, at 58–59 (describing the Washington Administration’s focus on finding administrators with “fitness of character” as the primary means of exercising internal discipline); id. at 175 (explaining the transition to a more bureaucratized approach to internal control during the Second Party System, when there was a much broader electorate). Although there were differences and debates within the Jeffersonian coalition, there were also sustained efforts by party leadership to maximally expand the party coalition. 69See generally George Dangerfield, The Era of Good Feelings (1989). Given the party’s reach, the group of respected party members overlapped more or less perfectly with social elites. So, selecting government officials based on party loyalty or professed ideology was redundant.

The small government and legislature-focused ideology of the Democratic-Republican party sometimes clashed with its policy ambitions, leading to the seemingly contradictory broad delegation of power to the President or administrative official to oversee complex regimes, such as the development of western lands and the embargo of French and British commerce. 70See Mashaw, supra note 8, at 98, 126 (discussing embargo and public lands management, respectively). Other peculiarities of administration law of the era were a better fit for patrician, small government ideology. In early America, judicial review took the form of suits against officers personally, rather than the government. 71 Federal sovereign immunity still bars some claims against the government directly, and suits against officers are still common. Nonetheless, statutes like the Federal Tort Claims Act, the Tucker Act, and the Administrative Procedure Act now waive sovereign immunity in many contexts for which relief was barred in Early America. See generally Gregory C. Sisk, A Primer on the Doctrine of Federal Sovereign Immunity, 58 Okla. L. Rev. 439, 458 (2005) (documenting “the proliferation of statutory waivers of sovereign immunity”). In the era before qualified immunity, the officer’s only defense to such a private suit was to claim authorization by law. 72 The legality of government conduct would often be tested by determining the validity of the authorization defense. This was the posture of classic cases like Little v. Barreme, where contests about inherent presidential authority and statutory interpretation arise in the context of personal actions against officers. 6 U.S. (2 Cranch) 170, 177 (1804). These cases could arise in state as well as federal courts. See, e.g., Olney v. Arnold, 3 U.S. (3 Dall.) 308, 308 (1796). The resulting court oversight, while capable of determining the legality of official conduct, also led to a “system of common law remedies [that was able to] disable administration” by paralyzing officials with personal liability. 73Mashaw, supra note 8, at 114. The relatively thin administrative apparatus and the personal posture of suits against government officials also placed courts in the position of deciding quasi-administrative questions, such as whether a vessel was seaworthy. 74Id. at 74. A lack of national-level policymaking capacity as well as notions of local self-governance popular at the time also fit with the regulatory importance of state and local governments. 75See generally William J. Novak, The People’s Welfare: Law and Regulation in Nineteenth-Century America 1 (1996).

Even when Congress did decide that a larger federal bureaucracy was necessary during this period, it included features that empowered private actors. This took the form of qui tam statutes, which allow private persons to assert the rights of the government. These statutes were common in the colonial period and used by states for much of early American history, 76 Randy Beck, Promoting Executive Accountability Through Qui Tam Legislation, 21 Chap. L. Rev. 41, 44–45 (2018) (“Regulation of government officials through qui tam legislation was widely practiced in the American colonies and early states. Qui tam monitoring was used to promote statutory compliance by an enormous variety of state officials, particularly those performing decentralized functions such as road construction and maintenance, judicial administration, and regulation of commercial activities. It was common for early states to rely on qui tam oversight to ensure lawful conduct by officials performing functions critical to public confidence in government, such as conducting elections and collecting taxes.”). helping solve the “difficult management challenge created by a large and dispersed federal workforce.” 77See Randy Beck, Qui Tam Litigation Against Federal Officials: A Neglected History, 93 Notre Dame L. Rev. 1235, 1294 (2018) (“Why did Congress authorize qui tam enforcement of two forfeitures directed at customs officials under the Collection Act when it opted not to use qui tam enforcement for other aspects of the legislation? The decision likely flowed from the difficult management challenge created by a large and dispersed federal workforce.”). In tandem with the other common law remedies, these statutes helped maintain control in a period before a bureaucratic and professional civil service. They also reinforced a political ideology skeptical of government intrusion.

Jackson’s election changed American politics, and administrative form quickly followed. The reformed Democratic Party was engineered by figures, such as Martin Van Buren, with a specific set of ideological and political goals. Partially, the party’s leaders sought a return to the principles of small and weak federal government at the root of Jefferson’s party, which had gradually drifted in the direction of the Federalists’ orientation toward nation-building. 78See generally David N. Mayer, The Constitutional Thought of Thomas Jefferson 123 (1994) (examining Jefferson’s politics in comparison to Federalist and Whig viewpoints). Additionally, Van Buren viewed with great fear the prospect that, with the dissolution of the ecumenical Democratic-Republicans, partisan differences might map onto regional differences, allowing the issue of slavery—largely submerged during the early years of the Republic—to rise to the forefront of American politics. 79See, e.g., Robert Remini, Martin Van Buren and Making of the Democratic Party 123 (1959) (detailing Van Buren’s attempt to build the Democratic Party during the period of increased suffrage before the election of Andrew Jackson). The Democratic Party coalition was largely structured to avoid this possibility by threading together a geographically diverse set of interests. 80See Richard P. McCormick, The Second American Party System: Party Formation in the Jacksonian Era 14 (1966) (describing how the Democratic Party competed in different regions of the country).

For the first time in U.S. history, political parties oriented toward mass mobilization took up a central place in the organization of political life. 81Silbey, supra note 56, at 46. Anti-party sentiment that had characterized American political culture since the colonial period was abandoned in favor of an embrace of a collective approach to politics that emphasized the inevitability of conflict in a pluralistic society and the need for institutions capable of mediating that conflict. 82See generally id. (discussing embrace of parties during this time). The increased emphasis on enhancing the organizational capacity of parties was both a reason for and response to an exploding level of electoral participation among voters, especially in national elections. 83 There was a threefold increase in the absolute number of votes cast in the 1824 and 1828 presidential elections, primarily driven by a growth of voter turnout. Popular vote numbers are collected by the American Presidency Project, housed at U.C. Santa Barbara. See John Woolley & Gerhard Peters, The American Presidency Project, http://www.presidency.ucsb.edu. Population statistics are available at Bureau of the Census, U.S. Dep’t of Commerce, Historical Statistics of the United States: Colonial Times to 1970 (1975), https://www.census.gov/history/pdf/histstats-colonial-1970.pdf.

The politics of the Jacksonian period favored a very different approach to administration. With the most divisive issue of the day driven out of focus by the geographic structure of the parties, political contestation focused on distributive questions, with ideological disputes between the dominant Democratic Party and the opposition Whigs limited to the old limited-government-versus-nation-building question. In addition, the mass mobilization politics of the era made government by patrician substantially less palatable. Together, the redistributive and anti-elite components of Jacksonian populism led to the spoils system, an approach to administration that lasted in one form or another until the Progressive Era at the turn of the century. In place of the Jeffersonian emphasis on the longstanding office holder, the spoils system placed party loyalists into administrative posts on a rotating basis. This system of rewards provided immediate financial incentives for party participation, helping to fuel mass mobilization. It also helped spur, perhaps ironically, a more bureaucratic civil service: Clear roles were far more important when governance was carried out by rotating party members. 84See Mashaw, supra note 8, at 177–78. As the bureaucracy, including law enforcement, became more professional, the need for extensive qui tam actions also diminished. 85 Note, The History and Development of Qui Tam, 1972 Wash. U. L.Q. 81, 101 (1972) (“What emerges from this historical evolution is recognition that in America, as earlier in England, qui tam proceedings began as a useful and perhaps necessary supplement to the efforts at law enforcement of inadequate public agencies. As the public agencies became more effective [during the latter part of the nineteenth century], the need for qui tam actions diminished.”).

Jacksonian America was undone by the Civil War and subsequent Reconstruction, replacing Democratic dominance with a period of closely divided elections. 86 The intervening Civil War and Reconstruction were perhaps the most important remaking of American politics, and they both had transformative effects on the administrative state. See generally Bruce Ackerman, We the People: Foundations 81 (1991) (presenting the Civil War and the subsequent work of the Reconstruction Republicans as a constitutional moment). Control of the White House changed four times during this period. In addition, there were two elections in which the popular vote winner did not prevail in the Electoral College. See Woolley, supra note 83. This was driven by Democrats’ ability to attract new immigrants to their party. See Mashaw, supra note 8, at 229. The years after the war also strained the federal government’s capacity, as Congress and the President found it difficult to manage the growing bureaucracy. 87Mashaw, supra note 8, at 233, 241–42. This period is also the starting point for Skowronek’s review of administrative state building. See generally Skowronek, supra note 10, at 39–45. Administrative law responded to these forces. First, the patronage system, which was now so large that presidential management was no longer possible, was gradually replaced by a civil service system. The merit-based system allowed each presidential administration to insulate personnel from their successors, who were often of a different party. 88Mashaw, supra note 8, at 239. Second, the agencies of the time developed their own internal procedures to fill the void left by an overwhelmed legislative branch, often proposing legislation to codify their existing practices. 89Id. at 243, 257–58 (providing an example involving military pensions). Together, these two changes empower relatively insulated administrators to weather a period of political volatility.

Another moment when a transition in American politics led to a remaking of administrative law and practice was after the election of William McKinley in 1896. This election ended the postbellum period of relatively evenly matched contests between the Republican and Democratic parties in favor of a period of Republican dominance that is sometimes referred to as the fourth party system. 90 See James E. Campbell, Party Systems and Realignments in the United States, 1868–2004, 30 Soc. Sci. Hist. 359, 380 (2006) (finding that 1896 reflected a realignment of political power towards Republicans at the presidential and congressional levels). For three decades, the Republicans retained a near lock on the White House, as well as control over both houses of Congress. 91Republican Party, Encyclopedia Britannica, https://www.britannica.com/topic/Republican-Party (last visited Aug. 12, 2019). The only deviation during this period was the election of Democrat Woodrow Wilson, who won with 42% over a fractured Republican field featuring a third-party challenge from Teddy Roosevelt’s Progressive Party, which finished second. United States Presidential Election Results, Encyclopedia Britannica, https://www.britannica.com/topic/United-States-Presidential-Election-Results-1788863 (last visited Aug. 17, 2019). The only real interruption of congressional control came during World War I, after which Republicans once again gained control of Congress in the fourteen years leading up to the Great Depression. Party Divisions of the House of Representatives, 1789 to Present, History, Art & Archives: U.S. House of Representatives, https://history.house.gov/Institution/Party-Divisions/Party-Divisions/ (last visited Aug. 17, 2019). The Democratic party of the time was largely relegated to regional status, where it dominated in the Jim Crow South. 92See R. Hal Williams, Realigning America: Williams, Bryan, and the Remarkable Election of 1896, 11 (2010). With a largely free hand to shape national policy, the Republicans pursued an incongruous mix of laissez-faire economic policies, concessions to the growing labor movement, and Progressive Era initiatives that included prohibition, trust-busting, and political reform. 93See Robert L. Rabin, Federal Regulation in Historic Perspective, 38 Stan. L. Rev. 1189, 1216–29 (1986).

Changes to administrative law and practice accompanied, and to some degree preceded, these political shifts. 94 For an important overview of bureaucrat development of the time, see Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies 1862–1928, 7 (2001). For other foundational work, see Skowronek, supra note 10, at 39–45. The Interstate Commerce Commission (ICC) was established roughly a decade before the McKinley wave election, creating a new paradigm for governmental organization. 95 The passage of the Interstate Commerce Act is often seen as the start of “traditional” administrative law. See Stephen Skowronek, supra note 10, at 121 (beginning his analysis with 1877 with a discussion of the Interstate Commerce Act); see also Rabin, supra note 93, at 1189 (“[W]hen Congress established the Interstate Commerce Commission, it initiated a new epoch in responsibilities of federal government. For the first time, a national legislative scheme was enacted that provided for wide-ranging regulatory controls over an industry that was vital to the nation’s economy—the railroads.”). The Act itself was new in how it “defines the structural position of administrative agencies within the governmental system, specifies the decisional procedures those agencies must follow, and determines the availability and scope of review of their actions by the independent judiciary.” Richard B. Stewart, Administrative Law in the Twenty-First Century, 78 N.Y.U. L. Rev. 437, 438 (2003). The Pendleton Civil Service Reform Act, which established the merit-based civil service system that ultimately eliminated the spoils system at the federal level, was adopted only a few years earlier. 96 27 Stat. 403 (1883). The growing concerns regarding excess by public servants led to a shift away from bounties and towards official salaries as well. See Nicholas Parrillo, Against the Profit Motive: The Salary Revolution in American Government 1780–1940, 17 (2013). The Sherman Antitrust Act, the first major competition law at the federal level, was adopted in 1890. 97 26 Stat. 209 (1890). In the ensuing years, it was the dominant Republicans that administered these laws, translating the general statutory directives into policies.

The administrative style during the fourth party system was markedly different than the Jacksonian or postbellum period. Broader cultural trends enhanced the legitimacy of experts, a position starkly at odds with Jackson’s populism. 98Skowronek, supra note 10, at 175. Coupled with civil service protections meant to root out the patronage networks that had supported the earlier political system, the result was a professionalized federal bureaucracy that was relatively insulated from electoral politics. 99Ronald N. Johnson & Gary D. Libecap, The Federal Civil Service System and the Problems of Bureaucracy 12 (1994) (“[I]n the late nineteenth century, with the enthusiastic support of the president Congress voted to restrict the number of patronage positions that were available. With the enactment of the Pendleton Act (22 Stat. 403) on 16 January 1883, the process was established by which patronage was to give way gradually to merit-based employment. By 1904, only twenty-one years after the Pendleton Act was passed, over 50 percent of the total federal civilian labor force was under merit provisions.”). The laissez-faire ideology and business constituency of the dominant Republicans, however, meant that direct regulation was largely disfavored, with courts continuing to play a central role—the Sherman Act’s private cause of action and criminal enforcement is perhaps the paradigmatic instance of this court-centric approach. 100 Rabin, supra note 93, at 1216–18. Even when bolder regulatory action was taken during this period, such as through the passage of the Pure Food and Drug Act, the resulting law was often friendly to industrial interests that saw the law as a stabilizing force for business. 101Id. at 1224–26.

The administrative system between the Civil War and the New Deal was a time of deep transition. There is some disagreement over whether this change was a fundamental break with the past, as Skowronek argues, 102Skowronek, supra note 10, at 285–87 (describing the period as one of “Modern American state-building”). To Skowronek, the system that emerged in this period was in direct contrast to the prior “state of courts and parties,” supported by “an intellectual vanguard” that “articulated the limitations” of the old system. Id. at 286. or simply the further development of existing administrative capacities. 103See Mashaw, supra note 8, at 4–9. What is clearer, however, is that many features of our modern administrative law—direct judicial review of agency action, independent regulators, and broad support for agency expertise—grew more prominent as the federal government expanded. In doing so, these features of the regulatory state were responding to a political movement that was supportive of broad national policymaking to address health, safety, and economic issues, but that was still skeptical of a fully administrative model. 104 Rabin, supra note 93, at 1219–20 (summarizing the competing view of Progressive Era and concluding that “neither position anticipated government intervention of the kind that was to result from the New Deal perception of wholesale market malfunction”).

The New Deal is the most widely recognized period when political transformation resulted in a radically altered administrative landscape. After the landslide election of 1932, and in the midst of the crisis of the Great Depression, an electoral majority existed unlike any other in modern politics: From 1932–1938, Democrats never held fewer than 59 Senate seats or 313 House seats. 105Shafer, supra note 13, at 7. Most critically, these partisan advantages were so great that a single part of the Democratic coalition, the more liberal Northern Democrats, constituted an absolute majority on its own. 106Id. at 6 (noting that the “elections of 1934 and 1936 were to be the only ones in American history in which Northern Democrats constituted a majority of the House of Representatives”). The result of this partisan asymmetry was a flurry of legislative activity, including both temporary responses to the Great Depression and lasting legislation like the Social Security Act. 107 Legislation from the first hundred days included the Federal Securities Act, Pub. L. No. 73-22, 48 Stat. 74 (1933); Agricultural Adjustment Act, Pub. L. No. 73-10, 48 Stat. 31 (1933); Glass-Steagall Act, Pub. L. No 73-66, 48 Stat. 612 (1933) (Glass-Steagall is part of the broader U.S.A. Banking Act); and the National Industrial Recovery Act, Pub. L. No. 73-66, 48 Stat. 195 (1933). The second wave of legislation included the Banking Act of 1935 and the Social Security Act, Pub. L. No. 74-271, 49 Stat. 620 (1935). For a general discussion of these two waves, see Shafer, supra note 13, at 12, 16 (collecting legislation). Political priorities, moreover, were set from the top, with the President clearly pushing a platform and controlling the party. 108 Almost all of these legislative initiatives began in the White House itself, rather than in the congressional committees. Id. at 16. In short, this six-year period featured a powerful executive at the head of a powerful party that held, essentially, unchecked power over the political branches.

These political incentives led to substantial changes to the administrative state, both in its size and its authority. Administrative agencies were created, both to manage the new regulatory programs and to directly provide sources of federal employment. 109 Agencies like the Tennessee Valley Authority (TVA) exemplify how these two aims coexisted. Created as a public corporation at the start of the New Deal program in 1933, the TVA pursued an agenda of improving resource management along the Tennessee River and Mississippi River Basins, while also serving as a source of short-term jobs for millions of unemployed Americans. Tennessee Valley Authority Act, Pub. L. No. 73-17 (1933). Much like Jackson’s spoils system or civil service system of the 1870s, federal administrative employment itself once again became a forum for policymaking, not just a means to an end. Legal constraints on the administrative state were also greatly reduced, allowing for unprecedented congressional innovation. After initial setbacks, the Supreme Court soon approved of both broad delegations of powers to agencies 110See Mistretta v. United States, 488 U.S. 361, 371–72 (1989) (discussing the history of the non-delegation doctrine before and after 1935). In 1935, the Court had invalidated two provisions on non-delegation grounds. See A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495, 529, 542 (1935) (invalidating a provision of the National Industrial Recovery Act allowing the president to designate “codes of fair competition”); Panama Refining Co. v. Ryan, 293 U.S. 388, 414–20 (1935) (invalidating a provision of the National Industrial Recovery Act that authorized the President to prohibit petroleum commerce). and broad assertions of power under Article I. 111See Wickard v. Filburn, 317 U.S. 111 (1942); NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1, 43 (1937). At no point during this period did the Court invalidate delegations to the New Deal agencies, 112 Kagan, supra note 3, at 2365–66 (noting that the two statutes invalidated on non-delegation grounds involved delegations directly to the President). despite the fact that some of these delegations were incredibly broad. 113 Stewart, supra note 2, at 1677. The Court also limited the reach of substantive due process as a restriction on economic regulation. 114See W. Coast Hotel Co. v. Parrish, 300 U.S. 379, 391 (1937). While this moment is frequently framed in “external terms,” as the Court responding to political pressure, Barry Cushman has offered an “internal” account, relying on the “legal intellectual dimension” of the shift in doctrine. See Barry Cushman, Rethinking the New Deal Court, 80 Va. L. Rev. 201, 207 (1994). Just as American politics demanded unprecedented national legislation to tackle the depression, the legal constraints on those actions—whether through powers (Commerce Clause), rights (substantive due process), or structure (non-delegation)—gave way. 115 These changes were supported in a new scholarly movement in favor of expert regulation, reflected in James Landis’ Administrative Process, published in 1938. James M. Landis, The Administrative Process (1938). Defending the administrative state against its detractors, Landis argued that the new administrative process was a superior means of exercising both judicial and legislative functions and offered a way to “modernize” the separation of powers. This view was particularly skeptical of the courts, with Landis noting that “judicial review over administrative action gives a sense of battle” leading courts to “thwart the effects of legislative judgments ... under the guise of constitutional and statutory interpretation.” James Landis, Administrative Policies and the Courts, 47 Yale L.J. 519, 519 (1937). His explanation for why administrative agencies were superior to courts not only included the familiar description of expertise and practicality, but also an assault on the “judicial conservatism” that was, in his account, ill-suited to the demands of the modern economy. James Landis, The Place of Administrative Law, 13 Conn. Bar J. 71, 72–77 (1939). The new administrative state was also buoyed by legal doctrine, developed during the New Deal period, that allowed for statutory independence from presidential control. See Wiener v. United States, 357 U.S. 349 (1958); Humphrey’s Ex’r v. United States, 295 U.S. 602, 631–32 (1935).

At the same time that the structure of party politics provided the Roosevelt Administration with unprecedented scope, the build-out of the administrative state was used purposefully by Roosevelt to undercut the importance of traditional parties, in particular by substituting agencies and formal programs for party patronage as the primary method for distributing state benefits. 116Sidney M. Milkis, The President and the Parties: The Transformation of the American Party System Since the New Deal 149–161 (1993).

 

After the high-water mark of the New Deal, the politics of the subsequent three decades largely entrenched the existing policy program, while making concessions to opposition interests when necessary. From 1940 to 1968, the Democrats sustained congressional majorities for all but a few years and won five out of seven presidential elections. But the margins were much closer, and internal division within the Democratic Party over civil rights created frequent opportunities for cross-party coalitions: Congressional voting during the time is markedly nonpartisan when compared to later periods. 117See Harry J. Enten, Were Republicans Really the Party of Civil Rights in the 1960s?, Guardian (Aug. 28, 2013, 9:15 AM), https://www.theguardian.com/commentisfree/2013/aug/28/republicans-party-of-civil-rights (documenting the regional, rather than partisan, difference in the vote). Accommodation to interests unhappy with the New Deal program and the administrative apparatus that accompanied it became necessary. Administrative law again shifted to account for this new political dynamic, most importantly with the Administrative Procedure Act (APA), passed in 1946. 118 Pub. L. No. 79-404, 60 Stat. 237 (1946).

Under the understanding of the politics/administration dichotomy popular at the time, agencies served as a “transmission belt” for legislative policy. 119 Stewart, supra note 2, at 1684. Concerns that agencies under presidential control might malfunction in that role were addressed through judicial review, which ensured that agencies were acting within their appropriate delegation of authority and not infringing upon individual private rights in the process. 120Id. at 1761–76. More probing judicial review, and in particular review on behalf of regulatory beneficiaries, might well have seemed unnecessary when the nation was still governed by a dominant single party, the prospect of regulatory legislation was relatively high, and the agencies could fairly be understood as an extension of the legislature.

As these examples demonstrate, administrative law adjusts to surrounding political forces through both doctrinal change and the development of new institutional arrangements. These illustrations are far from complete, but they do illuminate a system of administrative control that is highly contextual. The institutions that comprise this system—courts, agencies, and elected officials—are versatile, adjusting their role in administration to new political incentives. This adaptability does not make good administration inevitable, but it does suggest that there is potential for responsive change.

Our discussion of the first two centuries of American administration was abbreviated, but starting in 1968 it becomes necessary to slow things down. The political changes that followed the late New Deal era would bring about a complete overhaul in administrative law. Since the doctrines that arose from this period are the doctrines we still live with today, a more thorough examination of the period is needed.

C. The Last Era of Administrative Law, 1969–1992

Richard Nixon’s election in 1968 ushered in a new structure to U.S. party politics. The era of Democratic domination was over, and Republicans would hold the White House (with the limited exception of Jimmy Carter’s one term) for over two decades. Republican dominance was not complete, however, because the Democrats maintained majorities in Congress, and especially the House. 121Shafer, supra note 13, at 72–74. Writing at the time, Morris Fiorina argued that this pattern of consistent divided government signaled that the idea of realignment was “a dead concept.” See Morris P. Fiorina, An Era of Divided Government, in Developments in American Politics 324 (G. Peele et al. eds., 1992). He also noted that the era from 1968 to 1988 was a historical outlier for the prevalence of divided government. Id. at 325–26. Thus, the characteristic feature of this period was divided government, with the Democrats controlling the legislature and the Republicans controlling the White House. This was also a time of realignment in the ideological direction of the parties, with the ascendance of the conservative wing of the Republican Party and the decline of labor in the Democratic Party, leading to an increase in polarization and a shift in the substantive cleavage on issues that separated the parties. 122See Lawrence C. Dodd & Bruce I. Oppenheimer, Congress Reconsidered 94–97 figs.4-4 to 4-8 (11th ed. 2016) (empirically demonstrating increased polarization throughout the period and noting that “survey data confirm that the congressional parties represent increasingly distinctive sets of voters.”); Shafer, supra note 13, at 94–98 (describing the increased ideological polarization of party activists).

Changes in the organization of politics resulted in shifts in administrative law during this time that can be roughly divided into two periods: a reformation period that was characterized by an expansion of judicial oversight of administrative decision making, followed by a presidentialization period characterized by the increasingly successful imposition of presidential authority over agencies. 123See Mark Seidenfeld, A Big Picture Approach to Presidential Influence on Agency Policy-Making, 80 Iowa L. Rev. 1 (1994) (discussing justifications for growth of presidential power over agencies); Sidney A. Shapiro, Administrative Law After the Counter-Reformation: Restoring Faith in Pragmatic Government, 48 U. Kan. L. Rev. 689, 707 (2000) (“Reformers ... seek[ing] to combat government failure ... initially chose presidential oversight as the principal procedural vehicle for reform, and the Reagan and Bush Administrations put these plans into effect.”). By the end of these two waves of reform, the system of administrative law that is familiar today was in place.

The advent of divided government did not spell the end of legislative productivity, and bipartisan legislation was common. 124 For example, the Occupational Safety and Health Act, Pub. L. No. 91-596 (1970), passed with 83 votes in the Senate (35 of them Republican) and 310 votes in the House (138 of them Republican). See Votes to Pass S. 2193 (91st Congress), GovTrack, https://www.govtrack.us/congress/votes/91-1970/s599 (last visited Aug. 12, 2019) (showing the vote count in the Senate); Votes to Adopt the Conference Report on S. 2193 (91st Congress), GovTrack, https://www.govtrack.us/congress/votes/91-1970/h419 (last visited Aug. 12, 2019). The Clean Water Act, Pub. L. No. 92-500 (1972) was passed by bipartisan majorities following a veto by President Nixon. See Clean Water: Congress Overrides Presidential Veto, in CQ Almanac1972, at 11–17 (28th ed. 1973). Scholars of the period were struck with how divided government did not seem to undermine effectiveness. See generally Mayhew, supra note 51 (arguing that divided government was an effective system for developing public policy). The effectiveness of divided government during this period came from its stability. As Shafer put it, “cross-partisan and cross-institutional negotiations ... characterize[d] the policymaking process of the era of divided government [because the] particular and recurrent institutional split implied that the two parties could not hope to be rescued electorally, at least in the short run.” Shafer, supra note 13, at 104. The era of partisan volatility created very different incentives, a point we return to below. From the perspective of federal administration, however, this political dynamic created a complication. Many of the regulatory statutes, passed during the New Deal or pushed by liberal members of Congress, were now being administered almost exclusively by Republicans. 125See supra Schafer note 13, at 72–74. Allegations of agencies “captured” by industry abounded, both from the outside and in the courts. 126 See Thomas W. Merrill, Capture Theory and the Courts: 1967-1983, 72 Chi.-Kent L. Rev. 1039, 1043 (1996) (“There was during the period [1967 to 1983] no loss of faith in activist government. But a key instrumentality of activist government—the administrative agency—came to be regarded as suffering from pathologies not shared by other governmental institutions such as legislatures or courts. The principal pathology emphasized during these years was ‘capture,’ meaning that agencies were regarded as being uniquely susceptible to domination by the industry they were charged with regulating.”); George Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3, 3 (1971). To many on the left, the broad discretion asserted during the New Deal era, expressed in guidance like the Attorney General’s Manual, 127U.S. Dep’t of Justice, Attorney General’s Manual on the Administrative Procedure Act (1947). The manual was understood as embodying a hands-off role for the federal courts. Professor John Duffy described the manual as “a highly political document designed to minimize the impact of the new statute on executive agencies, shrewdly characterized the APA provisions governing judicial review as merely a ‘restatement’ and thereby invited courts and the bar to treat the Act as something less than a statute, as subservient to judge-made doctrine.” John F. Duffy, Administrative Common Law in Judicial Review, 77 Tex. L. Rev. 113, 119 (1998). looked less attractive when exercised by presidents less committed to the statutes’ regulatory objectives. This was the backdrop for the fundamental transformation of the 1960s and 1970s, famously dubbed by Professor Stewart as “the reformation of American administrative law.” 128 Stewart, supra note 2, at 1669.

The reformation was characterized by a rethinking of the role of judicial review. 129 This rethinking of the judicial role often played out in disagreements among the justices. In Abbott Laboratories v. Gardner, for instance, a majority of Court adopted a view of ripeness that allowed more private challenges to agency action. See 387 U.S. 136, 141 (1967) (finding that agency action could be ripe for judicial review prior to enforcement). Justice Fortas, long a political supporter of the New Deal Democratic Party, disagreed, advocating for adherence to the traditional view. See Toilet Goods Ass’n v. Gardner, 387 U.S. 167, 175 (1967) (Fortas, J., concurring in part and dissenting in part) (disagreeing with the holding in Abbott Laboratories and noting “that established principles of jurisprudence, solidly rooted in the constitutional structure of our Government, require that the courts should not intervene in the administrative process at this stage ....”). In rough strokes, during the long New Deal period, a (largely) Democratic administration was carrying out the policy program of a Democratic legislature, and the role of judicial review was to accommodate negatively affected interests, who were represented by the weak but not entirely marginalized Republicans. Once the Republican Party took over the White House, this dynamic shifted dramatically, drawing attention to the importance of discretion under vague and open-ended statutes. 130 Stewart, supra note 2, at 1683 (“[T]he exercise of agency discretion is inevitably seen as the essentially legislative process of adjusting the competing claims of various private interests affected by agency policy.”). Judges appointed during the long New Deal began to see the purely negative program of the prior model as inadequate. Instead, courts expanded their role to warding against agencies that failed to “affirmatively carry out the legislative mandates.” 131Id. at 1682. The court most closely associated with the reformation program was the D.C. Circuit. 132See, e.g., Greater Bos. Television Corp. v. FCC, 444 F.2d 841, 851 (D.C. Cir. 1970) (“Its supervisory function calls on the court to intervene not merely in case of procedural inadequacies, or bypassing of the mandate in the legislative charter, but more broadly if the court becomes aware, especially from a combination of danger signals, that the agency has not really taken a ‘hard look’ at the salient problems, and has not genuinely engaged in reasons decision-making.”). Gary Lawson’s administrative law casebook provides a colorful account of this period on the D.C. Circuit; see Gary Lawson, Federal Administrative Law 357 & nn.28–31 (7th ed. 2016). Components of the reformation included relaxed standing requirements as well as new procedural and substantive requirements that were placed on agency rulemaking. 133See, e.g., Portland Cement Ass’n v. Ruckelshaus, 486 F.2d 375, 394 (D.C. Cir. 1975) (requiring that agencies make available information that was relied on during the rulemaking as part of “hard look” review); Auto. Parts & Accessories Ass’n v. Boyd, 407 F.2d 330, 338 (D.C. Cir. 1968) (noting that judicial review ensures “major issues of policy were ventilated” during rulemaking process); Scenic Hudson Pres. Conference v. FPC, 354 F.2d 608, 616 (2d Cir. 1965) (finding that parties had standing to sue because of their recreational interest in “the conservation of natural resources”). For a discussion of the expansion of standing throughout this period, see generally Stewart, supra note 2, at 1723–48. For a thorough description of the debate on the D.C. Circuit at the time, see generally Reuel Schiller, Rulemaking’s Promise: Administrative Law and Legal Culture in the 1960s and 1970s, 53 Admin. L. Rev. 1139 (2001). Whether expanding judicial review along these lines actually helped regulatory beneficiaries is a separate empirical question. See, e.g., Wendy Wagner, Revisiting the Impact of Judicial Review of Agency Rulemakings: An Empirical Investigation, 53 Wm. & Mary L. Rev. 1717, 1789 (2012) (concluding that judicial review is less influential than generally expected in the administrative law context); Melissa F. Wasserman, Deference Asymmetries: Distortions in the Evolution of Regulatory Law, 93 Tex. L. Rev. 625, 645–49 (2015) (discussing ways in which deference regimes may favor regulated industry).

In tandem with reformation in the courts, the President and Congress were also taking steps to accommodate themselves to the new reality. A shift in the substantive focus on federal policy from sector-oriented economic regulation (e.g., ratemaking) to economy-wide substantive regulation on matters such as workplace safety or environmental protection encouraged the substitution of informal notice-and-comment rulemaking for case-by-case adjudication and formal rulemaking, which were too cumbersome for the new regulatory program. 134See Paul Verkuil, Judicial Review and Informal Rulemaking, 60 Va. L. Rev. 185, 186 (1974) (“Rulemaking is unique to the administrative process. And properly applied it represents the ultimate form of that process. Even though some adjudication occurs in the administrative decisionmaking model, the role assigned to administrative agencies is in many ways different from that assigned the judicial. Administrative agencies frequently must act prospectively to plan, organize, and set standards. The legislative branch recognizes these functions by accompanying most agencies’ mandates with grants of rulemaking power.”). The new, more flexible style of rulemaking, along with the expanded delegated authority of the new regulatory statutes, placed substantial policymaking power in the agencies. Responding to this fact, as well as the reality that agencies were largely staffed with civil servants committed to the prior regime’s policy program, the Nixon Administration embarked on an effort to exert central influence over the bureaucracy, in part by building out the presidential bureaucracy in the White House. 135See Richard P. Nathan, The Administrative Presidency 34–42 (1983) (describing the Nixon Administration’s increase of staff within the White House); see also Jim Tozzi, OIRA’s Formative Years: The Historical Record of Centralized Regulatory Review Preceding OIRA’s Founding, 63 Admin. L. Rev. 37 (2011) (chronicling the evolution of regulatory review). Congress, responding to the new partisan reality, engaged in the first major overhaul of administrative procedure statutes, adding procedural, analytic, and transparency requirements through legislation such as the Freedom of Information Act, the Federal Advisory Committee Act, and the National Environmental Policy Act. 136 Freedom of Information Act, 5 U.S.C. § 552 (2012); Federal Advisory Committee Act, 5 U.S.C. app. (2012); National Environmental Policy Act, 42 U.S.C. § 4321 (2012).

The election of Ronald Reagan marked a shift from the reformation period of accommodation with the old order to the presidentialization period that constructed the current edifice of administrative law. Reagan expanded on the earlier steps of the Nixon administration to exert control over agencies in a variety of ways. One of the most important changes was the centralization of control over the appointments process. 137 Donald P. Moynihan & Alasdair S. Roberts, The Triumph of Loyalty Over Competence: The Bush Administration and the Exhaustion of the Politicized Presidency, 70 Pub. Admin. Rev. 572, 573 (2010). This centralization helped select appointees who were loyal to the more conservative-oriented policy program of the new administration—it was not enough to be a Republican; the Administration wanted to screen for Reaganites. 138See Thomas J. Weko, The Politicizing Presidency: The White House Personnel Office, 1948–1994, 96–97 (1995) (noting the importance of the screening process and of the appointees’ political beliefs). A second innovation was a substantial expansion of executive review of regulatory proposals. 139 Some form of regulatory review had been in place since Nixon, but Reagan substantially expanded the substantive and institutional strength of this requirement. Office of Mgm’t & Budget, Office of Information and Regulatory Affairs (OIRA) Q&A’s, White House (Nov. 2009), https://obamawhitehouse.archives.gov/omb/OIRA_QsandAs (“The issuance of Presidential regulatory principles, and the centralized review of draft regulations, has been an accepted part of regulatory development for 30 years in one form or another. This began with President Nixon’s ‘Quality of Life’ program, and continued in the 1970s with President Ford’s requirement in Executive Orders 11821 and 11949 for agencies to prepare inflation/economic impact statements and with President Carter’s Executive Order 12044 on ‘Improving Government Regulations.’”). Under Reagan’s executive order 12,291, the Office of Information and Regulatory Affairs (OIRA) was charged with reviewing all new regulations that were issued by executive agencies against a cost-benefit standard as well as the President’s substantive priorities, which included a program of deregulation aimed especially at New Deal industrial planning, but with a secondary focus on the expanded economic regulation program initiated under Nixon.

Congress understood that the Reagan expansion of presidential power would come at the expense of congressional influence over administrative agencies, and it took several steps to resist these changes, including:

hyper-detailed substantive statutory amendments; the imposition of rulemaking deadlines, regulatory hammers, and similar statutory action-forcing mandates; the increase in reports and other oversight data that agencies are required to give Congress; the refusal to confirm nominees to key White House oversight positions; and the use of riders in appropriations bills to regain control over specific agency programs. 140 Cynthia R. Farina, Undoing the New Deal Through the New Presidentialism, 22 Harv. J.L. & Pub. Pol’y 227, 235 (1998).

Conflict over OIRA lasted through the Reagan Administration and eventually led to the refusal of Congress to confirm any administrator to OIRA during the George H.W. Bush Administration. 141See James F. Blumstein, Regulatory Review by the Executive Office of the President: An Overview and Policy Analysis of Current Issues, 51 Duke L.J. 851, 855 (2001).

The gradual replacement of judges from the New Deal era by Nixon’s appointees also led to a rethinking of the role of the courts in overseeing administrative action. 142See generally Patricia M. Wald, The D.C. Circuit: Here and Now, 55 Geo. Wash. L. Rev. 718 (1986) (discussing the changing makeup of the court and its emerging practices and jurisprudence). In Vermont Yankee v. Natural Resources Defense Council, Inc., then-Associate Justice William Rehnquist (a Nixon appointee) sharply upbraided the D.C. Circuit for expanding procedural requirements to the point of the “Kafkaesque.” 143See Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 557 (1978). Although that decision did not immediately lead to a reverse course on the lower courts, 144 Richard J. Pierce, Jr., Waiting for Vermont Yankee II, 57 Admin. L. Rev. 669, 673, 683 (2005) (noting practice of the First Circuit to maintain precedents eroded by Vermont Yankee). the era of ever-increasing judicial management of the administrative process had come to an end. 145See, e.g., Air Transp. Ass’n of Am. v. Fed. Aviation Ass’n, 169 F.3d 1, 7 n.5 (D.C. Cir. 1999) (suggesting that prior ex-parte limitation was undermined by Vermont Yankee); Sierra Club v. Costle, 657 F.2d 298, 402 (D.C. Cir. 1981) (declining to extend prior decision limiting ex-parte communications in agency decision making); In Heckler v. Chaney, the Court found an agency’s enforcement authority to be all but unreviewable, another important protection for presidential prerogative. 146 Heckler v. Chaney, 470 U.S. 821, 831–32 (1985). See generally Cary Coglianese & Christopher S. Yoo, Introduction: The Bounds of Executive Discretion in the Regulatory State, 164 U. Pa. L. Rev. 1587, 1594 (2016) (discussing importance of the “power to defer”). The Court also released other external checks on presidential authority in Immigration and Naturalization Service v. Chadha147 462 U.S. 919, 955–59 (1983). which struck down the legislative veto as unconstitutional while maintaining the broad delegations of authority to agencies in related statutes, a breathtaking transfer of authority to the executive. 148See E. Donald Elliott, INS v. Chadha: The Administrative Constitution, the Constitution, and the Legislative Veto, 1983 Sup. Ct. L. Rev. 125, 126 (1983) (describing the decision in Chadha as a “shock” following the courts’ prior balancing approach to the separation of powers); cf. Richard I. Goldsmith, INS. v. Chadha and the Nondelegation Doctrine: A Speculation, 35 Syracuse L. Rev. 749, 750–51 (1984) (wrongly predicting that the decision in Chadha would lead to subsequent development of the non-delegation doctrine). Although the initial decision shifted power to the executive, dynamically the elimination of the legislative veto may, paradoxically, have led to an increasingly assertive legislature, ultimately canceling out the executive-enhancing nature of the decision. See Edward H. Stiglitz, Unitary Innovations and Political Accountability, 99 Cornell L. Rev. 1133, 1154 (2014) (drawing lessons from history of the legislative veto at the state level).

Two other foundational cases for contemporary administrative law are Motor Vehicles Manufacturers Association v. State Farm and Chevron v. Natural Resources Defense Council, Inc149 Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837 (1984); Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mutual Auto. Ins. Co., 463 U.S. 29 (1983). By way of illustration, State Farm has been cited more than 700 times by the D.C. Circuit alone, while Chevron has nearly 1,400 citations by the Circuit. Much (arguably too much) has been written about these two cases, so our discussion will be brief. Although Chevron marked an important expansion of authority by the executive, it also drew a line, with the Court preserving the function of the judiciary in ensuring the legality of agency action by enforcing unambiguous congressional commands. 150Chevron, 467 U.S. 837. This accommodation to the continued power of Congress, and the institutional function of courts, marked an outer boundary of the President’s new authority. State Farm played a similar role. Although Vermont Yankee was meant to reel in the lower court’s imposition of procedural requirements on agencies, the Court in State Farm reaffirmed the judiciary’s central role in ensuring regularity with the core administrative value of reasoned decision-making.

When Bill Clinton assumed the presidency in 1993, he inherited the administrative law and institutions that had been forged over the past sixty years. Clinton’s election marked a return to unified Democratic control over the political branches. Like Carter before him, Clinton embraced a more centrist platform, placing him somewhat at odds with the older-style Democrats in Congress. 151See Jon F. Hale, The Making of the New Democrats, 110 Pol. Sci. Q. 207, 207 (1995) (“Clinton did not want to be regarded as a Democrat in the line of Michael Dukakis, Walter Mondale, Jimmy Carter, and George McGovern. The New Democrat rhetoric, however, was not simply campaign rhetoric concocted by Clinton’s coterie of talented campaign strategists. It stood for a substantively new Democratic approach to active government that had been developing since the early years of the Reagan administration.”). At the time, it was not clear whether Clinton’s victory presaged a return to Democratic dominance, or would be a Carter-style blip in the politics of divided government that had reigned since 1969. 152Compare Everett Carll Ladd, The 1992 Vote for President Clinton: Another Brittle Mandate?, 108 Pol. Sci. Q. 1, 1–2 (1993) (arguing that the 1992 election was a deviation, but not a realignment, noting that “little happen[ed] in the mix of party identification”), with Robin Toner, 1992 Elections: At Dawn of New Politics, Challenges for Both Parties, N.Y. Times, Nov. 5, 1992, at B1 (citing democratic sources who believed that the Republicans “‘electoral lock’ on the White House was ‘broken forever’”). The reality that Clinton’s election would usher in a new stage of wildly swinging party control and increasingly deep ideological divides was much less anticipated. As Clinton and his successors in the presidency, as well as courts and Congress, responded to the new reality, they left much of the administrative apparatus in place, turning doctrine and institutions to new ends. This syncretic approach functioned reasonably well for a time, creating channels that translated electoral preferences into policy outcomes while protecting administrative values of stability and expertise. But, as we will discuss in the following Part, this system of turning the old system to new purposes may have reached a breaking point.

II. The Challenges of Partisan Volatility

The current party system has two dominant features: polarization and volatility. The first feature leads to party coalitions that are further away from one another (and the median voter). As the parties drift further apart, opportunities to act in periods of divided government shrink. Executive actions replace legislation as the means of moving a policy program forward and Congress shifts its focus toward oversight and investigations. Current parties are more decentralized, with less of a role for party leadership. As a result, polarization leads to both more extreme outcomes and more isolated centers of decision-making, as issue advocates are given the reins to manage different areas of policy. The second feature, volatility, leads to frequent and consistent swings in control of government. Periods of unified government quickly yield to long stretches of divided government, followed by electoral success for the out party in the next open presidential election. This predictable pattern incentivizes obstruction over cooperation, as the party opposed to the executive can simply wait for a future election to enact its agenda. Taken together, these two characteristics lead to swings between radically different policy positions. The result is a political system where policy reversals are common, Congress is unresponsive, and decision-making is less centralized.

This partisan volatility threatens effective administration. Good public policy takes many years to develop and implement. A project started in one administration may not produce results until the next. When the electoral results were stable and polarization less extreme—as in the period from 1968 to 1992—bipartisan policy had a chance of weathering intervening elections. The current system instead creates a need for more short-term policy implementation, as signature pieces of legislation do not have bipartisan support at the time they are passed. Even if policies are not reversed outright, they are often weakened or ignored by the other party. For regulated communities, this back and forth results in higher costs, as the investments made to satisfy one regulatory regime are disregarded by the next. The constant concerns about the health of the Affordable Care Act (ACA) exchanges under Republican leadership are one salient example of a more pervasive problem. 153See, e.g., Linda J. Blumberg et al., How Would Coverage, Federal Spending, and Private Premiums Change if the Federal Government Stopped Reimbursing Insurers for the ACA’s Cost-Sharing Reductions? 1–3 (2017) (discussing the implications of the Trump Administration’s decision regarding reimbursements for ACA exchange insurers, one of the many issues over which the new Republican administration departed from the prior administration in its implementation of the ACA).

Many institutions of administrative law have the potential to limit these costs, either through promoting less extreme policy in the first instance or by stabilizing implementation once a policy is adopted. The strength of these institutions is all the more important as polarization and volatility become more severe. In the Executive Branch, OIRA has historically helped to both centralize the regulatory agenda and moderate policy. In Congress, agencies like the Congressional Budget Office (CBO), the Congressional Research Service (CRS), and the Government Accountability Office (GAO), staffed with career civil servants, inform the policy process through nonpartisan research and consistent methodology. Finally, the courts themselves have played a role in stabilizing policy through the development of doctrines to ensure procedural rigor and substantive rationality in agency actions. Rather than remaining insulated from political pressure, however, these institutions are threatened by the current system. In some cases, the same external forces that created increased partisan polarization have reached the institutions designed to resist them, such as the courts. For other actors, such as OIRA and CBO, their role has been reduced and their input dismissed when they are seen as an obstacle to partisan policy.

This Part discusses the current party system and its effect on administrative institutions. Sections II.A and II.B describe the political conditions of extreme polarization and volatility, respectively. Section II.C examines the effect of these political forces on administrative institutions, finding that they threaten the ability of stabilizing institutions to maintain reasoned decision-making.

A. Partisanship Without Parties

The era of partisan volatility began with the election of Bill Clinton in 1992. 154Shafer, supra note 13, at 121. This election led to unified Democratic control of the federal government. This dominance was short-lived, however, with Republicans taking control of Congress in the wave election of 1994. This led to six years of divided government, followed by unified Republican control in 2000 under President George W. Bush. Much like his predecessor, however, divided government followed, this time with a Republican President and a Democratic Congress. In just sixteen years, the American electorate had selected different parties to control the machinery of government in every possible combination. The result was “a rapid succession of all four partisan possibilities” that “had not been seen in American politics since the 1840s.” 155Id. at 122. These trends continued during the presidency of Barack Obama, who entered office with substantial majorities, only to see the House change hands two years later and the Senate change hands during his second term. Like his two predecessors, he also left office to a President and Congress of the opposing party. For President Trump, the story has been the same, with control of the House flipping in the first midterm. 1562018 Midterm Election Results, N.Y. Times (Nov. 27, 2018), https://www.nytimes.com/interactive/2018/us/elections/calendar-primary-results.html. Rather than creating a party system characterized by stable actors, like the party systems that came before it, the defining feature of this era is change itself, what Byron Shafer calls “shifting mixes of absolutely everything.” 157Shafer, supra note 13, at 122.

 

Changes in party control of government are only half of the partisanship story. The other is the increased ideological distance between the parties themselves. Many observers have noted the increasingly polarized Congress, as the moderating forces of Northeastern Republicans and Southern Democrats collapsed. 158See, e.g., Robin Toner, Southern Democrats’ Decline is Eroding the Political Center, N.Y. Times (Nov. 15, 2004), https://www.nytimes.com/2004/11/15/politics/southern-democrats-decline-is-eroding-the-political-center.html. Moreover, party activists, once kept in check by more moderate party elites, began to take control of elected officials, further exacerbating the divide. 159Shafer, supra note 13, at 141 (“[I]n … the Era of Partisan Volatility, the ideological dynamic that had captured the activist stratum in both parties broke through the previous firewall, capturing elected officialdom and fostering generalized polarization.”). As this extremism came to define Congress itself, major legislation (now adopted almost exclusively during united government) has passed without bipartisan support. This trend has worsened since 1992, with the “truly amazing incarnations of an exaggerated partisan outline” during the Obama years. 160Id. at 159. For instance, the ACA received zero Republican votes in either chamber, while the Trump Administration’s signature tax legislation received no Democratic votes. Since the start of the Bush Administration, instances of presidents aligning with the opposition party have been rare, and often in the face of financial crisis, wherein both actors had to come to the table, such as during the financial crisis of 2008 or the fiscal cliffs of 2011 and 2012. 161Id. at 160–61; see American Gridlock: The Sources, Character, and Impact of Political Polarization xxii (James E. Thurber & Antoine Yoshinaka eds., 2015) (listing as consensus views the proposition that “[t]he parties in Congress are as polarized … as at any time in history,” “[t]he fit between ideology and party is unusually strong,” and “[u]nder divided government and split chamber control, the current Congress has ceased to operate as an effective legislative body.”). The pieces in this collection go on to explore particular factors reinforcing this polarization, like changes to the media and political redistricting. See, e.g., Micah Altman & Michael McDonald, Redistricting and Polarization, in American Gridlock: The Sources, Character, and Impact of Political Polarization 45 (James E. Thurber & Antoine Yoshinaka eds., 2015).

A shift in parties themselves has accompanied the shift in partisanship. Political parties function as both an identity for voters and an independent organization, managed by its elite members. 162See Hershey, supra note 43, at 6; see also Tashjian v. Republican Party of Conn., 479 U.S. 208, 224–25 (1986) (recognizing associational rights of political parties as private associations); Aldrich, supra note 43, at 4–6 (adopting a more elite-driven and insider-driven understanding of political parties; Donald Green et al., Partisan Hearts and Minds: Political Parties and the Social Identities of Voters 4 (2002) (“To be sure, party issue positions have something to do with the attractiveness of partisan labels …. But causality also flows in the other direction: When people feel a sense of belonging to a given social group, they absorb the doctrinal positions the group advocates.” While partisanship affiliations have grown more prominent as an electoral force, the role of the parties as a system of organization and discipline has declined. 163See Julia Azari, Weak Parties and Strong Partisanship are a Bad Combination, Vox (Nov. 3, 2016, 4:40 PM), https://www.vox.com/mischiefs-of-faction/2016/11/3/13512362/weak-parties-strong-partisanship-bad-combination. For much of the twentieth century, party elites controlled the party nomination process, especially at the national level. 164Marty Cohen et al., The Party Decides: Presidential Nominations Before and After Reform 7 (2008). Despite reforms that attempted to wrest control from insiders and broaden the nomination process, highly involved party activists were able to work within the party system to both direct campaign resources and control messaging, therefore guaranteeing the success of insider candidates. 165Id. This work defines political parties in a broad way, arguing that the highly involved activists function as a single political party, even without formal affiliations with the national organization. As recently as the 2000 election, insider favorites were able to defeat outside candidates, even those with popular positions and biographies. 166Id. at 5–6. When “the party decides,” the real presidential contest is not for primary votes, but instead for the high profile endorsements that dominate the “invisible primary.” 167See id. at 5, 187–88.

If the nomination of Barack Obama over Hillary Clinton suggested a deviation from the model, the nomination of Donald Trump as the Republican nominee eight years later was a full-scale repudiation. Trump is the only President since at least 1980 to have fewer endorsements than another primary candidate and win the nomination. 168See Harry Enten, The Four Things I Learned from the Donald Trump Primary, FiveThirtyEight (May 6, 2016, 4:55 PM), https://fivethirtyeight.com/features/the-four-things-i-learned-from-the-donald-trump-primary/. While it is too soon to know whether this is just another aberration, the inability of Republican elites to control the nomination process has coincided with other forces that suggest a declining role for party elites. A variety of legal changes, as well as changes in the practices of political donors, have reduced the influence of the official party apparatus. 169 Samuel Issacharoff, Outsourcing Politics: The Hostile Takeover of Our Hollowed-Out Political Parties, 54 Hous. L. Rev. 845, 862–64 (2017) (summarizing the Court’s decisions in FEC v. Colo. Republican Fed. Campaign Comm. (Colorado Republican II), 533 U.S. 431, 455 (2001) and Colo. Republican Fed. Campaign Comm. v. FEC (Colorado Republican I), 518 U.S. 604, 614 (1996)). At least anecdotally, this has contributed to a string of upsets in party nominations in recent years on the Republican side 170 Examples include Christine O’Donnell, Ken Buck, Roy Moore, and Todd Akin. and conflicts between donors and party leaders. 171See, e.g., Edward-Isaac Dovere, Why Tom Steyer Doesn’t Care What Nancy Pelosi Thinks, Politico (May 29, 2018), https://www.politico.com/magazine/story/2018/05/29/trump-impeachment-democrats-steyer-pelosi-218548.

These two trends produce a paradoxical result. Ideological separation is more pronounced than ever, while the parties themselves retain less control. The interaction between strong partisans and weak parties has gained attention following the 2016 election. 172See Julia Azari, Weak Parties and Strong Partisanship Are a Bad Combination, Vox (Nov. 3, 2016, 4:40 PM), https://www.vox.com/mischiefs-of-faction/2016/11/3/13512362/weak-parties-strong-partisanship-bad-combination. The critiques have focused on how these two features of the current political climate change the policymaking process, noting the declining importance of political norms. 173See id. (focusing on the collapse of political norms that results from weak institutional parties); see also Issacharoff, supra note 169, at 879–80. These same features also have an impact on administration. When parties are operating effectively, they help aggregate policy preferences, determine priorities, and push a unified policy agenda. 174Hershey, supra note 43, at 11–12. See generally Gary W. Cox & Mathew D. McCubbins, Legislative Leviathan: Party Government in the House (2d ed. 2007). For the White House, these same forces lead to centralized decision-making. As the parties have declined in their ability to exercise these functions, the traditional tools of managing the administrative state, both in the Legislative and the Executive Branches, have changed with it.

As discussed above, the era of divided government was characterized by the exertion of greater control over regulatory policy by Republican Presidents, including through the establishment and growth of OIRA. OIRA review procedures served a gatekeeping function for federal regulation, helping to ensure substantive outcomes favored by the Republican-dominated White House. 175Richard L. Revesz & Michael A. Livermore, Retaking Rationality: How Cost-Benefit Analysis Can Better Protect the Environment and Our Health 21–30 (2008). In the agencies themselves, personnel selection was heavily influenced by partisan affiliation, leading to posts filled by Republican Party loyalists who reflected the priorities of the national organization. 176See, e.g., Devins & Lewis, supra note 36, at 480–83 (describing the growth in the nomination of party loyalists to independent agency positions since the 1980s).

During much of the era of partisan volatility, presidential control of the administrative state accelerated. The structures that facilitated greater control by Republican presidents during the era of divided government were relatively easily repurposed for use by both Democratic and Republican presidents in the early years of partisan volatility. 177See Kagan, supra note 3, at 2249–50. In addition, like their predecessors during the era of divided government, the Clinton, George W. Bush, and Obama Administrations used both personnel decisions and regulatory review in ways that broadly promoted the consensus agenda of the respective parties. 178See Livermore, supra note 3, at 73–80. This reflected the posture of each President as broadly representative of elites within their respective parties. Even President Obama, whose 2008 Democratic primary victory upset longstanding norms of elite party control over the nomination process, nevertheless governed as a party moderate, relying heavily on appointees with deep connections to the party establishment, and adopting a policy agenda that fell well within the norms of the Democratic Party. 179Id. at 91–92.

The current Administration has struck a different path, although it is too soon to know whether this is part of a larger trend. Two important differences stand out. The first is the lack of a coherent policy agenda that reflects the consensus view of party elites. 180Cf. Lindsay McPherson, 2018 Republican Agenda Not What Lawmakers Envisioned, Roll Call (Feb. 5, 2018, 5:04 AM), https://www.rollcall.com/news/politics/2018-republican-agenda-not-lawmakers-initially-envisioned (noting the role of intraparty disputes in limiting the legislative agenda). Trump’s political appeal was built on a record outside of traditional politics, and he did not develop particularly robust policy proposals during the campaign. The policy platform he did enunciate deviated from preexisting Republican orthodoxy in many areas, most markedly on trade and national security. 181 Max Boot, Trump Security Strategy a Study in Contrasts, CFR (Dec. 18, 2017), https://www.cfr.org/expert-brief/trump-security-strategy-study-contrasts. On immigration, arguably President Trump’s signature issue, his policy views took a harder line than the one favored by Republican Party elites. It may be that this departure from the party establishment fits into a pattern where party elites have declining control over American politics.

The second difference is the breakdown of the coordinating and moderating influence of the White House in favor of diffusion of authority to the political leadership at agencies. 182 The diffusion of authority to political appointees at agencies has been accompanied by a sidelining of civil service experts. Daniel A. Farber, Presidential Administration Under Trump 3–4 (Aug. 8, 2017) (unpublished research paper) (on file with U.C. Berkeley School of Law), https://ssrn.com/abstract=3015591 (“[R]ather than deferring to agency experts, the Administration has often cut them out of the loop and has shown itself hostile in important ways to traditional forms of expertise.”) The current Administration has had a great deal of difficulty identifying competent long-term appointees for key White House posts, and has been continually plagued by low-grade crises and shifts in leadership. 183 See Kathryn Dunn Tenpas, Record-Setting White House Staff Turnover Continues with News of Counsel’s Departure, Brookings Inst. (Oct. 19, 2018), https://www.brookings.edu/blog/fixgov/2018/10/19/record-setting-white-house-turnover-continues-with-news-of-counsels-departure/) (“Much like turnover within the larger sample of senior White House staff, President Trump is breaking records.”); Kathryn Dunn Tenpas, With the Revelation of Marc Short’s Impending Departure, President Trump Has Lost the Vast Majority of Tier 1 Staff Members, Brookings Inst. (June 27, 2018), https://www.brookings.edu/blog/fixgov/2018 /06/27/trump-has-lost-the-vast-majority-of-tier-one-staff-members/ (describing record levels of turnover in the Trump Administration for the twelve most senior positions within the White House). With less high-level coordination over policy priorities, the political leadership of agencies have space to follow more idiosyncratic or ideologically extreme agendas. For instance, the members of the Republican coalition with a high level of interest in education policy pushed for Betsy DeVos’s nomination to the Department of Education, despite her lacking comparable experience to her predecessors. 184 Amber Phillips, That Confirmation Hearing, and Three Other Reasons Betsy DeVos’s Nomination Is on the Rocks, Wash. Post (Feb. 2, 2017), https://www.washingtonpost.com/news/the-fix/wp/2017/02/02/that-confirmation-hearing-and-3-other-reasons-betsy-devoss-nomination-is-on-the-rocks/?utm_term=.ae066a6face9. Similar stories could be told regarding Scott Pruitt’s nomination at the EPA or Ryan Zinke’s leadership of the Department of the Interior. 185 For a description of how decentralization can lead to greater ideological separation in the context of congressional oversight, see Broker, supra note 39, at 270–73. Judicial appointments have seen a similar trend, with outside groups playing a prominent role in Trump’s court selections. 186 Leonard Leo, the Executive Vice President of the Federalist Society, served as President Trump’s judicial adviser during both Supreme Court confirmations. See Jay Michaelson, The Secrets of Leonard Leo, the Man Behind Trump’s Supreme Court Pick, Daily Beast (July 9, 2018, 5:16 AM), https://www.thedailybeast.com/the-secrets-of-leonard-leo-the-man-behind-trumps-supreme-court-pick. Without a moderating central influence from the White House, agency agendas have become more responsive to the most engaged (and likely extreme) voices within the party coalition. 187 Farber, supra note 182, at 19.

Even when coordinated through powerful and coherent signals from the White House, as they were during the Clinton, Bush, and Obama Administrations, the dynamics of partisan volatility favored more aggressive use of regulatory authority to achieve partisan policy goals. These same forces, coupled with some features of the current Administration—including high turnover of senior White House staff, the lack of a coordinated policy program, greater skepticism towards agency expertise, and less reliance on traditional experience in selecting personnel 188Id. at 28.—has unmoored decision-making from stabilizing influences. If successful outsider candidates with few connections to party institutions and little allegiance to party programs and governance practices become the norm, then regulatory policymaking is likely to become increasingly extreme.

B. Oscillating Control

Taken in isolation, a more polarized form of politics might not be destabilizing on its own. If parties have unified and cohesive ideological preferences, then pursuing these goals as a single party can lead to effective results, so long as that party remains dominant. 189See generally Aldrich, supra note 43. In a system like that of the New Deal era, characterized by single-party dominance, increasing polarization would not necessarily lead to uncertainty. The outcomes would be predictable, even if extreme. As outlined above, however, the modern era is instead characterized by frequent partisan change. 190See supra Part II. While the periods of divided government produce very little legislation, the short waves of unified government produce partisan legislation. 191Frances E. Lee, Insecure Majorities: Congress and the Perpetual Campaign 1 (2016). Because of the polarized political climate, these results are extreme and often directly contradict the policy of the prior administration. 192See generally id. at 1–3 (arguing that the electoral parity between the two major parties contributes to gridlock because each party has a plausible chance of winning the next election, and accordingly has less of an incentive to cooperate).

This cycle of change structures the incentives of political actors at each step, leading to different outcomes even when political conditions are the same as in prior eras. For instance, divided government has not always produced inaction. From the time of Woodrow Wilson, many students of American politics had assumed that an opposition Congress leads to less new legislation and more antagonistic oversight. 193See, e.g., James L. Sundquist, Needed: A Political Theory for the New Era of Coalition Government in the United States, 103 Pol. Sci. Q. 613, 617 (1988) (“As soon as political science emerged as a scholarly discipline, its adherents began to pronounce and elaborate the theoretical foundation of the system of party government…. [T]he scholars concluded that [parties] were useful and necessary … in unifying a government of dispersed powers and thereby making it more effective.”). Professor Mayhew’s study of the post-World War II Congress persuasively challenged these assumptions. 194Mayhew, supra note 51, at 3–5. Examining congressional activity from 1946 to 1990, his work found that congressional productivity was not demonstrably higher during periods of unified government. 195Id. at 79–80. Major policy changes like the expansion of social security or the growth of the regulatory state 196Id. at 78, 85–87. were enacted during periods of divided government. Examining this data, he concluded that divided government could be effective. 197Id. at 3-4.

Before 1992, this may have been right. 198See Lee, supra note 191, at 1–2, 5 (describing the differing incentives for cooperation during the era of divided government and the current era). The benefits of divided government, however, rely on some degree of political stability. In the years examined by Professor Mayhew, congressional Democrats could expect both a stable period as a congressional majority and a high likelihood of Republican presidential success. At the time, these conditions were atypical: “In the fifty-eight years from 1897 through 1954, the country experienced divided government during only eight years—all in the last half of the presidential term—or 14 percent of the time.” 199 Sundquist, supra note 193, at 613. The electorate during this period, moreover, had stable preferences. When President Bush was elected in 1988, it marked the sixth time in eight presidential elections that the voters had simultaneously chosen a Republican President and a Democratic Congress. 200Id. This stability created an incentive for both sides to play the hand they were dealt. Democrats knew that advancement in Congress required working with Republican presidents, and these presidents knew that delivering on their agenda would require a measure of bipartisanship. 201Id. at 627. By contrast, congressional opponents today can simply wait for a new election, where the deck will be reshuffled. The result is that divided government, while structurally the same as ever, leads to drastically different results in the two eras. Today, divided government means legislative stagnation, therefore putting more pressure on the executive to act unilaterally through the administrative state.

Congressional action since 1992 bears this out. While there were some bipartisan accomplishments in the Clinton and Bush presidencies, they have become less common. Most of the major legislation passed in recent years—such as the Affordable Care Act, the Dodd-Frank Act, and the Tax Cuts and Jobs Act—was passed with substantial support in only a single party. 202 Alex Pareene, Political Power Never Lasts. Democrats Need To Use Theirs While They Have It, Wash. Post (Nov. 9, 2018), https://www.washingtonpost.com/outlook/political-power-never-lasts-democrats-need-to-use-theirs-while-they-have-it/2018/11/09/f5f050d2-e396-11e8-8f5f-a55347f48762_story.html?noredirect=on&utm_term=.753dbf326258 (summarizing legislation passed largely by Democrats with unified government in 2009 and 2010). As a result, it is enacted during the period of unified government at the start of each new presidency. 203 Even during periods of unified government, legislation may still be thwarted by the increasingly aggressive use of the filibuster. The willingness to use the filibuster to block legislation has grown over time. As a result, major legislation has often been passed through “reconciliation bills,” which only require a simple majority. Both the Affordable Care Act and the Tax Cuts and Jobs Act were passed this way. These reconciliation bills, however, are limited to budget-related legislation and are capped at three per year. See generally Megan S. Lynch & James V. Saturno, Cong. Research Serv., R44058, The Budget-Reconciliation Process: Stages of Consideration (2017). When legislation is bipartisan, it is usually the result of a crisis. Given this dearth of vehicles to move policy—and the looming threat of the filibuster in the Senate—most congressional action during this period has taken the form of “omnibus legislation,” which “addresses numerous and not necessarily related subjects.” 204Barbara Sinclair, Unorthodox Policymaking: New Legislative Processes in the U.S. Congress 64 (1997). Under these conditions, policymaking is tied to “must-pass” budget legislation, often facilitated by procedural mechanisms like the reconciliation process. 205Shafer, supra note 13, at 163–66. Since any new signature legislation that is passed lacks bipartisan support, it is attacked as soon as the opposition takes control of any of the political branches. For example, the implementation phase of the ACA exposed “unanticipated difficulties in the statutory design” but “congressional antipathy to health reform has precluded looking to the legislature to iron out those difficulties,” leading President Obama to “repeatedly test[] the limits of executive authority in implementing the [law].” 206 Nicholas Bagley, Legal Limits and the Implementation of the Affordable Care Act, 164 U. Pa. L. Rev. 1715, 1716 (2016). After the 2016 presidential election placed unified power in Republican hands, much of the 2017 legislative calendar was spent on an effort to repeal the ACA. 207 Pareene, supra note 202. While this attempt was ultimately unsuccessful, failing by one vote in the Senate, 208 Ryan Lizza, Why John McCain Killed Obamacare Repeal—Again, New Yorker (Sept. 22, 2017). Notably, no Democrat supported the Obamacare repeal efforts, even though many represented states or districts that lean heavily Republican. This highly partisan result further supports the recent trends for major legislation during periods of unified government. Supra notes 191–192 and accompanying text. the prominence of ACA repeal on the policy agenda reflected the partisan nature of its passage in the first instance.

While frequent electoral swings incentivize Congress to obstruct and oppose compromise, the President turns to unilateral action for policy progress during divided government. Policymaking comes in the form of new regulations, enforcement priorities, executive actions, and presidential agreements, rather than bipartisan legislation. Some of these efforts come in the final days of the administration as so-called “midnight rules,” just before the Executive Branch is turned over to the other political party. 209See Jack Beermann, Midnight Rules: A Reform Agenda, 2 Mich. J. Envtl. & Admin. L. 285, 333–34 (2013). Not surprisingly, the study of midnight rules began to gain force during the Clinton Administration, which was the first presidency of the era of partisan volatility. Id. at 291. While midnight rules are frequently critiqued, they may not be all that different in kind from rules promulgated during other periods in a President’s term, and therefore do not get significant focus here. See generally Edward H. Stiglitz, Unaccountable Midnight Rulemaking?: A Normatively Informative Assessment, 17 N.Y.U. J. Legis. & Pub. Pol’y 137 (2014) (empirically examining the practice of midnight rulemaking). For instance, Obama’s second-term agenda was almost entirely characterized by executive actions in an area of policy where legislation had previously failed. 210See Nancy LeTourneau, Democrats Should Start Talking About Comprehensive Immigration Reform, Wash. Monthly (June 1, 2018) (“When it became clear that their attempts would be successful in blocking comprehensive immigration reform, President Obama signed DAPA, granting deferred action to parents of children who were citizens or lawful permanent residents.”); Editorial Board, Don’t Like EPA’s Power-Plant Plan? Complain to Congress, Bloomberg (Sept. 29, 2016), https://www.bloomberg.com/opinion/articles/2016-09-29/don-t-like-epa-s-power-plant-plan-complain-to-congress (“Congress left the Obama administration no choice. The only way to avert potentially disastrous climate change is to lower emissions as soon as possible. Congress’s refusal since 2010 to consider any climate policy at all forced the EPA to look for alternatives.”).

The administrative state has felt the weight of these strong policy swings. Once a wave election has ushered in unified government, the new President immediately works to undo or undermine the accomplishments of the prior administration. 211Frances E. Lee, Insecure Majorities: Congress and the Perpetual Campaign 4 (2016). This can involve taking action in an area where the prior administration had decided to remain inactive, such as the Obama Administration’s action on climate change, or attempts to reverse or alter existing regulations. By way of example, the first year of the Trump presidency involved both legislative and executive attempts to unwind much of Obama-era policy. On the executive side, new Trump appointees quickly went to work reversing net neutrality regulations, 212 Restoring Internet Freedom, 83 Fed. Reg. 7852, 7852 (Apr. 23, 2018). ending the Deferred Action for Child Arrivals program (DACA), 213 Memorandum from Elaine C. Duke, Acting Sec’y, U.S. Dep’t of Homeland Sec., to James W. McCament, Acting Dir., U.S. Citizenship & Immigration Servs., et al. (Sept. 5, 2017) (on file with U.S. Dep’t of Homeland Sec.). withdrawing from the Paris Climate Accords and terminating parts of the Clean Power Plan, 214 Repeal of Carbon Pollution Emissions Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 82 Fed. Reg. 51787 (proposed Nov. 8, 2017). and removing national monument designations for Western lands. 215 Julie Terkowitz, Trump Slashes Size of Bears Ears and Grand Staircase Monuments, N.Y. Times (Dec. 4, 2017), https://www.nytimes.com/2017/12/04/us/trump-bears-ears.html. For all of these (except net neutrality), the policy change came either from the White House directly or from an executive agency subject to greater presidential control. The change to net neutrality originated from an independent agency, the FCC, under the leadership of Trump’s appointed Chairman, Ajit Pai. 216 Tony Romm, Here’s How the FCC Plans To Defend Its Net Neutrality Repeal in Federal Court, Wash. Post (Oct. 11, 2018), https://www.washingtonpost.com/technology/2018/10/12/heres-how-fcc-plans-defend-its-net-neutrality-repeal-federal-court/?utm_term=.7bf9402a311e. Once periods of divided government set in, the agencies become the forum for any new policy initiative and existing legal authorities are stretched to meet new needs.

In Congress, the regulations of the prior administration are also threatened by a newly invigorated Congressional Review Act (CRA). The CRA, passed in 1996, provides Congress with a window to undo “major regulations” of executive agencies following final agency action. 217Maeve P. Carey et al., Cong. Research Serv., R43992, The Congressional Review Act: Frequently Asked Questions 1 (2016). The CRA governs legislative procedure only; it is unlike the general management laws imposed on the executive to impose procedural obligations on agencies directly. If Congress decides to take up a resolution disapproving of the regulation, the CRA ensures expedited procedures, which do not allow for a Senate filibuster. 218Id. at 14. To have legal effect, CRA resolutions must go through bicameralism and presentment like any other law. As a result, the statute has little import during periods of divided government when the President could simply veto the resolution. The law becomes a very important tool, however, when two conditions are present: first, the same party controls both Congress and the presidency; second, the regulations of an opposition party president were promulgated recently enough to be within the CRA window. 219Id. at 12–13. Holdover appointees can create a context where Congress and the President can use the CRA effectively well into a new period of unified government. The CRA has only ever been used when both conditions were present. While both conditions were rare during prior party systems (occurring only once in the period of divided government, in 1977), they are an enduring feature of the current era, occurring in 1992, 2000, 2008, and 2016. As such, the CRA has become an important part of modern administrative governance. Prior to 2016, the statute had only been used once, when the Republican Congress passed a resolution disapproving a Clinton-era ergonomics regulation. 220Id. at 5. The current Congress has been much more aggressive, using the CRA to reverse environmental, 221 Stream Protection Rule, Pub. L. No. 115-5, 131 Stat. 10 (2017) (disapproving the Department of the Interior’s Stream Protection Rule). education, 222 Department of Education—Teacher Preparation Issues, Pub. L. No. 115-14, 131 Stat. 78 (2017) (disapproving the Department of Education’s teacher preparation rules). and labor regulations. 223 Clarification of Employer’s Continuing Obligation, Pub. L. No. 115-21, 131 Stat. 87 (2017) (disapproving the Department of Labor’s regulation of reporting employee injury and illness). Apart from the CRA, modern Congresses operating under divided government also have greater incentives to obstruct executive branch appointments, leading to uncertain and harmful consequences. 224See Nina A. Mendelson, The Uncertain Effects of Senate Confirmation Delays in the Agencies, 64 Duke L.J. 1571, 1576, 1603–04 (2015).

The combination of these frequent policy reversals and greater ideological polarization in agency leadership is destabilizing. The current party system incentivizes capitalizing on periods of unified government to push partisan political preferences with little prospect of achieving a bipartisan compromise. During periods of divided government, the President is more likely to achieve policy results through executive action than legislative compromise. When control of government changes hands (as it has consistently throughout the period), the new President is tasked with implementing laws and regulations that have no support from within his own party. When the old policies are replaced, the new regulation often pursues completely different objectives. This sudden reversal only lasts as long as the current party maintains control, at which point the policy swings back. This constant back and forth is a recipe for regulatory uncertainty, high compliance costs, and ineffective programs.

The example of climate change policy brings these costs into sharp relief. The Obama Administration finalized the Clean Power Plan (CPP) in October of 2015. As part of this program, the EPA developed state-specific plans with emission performance targets through 2030. 225 Carbon Pollution Emissions Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 80 Fed. Reg. 64,661, 64,664 (Oct. 23, 2015) (to be codified at 40 C.F.R. pt. 60). Interim performance targets started in 2022, seven years after the promulgation of the rule. The CPP was an ambitious plan to tackle the growing problems of anthropogenic carbon emissions, involving major obligations for state governments and the private sector. The costs and benefits of both totaled many billions of dollars. 226Id. at 64,679–82. Implementation of the rule was expected to take decades, with benefits only exceeding costs once the emissions targets went into effect. In a stable political environment, this policy design may be successful. Such stability could come from any one of (1) sustained single party dominance, (2) decreased polarization such that policy swings are less extreme, or (3) incentives for bipartisan compromise. The current political environment lacks any of these features.

The CPP demonstrates what happens when these foundations disappear. The CPP was only put into place after many attempts to pass legislation failed. It was heavily opposed by the Republican Congress at the time, many of whom challenged both the legal authority as well as the underlying policy goals. Lawsuits contesting the program began immediately. When President Obama left office, a new Republican Congress and administration immediately went to work repealing the regulation. The costs of this ambitious regulation and sudden about-face are clear. States and private actors began to incur compliance costs, but the repeal undermined the climate benefits of those investments. This repeal, however, is just as unstable as the initial policy, such that new compliance costs may emerge again soon. Faced with this political climate, the affected industry cannot rely on a stable regulatory environment, nor can nascent firms in the energy sector, such as solar or wind power developers, rely on favorable policy.

While the CPP is a salient and potent example of the inefficiencies created by oscillating control over regulatory decision making, it is hardly unique. Similar costs are imposed in any area where agencies swing between such extreme policy positions. Health insurers are currently raising premiums to respond to drastic changes between the policies of HHS during the Trump and Obama Administrations. 227See Blumberg et al., supra note 153. Higher education is adjusting to drastic changes in the rules related to for-profit universities. 228 Michael Stratford, Trump and DeVos Fuel a For-Profit College Comeback, Politico (Aug. 31, 2017, 4:58 AM), https://www.politico.com/story/2017/08/31/devos-trump-forprofit-college-education-242193. A similar situation exists for other environmental rollbacks, some of which have been opposed by the regulated industry they purport to help.

This threat is not unique to administrative regulation. The Republican tax legislation will also create incentives that may be unwound within a short period. These costs, however, do pose a unique threat to the legitimacy of agency actions, which continue to rely on technical expertise to justify their authority. Legislative outcomes are not (for better or worse) necessarily expected to proceed along a rational or scientific path. At one level, these experiences could simply be seen as the price of political responsiveness. Administrative regulation, however, is not the legislative process. It is governed by a legal regime premised on a rational relationship to an enduring statutory regime. Agency regulations are by design more incremental, accounting for uncertainty and ensuring that the benefits of a change outweigh the costs. 229 Michael A. Livermore, Cost-Benefit Analysis and Agency Independence, 81 U. Chi. L. Rev. 609, 612 (2014). This is the result of both internal executive branch guidance, whether in OIRA or the agencies themselves, and the modern doctrine of arbitrary and capricious review. 230Id. at 614; Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 844 (1984) (“Such legislative regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to statute.”). At present, political forces threaten this model, adding costs that are not currently part of the analysis. These costs are as real as others, such as health risks and compliance costs, that agencies are already bound to consider.

C. Erosion of Stabilizing Institutions

Many institutions in administrative law ensure that agency decision-making is supported by more than sheer political pressures. While the notion that administration can be exercised wholly independent from politics has disappeared, 231See supra note 1 and accompanying text. administrative law remains structured to ensure some degree of rationality in the policy process, even when politics is inevitably present. Institutions in all three branches of government serve this function, and therefore are most important in times of high volatility. Unfortunately, many of these institutions are threatened, either through being captured by the broader partisan dynamics or through marginalization by political actors. 232See infra note 238.

In the Executive Branch, OIRA is responsible for both ensuring consistent regulatory approaches and for considering the effects of regulations enacted across the government. 233See Exec. Order No. 12,291, 3 C.F.R. § 127 (1981), reprinted in 5 U.S.C. § 601 (1988), revoked by Exec. Order No. 12,866, 3 C.F.R. § 638 (1993), reprinted in 5 U.S.C. § 601 (2000). In the era of divided government, these features enabled OIRA to promote the Administration’s favored policy program in the face of opposition from Congress and, to some degree, agencies themselves. 234See generally Richard L. Revesz & Michael A. Livermore, Retaking Rationality: How Cost-Benefit Analysis Can Better Protect the Environment and Our Health 21–30 (2008) (discussing use of cost-benefit analysis during the Reagan and George H. W. Bush Administrations). The cost-benefit analysis standard also provided a means for agencies to use their relative substantive sophistication to protect their independence. See Livermore, supra note 229, at 613. After his election, President Clinton retained OIRA and cost-benefit analysis, despite objections from some in his party, 235Revesz & Livermore, supra note 234, at 31–32. and during the era of partisan volatility (at least prior to the Trump Administration), OIRA served a stabilizing function. 236 Michael A. Livermore & Richard L. Revesz, Regulatory Review, Capture, and Agency Inaction, 101 Geo. L.J. 1337, 1339 (2013). Several features of OIRA facilitate this role: the fact that it is a generalist organization, and therefore difficult for any particular special interest group to capture; the institution’s coordination function, which naturally exposes it to multiple constituencies within an administration; the use of cost-benefit analysis, which is comprehensive in the interests that it considers; and the appointment of relatively independent OIRA Administrators. 237Id. at 1361–62. OIRA’s stabilizing role naturally generates criticism when it dials back activities that might otherwise be pursued at the agency level. See generally Lisa Heinzerling, Statutory Interpretation in the Era of OIRA, 33 Fordham Urb. L.J. 1097, 1117 (2006) (arguing that OIRA has usurped the appropriate role of agencies in interpreting statutes).

Although OIRA’s generalist perspective and coordinating function may help maintain some semblance of its stabilizing function, indicators during the Trump Administration do not bode well for the methodology of cost-benefit analysis or the relative independence of OIRA’s Administrators. On the cost-benefit front, the Trump Administration has been quite aggressive in manipulating cost-benefit analyses to serve prearranged goals and has made ill-considered and unwise changes to cost-benefit analysis methods and the process of regulatory review. 238See Jason Schwarz & Jeffrey Shrader, Muddying the Waters: How the Trump Administration Is Obscuring the Value of Wetlands Protection from the Clean Water Rule 1 (2017) (discussing inconsistent treatment of prior studies of regulatory costs and benefits); Caroline Cecot & Michael A. Livermore, The One-In, Two-Out Executive Order Is a Zero, 166 U. Penn. L. Rev. Online 1, 16 (2017); Jeremy Berg et al., Joint Statement on EPA Proposed Rule and Public Availability of Data, Science (May 4, 2018), https://science.sciencemag.org/content/360/6388/eaau0116.full (discussing efforts to ignore relevant scientific findings on benefits of pollution reduction); Richard L. Revesz, Pruitt Would Like Us to Ignore the Indirect Benefits of Environmental Regulations, Slate (June 13, 2018, 12:38 PM), https://slate.com/technology/2018/06/scott-pruitt-is-trying-to-undermine-environmental-regulation-in-a-creative-way.html (quoting former EPA Administrator’s promise to alter method of treating secondary effects of regulation). In addition, the appointment of Neomi Rao as the OIRA Administrator also departed from longstanding tradition of selecting relatively moderate technocrats with deep expertise in cost-benefit analysis methodology for that role. 239 Neomi Rao, Curriculum Vitae (2016) (on file with author). Rao, like many other prior appointees from both parties, was drawn from an academic post. 240Id. But unlike prior appointees (including Sally Katzen, John Graham, Cass Sunstein, and Howard Shelanski) Rao has no formal background in economics, cost-benefit analysis, or regulatory practice. 241Id. Instead, her most relevant prior work focused on constitutional questions in administrative law, such as removal and the delegation doctrine, which are quite far afield from the technical methodological questions that were previously considered most relevant for OIRA. 242See, e.g., Neomi Rao, Administrative Collusion: How Delegation Diminishes the Collective Congress, 90 N.Y.U. L. Rev. 1463, 1525 (2015); Neomi Rao, Removal: Necessary and Sufficient for Presidential Control, 65 Ala. L. Rev. 1205, 1276 (2014). Rao, however, is regarded as having deep political connections, having served in both the Bush White House and the Senate Judiciary Committee. 243 Steve Eder, Neomi Rao, The Scholar Who Will Help Lead Trump’s Regulatory Overhaul, N.Y. Times (July 9, 2017), https://www.nytimes.com/2017/07/09/business/the-scholar-who-will-help-lead-trumps-assault-on-rules.html. Administrator Rao’s tenure recently ended when she was confirmed to serve as a judge on the D.C. Circuit.

Judicial review of agency actions also gives the courts a role as a stabilizing presence in the regulatory state. Through both procedural and substantive review, courts are tasked with ensuring that agency decision-making proceeds along a reliable and scientific course, arriving at decisions that are not “arbitrary and capricious.” 244 5 U.S.C. § 706 (2012). As Part II demonstrated, the rigor of judicial review has changed with the underlying political landscape, beginning as relatively hands-off during the New Deal era and becoming more rigorous in the era of divided government. 245See supra Part I.B. During the era of partisan volatility, courts have frequently checked the President’s most ambitious or controversial uses of executive action, whether in the form of regulatory overreach or inaction. For its part, the Clinton Administration’s attempt to regulate tobacco as a drug under the Food, Drug, and Cosmetic Act, following legislative failures in 1998, 246 Joan Biskupic, FDA Can’t Regulate Tobacco, Supreme Court Rules 5 to 4, Wash. Post (Mar. 22, 2000), https://www.washingtonpost.com/archive/politics/2000/03/22/fda-cant-regulate-tobacco-supreme-court-rules-5-to-4/3e3fd725-a469-4261-b660-a3415d2f99c1/?utm_term=.8364a7b7ab17 (describing how FDA action on tobacco was in response to failed legislative efforts). was rejected by the Supreme Court in FDA v. Brown & Williamson247 FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 161 (2000). The Bush-era EPA was also pushed by the Supreme Court, this time in the direction of greater regulatory action. Despite the growing consensus around the threats of climate change, the EPA refused to make a finding that would trigger regulation under the Clean Air Act. The Court rejected this approach in Massachusetts v. EPA, finding that the statute only permitted inaction in this situation if the agency provided a reasoned scientific explanation for its policy. 248 Massachusetts v. EPA, 549 U.S. 497, 534 (2007) (“EPA has offered no reasoned explanation for its refusal to decide whether greenhouse gases cause or contribute to climate change. Its action was therefore ‘arbitrary, capricious, … or otherwise not in accordance with law.’”). More recent executive actions have been met with similar treatment in the lower courts. The Obama Administration suffered a second-term setback on DACA, 249 Texas v. United States, 809 F.3d 134, 178 (5th Cir. 2015), aff’d by an equally divided court, 136 S. Ct. 2271, 2272 (2016). while President Trump saw appellate courts enjoin his early executive actions related to immigration, 250 NAACP v. Trump, 298 F. Supp. 3d 209, 249 (D.D.C. 2018). leading to changes to the policy to survive judicial review. 251 This ultimately happened to revised versions of the travel ban in June of 2018. See Trump v. Hawaii, 138 S. Ct. 2392, 2423 (2018).

The potential for judicial stabilization of administrative action, however, may be limited, with the current party system threatening to undermine the independent role of the judiciary. The increased politicization of the judicial nomination process is nothing new, growing steadily worse throughout the era of divided government. Much like executive nominations, the number of vacancies reached new heights during Obama’s second term, with more than 100 vacancies at the time he left office. 252Judicial Vacancy List for December 2016, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2016/12/vacancies (last visited July 5, 2019). This figure was much higher than when Bush left office (42) or Clinton left office (67). Judicial Vacancy List for December 2008, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2008/12/vacancies (last visited July 5, 2019); Judicial Vacancy List for December 2000, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2000/12/vacancies (last visited July 5, 2019);See also Anne Joseph O’Connell, Shortening Agency and Judicial Vacancies Through Filibuster Reform? An Examination of Confirmation Rates and Delays from 1981 to 2014, 64 Duke L.J. 1645, 1660–61 (2015) (describing failed nominations by administration). The pace of judicial nominations quickly picked back up during unified Republican control of Congress. If this trend continues, fewer and fewer judicial appointments will be the product of bipartisan compromise, creating the risk of a more polarized judiciary, or at the very least the appearance of greater partisanship on the courts. 253 Indeed, there appears to be widespread agreement that the judicial confirmation process is becoming more politicized. David Russell, Politicization in the Federal Judiciary and Its Effect on the Federal Judicial Function, N.Y.U. J. Legis. & Pub. Pol’y Quorum 21, 26 (2018). Many commentators ascribed this trend to the Robert Bork nomination. Nina Totenberg, The Confirmation Process and the Public: To Know or Not to Know, 101 Harv. L. Rev. 1213, 1213 (1988); John M. Walker, Jr., The Unfortunate Politicization of Judicial Confirmation Hearings, Atlantic (July 9, 2012), https://www.theatlantic.com/national/archive/2012/07/the-unfortunate-politicization-of-judicial-confirmation-hearings/259445/. There is an active debate regarding whether increased politicization is necessarily a bad thing. Benjamin Wittes, Confirmation Wars: Preserving Independent Courts in Angry Times 76–77 (Peter Berkowitz & Tod Lindberg eds., 1st ed. 2006); Henry Paul Monaghan, The Confirmation Process: Law or Politics, 101 Harv. L. Rev. 1202, 1203 (1988). The steady erosion of the filibuster for judicial nominees of all stripes will only exacerbate the ideological separation between the parties’ nominees.

Many of the stabilizing forces within Congress have also become less prominent. Permanent committees, which were so often a place for bipartisan compromise in prior eras, have decreased in importance as omnibus legislation becomes the only means of moving bills. 254See Lara E. Chausow, R. Eric Peterson & Amber Hope Wilhelm, Cong. Research Serv., R43947, House of Representatives Staff Levels in Member, Committee, Leadership, and Other Offices, 1977-2016 (2016) (describing the shift from committee staff to individual member office support over time). This decline in staff began right at the beginning of the era of partisan volatility, with the Republican takeover of Congress in 1994. See Russell W. Mills & Jennifer L. Selin, Congressional Committee Staff Have Shrunk. Here Is One Way Congress Makes up the Difference, Wash. Post (June 14, 2017), https://www.washingtonpost.com/news/monkey-cage/wp/2017/06/14/congressional-committee-staffs-have-shrunk-heres-one-way-congress-makes-up-the-difference/?noredirect=on. Many commentators have noted the declining role of congressional committees and called for a rebalancing of the committee-leadership relationship. See, e.g., Kevin R. Kosar & Adam Chan, R Street Policy Study No. 66: A Case for Stronger Congressional Committees, R St. Inst. 1 (Aug. 2016) (“[T]he clearest path ahead may be to go backward, away from a hierarchical, leadership-dominated model of operating the chambers to one that disperses more power to committees.”). In the place of strong committee chairs, policy now runs more through party leadership, exacerbating the ideological character of the legislation. 255Kevin R. Kosar & Adam Chan, R Street Policy Study No. 66: A Case for Stronger Congressional Committees 1 (2016). At the same time, the nonpartisan institutions of the Legislative Branch—GAO, CBO, and CRS—are also shrinking. 256 Curtlyn Kramer, Vital Stats: Congress Has a Staffing Problem Too, Brookings Inst.(May 24, 2017), https://www.brookings.edu/blog/fixgov/2017/05/24/vital-stats-congress-has-a-staffing-problem-too/. Like the decline in committee staff, this change can be traced to the cuts following the 1994 midterm elections, right at the start of the current party system. 257 These deep cuts also included the elimination of another Legislative Branch institution, the Office of Technology Assessment, which provided “nonpartisan scientific studies of policy decisions.” Ed O’Keefe, When Congress Wiped an Agency Off the Map, Wash. Post (Nov. 29, 2011), https://www.washingtonpost.com/blogs/federal-eye/post/when-congress-wiped-an-agency-off-the-map/2011/11/29/gIQAIt0J9N_blog.html. In addition to lower funding levels, the nonpartisan congressional agencies are also the subject of partisan attacks. For instance, Republican leadership challenged the CBO during the fight for ACA repeal in 2017, going so far as to threaten the agency’s traditional role in developing a budget score for pending legislation. 258 Vann R. Newkirk II, The GOP Escalates Its Battle with the CBO, Atlantic (July 22, 2017), https://www.theatlantic.com/politics/archive/2017/07/republicans-cbo-hhs-score-bcra/534480/. At one point in the debate, members of Congress “floated the unprecedented idea of using the HHS score alone, or in concert with an evaluation from the White House Office of Management and Budget, as a substitute for the CBO’s score.” 259Id. Rather than turning to these institutions to provide stability, the current party system encourages Congress to marginalize these agencies when they stand in the way of achieving partisan objectives.

A final erosion in a stabilizing institution that is characteristic of the time concerns the intersection of polarization and the states. Former House Speaker Tip O’Neill (whose leadership in Congress coincided with much of the era of divided government) is closely associated with the phrase “all politics is local,” but this truism has become increasingly outdated in a time of increasingly ideologically oriented and polarized partisan affiliation. 260See generally Daniel J. Hopkins, The Increasingly United States: How and Why American Political Behavior Nationalized 227–28 (2018). One consequence of this change is that, when states play a role in implementing federal policy (as they often do), it creates additional opportunities for partisan contestation, rather than an opportunity to accommodate local interests or secure bipartisan cooperation through the distribution of local benefits. 261See Bulman-Pozen, supra note 3, at 995. States can resist federal policy through their own independent actions, and often do. 262See Bulman-Pozen & Gerken, supra note 35, at 1267; see also Deil S. Wright, Understanding Intergovernmental Relations 49 (1978) (presenting a model of federalism premised on overlapping authority, in which the federal government was not in hierarchical relationship with state and local actors). As a result, intergovernmental programs like Medicaid or funds for elementary and secondary education face two distinct kinds of pressure. First, they are subject to the same instability over time as all other federal programs. The choice to implement the program through state actors, however, introduces a second pressure: political opposition from the states in real time. 263See Bulman-Pozen & Gerken, supra note 35. This not only emerges from the states taking steps as autonomous actors, such as passing laws that frustrate federal policy or refusing to participate in a federal scheme, but it also comes from within programs, as states use the flexibility provided by the statutory scheme to deviate from federal policy preferences. 264See id. at 1264 (“[W]e argue that contestation can and does take place in many areas of federalism where states lack policymaking autonomy…. We do not quibble with the idea that sovereignty confers one sort of power, but we think it is a mistake to neglect the possibilities associated with a different sort of power – the power of the servant.”). Liberal states, for instance, refused to administer parts of the Patriot Act during the Bush Administration, 265See id. at 1278 (documenting state opposition to the Patriot Act based on state interpretations of the federal Constitution). just as many Republican states refused to expand Medicaid through ACA during the Obama years. 266 This decision was made easier for states following the Supreme Court’s holding that all existing Medicaid funds could not be conditional on the expansion under the ACA. See NFIB v. Sebelius, 567 U.S. 519, 585 (2012). In addition, intergovernmental administrative schemes are not the only forum where state-level actors use their position to engage in partisan conflict over regulatory or administrative matters. In recent years, state attorneys general, acting through expanded solicitor general offices, have become very active in litigation involving federal law. 267See Claire McCusker, The Federalism Challenges of Impact Litigation by State and Local Government Actors, 118 Yale L.J. 1557, 1557 (2009). For example, the Solicitor General of Utah filed an amicus brief on behalf of Utah and fifteen other states in Lucia v. SEC, a case about the Appointments Clause. 268 Brief for Utah et al. as Amici Curiae Supporting Petitioners, Lucia v. SEC, 138 S. Ct. 2044 (2018) (No. 17-130). In doing so, they cited an interest in “protect[ing] the structural safeguards embedded in the Constitution,” even though federalism was not at issue. 269Id. at 1. States are now a regular voice in highly charged litigation throughout the federal courts, offering yet another means of engaging in partisan disagreement.

The combined effects of increasingly nationalized politics, competitive partisan balance, ideological polarization, and weak party control prevent any single actor from limiting the harms that the current political system imposes on reasoned decision-making. Government is often divided, but not consistently enough to create the cooperative incentives of the 1970s and 1980s. Presidents are empowered to act unilaterally, but frequent changes in White House control limit the reach of policymaking to the next election. The courts and the states are drawn into the national political fight, rather than remaining in separate, more independent spheres. And meanwhile, political actors respond to increased polarization by pursuing ever-more extreme policy, unchecked by moderating forces in the elite party leadership.

The political trends discussed in this Part exist against a backdrop of administrative law doctrines that shape the incentives and options available to the relevant players. As seen in prior eras, administrative law has proven to be remarkably flexible, evolving in tandem with other political and administrative institutions. But, the mere fact that administrative law can and does change does not imply that those changes are normatively desirable—just as administrative law reforms might serve as a useful counterweight to negative pressures within a party system, they might also exacerbate its worse impulses. As we discuss in the following Part, the direction of U.S. administrative law does not appear well suited to meeting the growing challenges posed by the current political climate.

III. (Mal)adaptive Doctrine

The current party system and its interaction with administrative law pose serious threats to many values that are central to effective administration. Policies are more extreme and volatile and stabilizing institutions are less effective. Many of the administrative law staples that were at least arguably well suited to an era of sustained divided government—such as a strong OIRA, a lighter touch approach to judicial review of agency action, and a more political agency leadership—do little to mitigate the worst impulses of the new system.

Administrative law doctrine that was adapted to the current political environmental, however, could alleviate some of the worst tendencies of partisan volatility toward instability and irrationality. For example, the era of partisan volatility is especially likely to produce extreme policy reversals, whereby the major initiatives of one administration are rolled back in toto within a matter of years. Given this concern, doctrines that scrutinizes these reversals more rigorously than other sorts of agency action could at least reduce the extent of this negative tendency. Similarly, it may be that certain ways of structuring the government, such as limiting partisan influence through for-cause removal restrictions or the multi-member commission structure, facilitate stability and incrementalism and should be supported.

Unfortunately, the Court has shown little interest in doctrines that would reduce the risks of partisan volatility and, instead, seems more inclined toward administrative law reforms that are likely to exacerbate these risks. With respect to policy reversals, the Court’s most recent word on the issue came in FCC v. Fox Television Stations, Inc., which did not embrace heightened review of policy changes. 270 556 U.S. 502, 514 (2009). In addition, two signature areas of potential administrative law reforms for the Roberts Court—reducing the insulation of independent agencies from presidential control, and erosion of Chevron deference to agency legal interpretations—have substantial downside risks in the current political environment. With respect to independent agencies, the Court’s path is likely to exacerbate some of the worst pathologies of partisan volatility. The Court’s formal approach to these questions has pushed in the direction of a more unified Executive Branch with clearer lines of accountability to the political process. While the current doctrine leaves many questions unanswered, a continuation of this trend would further expose agency decision-making to back and forth of partisan volatility.

The consequences of the decline of Chevron are more ambiguous. The Court’s current move toward implementation of a major questions doctrine would insert courts into exactly those questions that are most politically salient, and therefore the most polarized along partisan lines. 271 The major questions doctrine can be traced to FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000). See Kevin O. Leske, Major Questions About the “Major Questions” Doctrine, 5 Mich. J. Envtl. & Admin. L. 479 (2016) (tracing the history of the doctrine). Repeated forays into the heart of the most deeply contested policy questions of the day runs the risk of undermining the legitimacy of judicial review and increasing partisan pressure on the process of judicial nomination and confirmation. However, revisions to Chevron deference also hold out some promise for promoting stability in the administrative system. Under existing deference doctrines, agencies can adjust legal interpretations in a way that stare decisis would typically not allow. Given the volatile nature of the current party system, placing more interpretive power in courts could have a stabilizing effect. But rather than focusing on the risks where courts could add some value—namely reversals of earlier legal interpretations—the Court seems more likely to shift judicial attention where it is least well suited by expanding the major questions doctrine.

In this Part, we take up each of these issues: arbitrary or capricious review of policy reversal, constitutional questions related to agency independence, and deference to agencies’ legal interpretations. We argue that the effect of future decisions in each of these areas can only be understood by explicitly considering the context set by partisan volatility. Doctrine that fits the needs of the day can promote effective administration and protect the core values of administrative law. Maladaptive doctrine will leave more harmful effects of partisan volatility unchecked, continuing our current experience of destabilizing policy swings and extreme policymaking. In any event, knowing how each doctrine interacts with the dynamics of the current party system is essential to understanding its implications.

A. Review of Policy Reversals

The era of partisan volatility leads to frequent reversals of position on significant policy questions. One highly ideological political party is replaced in the White House by a President at the other ideological extreme, often equipped with a only short period of unified government. As a result, sudden policy reversals are common, such as the Trump Administration’s about-face on climate change regulations, net neutrality, and deportation policy for immigrants who arrived in the United States as children.

At the outset, it is important to distinguish policy reversals from policy development. Any competent agency will reevaluate past decisions, just as it will look to new facts in determining if further action is needed. In keeping with this practice, OIRA expects agencies to take a look at old regulations at least once every five years. 272See Cary Coglianese, Moving Forward with Regulatory Lookback, 30 Yale J. on Reg. 57, 58–59 (2013) (describing regulatory “look-back” initiatives of the Obama Administration). We distinguish this sort of updating from a true “policy reversal.” Whereas the former is motivated by new facts in the world, the latter is motivated by a simple disagreement of policy, itself rooted in ideological and partisan concerns.

To illustrate, consider the Obama-era change to the overtime rule for white-collar workers. In 2004, the Bush Administration updated the overtime rule to raise the exemption threshold to $455 per week (or $23,660 per year). 273 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 69 Fed. Reg. 22,122, 22,123 (Apr. 23, 2004) (to be codified at 29 C.F.R. pt. 541). Under that policy, workers making less than that amount would receive overtime pay, while those making more would be exempt. The Obama Administration’s update to this rule doubled that amount to $913 per week (or $47,476 per year). 274 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 81 Fed. Reg. 32,391, 32,393 (May 23, 2016) (to be codified at 29 C.F.R. pt. 541). Part of this increase reflects little more than the usual updating you might expect an agency to undertake in light of new economic data and a rising cost of living. In the final rule, the Department of Labor noted that the salary level had been updated seven times since 1938, of which this was simply the latest iteration. 275Id. at 32,392. Other motives for the new rule, however, bottom out on policy disagreement. The Obama-era rule, for instance, noted that it disagreed with the Bush Administration methodology that made it easier to exempt lower-paid employees from the overtime rules. 276Id. at 32,396 (explaining that “the FLSA’s overtime protections are a linchpin of the middle class”).

It is this latter sort of policy change, which is driven by ideological and value-based concerns, that is likely to become both more common and more destabilizing as partisan volatility exerts pressure on the administrative state. Apart from the recent examples cited above, policy reversal has characterized each transition since 1992. Examples include the Bush Administration rollback of Clinton-era mining rules 277See Katharine Q. Seelye, Bush White House Reverses Clinton Decision on Mining, N.Y. Times (Oct. 26, 2001), https://www.nytimes.com/2001/10/26/us/bush-white-house-reverses-clinton-decision-on-mining.html. and Obama Administration reversal of Bush-era rules on hours of service for long haul truck drivers. 278 The Bush Administration’s rule concerning hours of service for commercial truckers was promulgated in 2008, after years of litigation challenging earlier iterations. See Hours of Service of Drivers, 73 Fed. Reg. 69567, 69568 (Nov. 19, 2008) (to be codified at 47 C.F.R. pt. 76). The new Obama-era rule was promulgated in 2011. See Federal Motor Carrier Safety Administration, Hours of Service of Drivers, 76 Fed. Reg. 81134, 81134 (Dec. 27, 2011) (to be codified at 47 C.F.R. pts. 385, 386, 390, 395).

Current practices do not fully account for the costs of these reversals, either as a matter of judicial doctrine or Executive Branch review. First, the Supreme Court does not explicitly apply additional scrutiny to agency actions that reverse previous policy determinations. This position is most clearly stated in FCC v. Fox Television Stations, Inc., where Justice Scalia wrote for the Court that there “is no basis in the Administrative Procedure Act ... that all agency change be subject to more searching review,” or that “agency action representing a policy change must be justified by reasons more substantial than those required to adopt the policy in the first instance.” 279 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 514 (2009). Under this view, an agency must be aware that the new policy is a departure from existing practices and must have reasons for the departure, but it does not need to compare the reasons to those that justified the previous regulation. 280Id. at 515. As a result, under the logic adopted in Justice Scalia’s opinion, agencies can work from the existing regulations in the same way as they would from a blank slate. 281 As in many areas of law, the Justices’ proclivity to produce numerous concurring and dissenting opinions muddies the waters of Fox Television Stations somewhat. In particular, Justice Kennedy provided the deciding vote and wrote separately to note that he agreed with some of the arguments found in Justice Breyer’s dissent, which did urge more probing review of agency policy reversals. See Jeffrey S. Lubbers, A Guide to Federal Agency Rulemaking 509–13 (6th ed. 2018).

In a vacuum, this interpretation of the APA might make sense. Such a practice ensures that agencies are aware of prior policy, recognize the departure, and provide a justification for the new practice. By not requiring more, the decision in Fox Television Stations allows for greater political responsiveness in agency decision-making. Since the standard applied to the policy reversal at Time 2 is the same as the initial determination at Time 1, the political party with control over the agency at Time 1 has no advantage. When an agency comes under new leadership, this standard simply allows for different conclusions on the same set of facts, without more.

Whatever merits this approach has, it’s clear disadvantage is regulatory instability. As the Court subsequently explained (in the context of Chevron deference), “an agency must also be cognizant that longstanding policies” may have “engendered serious reliance interests that must be taken into account.” 282See Encino Motorcars LLC v. Navarro, 136 S. Ct. 2117, 2125–26 (2016) (quoting Fox Television Stations, 556 U.S. at 515–16). As Justice Breyer’s dissent in Fox Television Stations recognized, the majority’s standard allows for policy change on the basis of “unexplained policy preferences,” not new facts. 283Fox Television Stations, 556 U.S. at 547 (Breyer, J., dissenting). Rather than treat the FCC decision as unconstrained by prior agency actions, Justice Stevens’s dissent went further, connecting the doctrine to the FCC’s role as an independent regulator and expert in its field: “There should be a strong presumption that the FCC’s initial views, reflecting the informed judgment of independent Commissioners with expertise in the regulated area, also reflect the views of the Congress that delegated the Commission authority to flesh out details not fully defined in the enacting statute.” 284Id. at 541 (Stevens, J., dissenting). As the dissent understood the issue, a failure to require a factual basis for the departure from prior policy elevated political responsiveness over independent expertise. 285See id. The actual difference between the majority and dissenting opinions in Fox Television Stations is uncertain. The majority references the importance of “reliance interests,” but the rule explicitly declines to impose a higher burden for situations in which a prior decision created such interests. Some lower courts, however, have read the majority to arrive at a position comparable to that of the dissent. Judge Williams’ opinion in U.S. Telecommunications is such an example. U.S. Telecomm. Ass’n v. FCC, 825 F.3d 674, 744-45 (D.C. Cir. 2016). Both sides seem to agree that the reliance created by prior regulations is important; the Breyer dissent, however, would require a higher justification to overcome these interests.

The reliance interests associated with policy reversals impose serious costs in any political context. They pose especially high costs in the current system. Control of Executive Branch policy has changed course every eight years since 1992. These changes, moreover, have been exacerbated by extreme polarization. A rational actor in today’s government, therefore, should recognize that the policies she puts in place are not only likely to erode or be ignored after the next election, but may be completely overhauled. The experience of the Clean Power Plan, discussed above, demonstrates how substantial these changes can be. 286See supra notes 225–229 and accompanying text. The current approach to judicial review of agency action does generally require an agency to take costs into account, and this would presumably include the sunk costs or reliance interests of prior rules. 287See, e.g., Bauer v. DeVos, 325 F. Supp. 3d 74, 101 (D.D.C. 2018). The district court in Bauer, applying the analysis of the majority in Fox Television Stations, found that the agency has insufficiently explained its reasons for changing its view of the legality of regulations related to student loan borrowers. Id. As Fox Television Stations demonstrates, however, these requirements often will not impose real constraints on sudden about-faces from an agency. 288See also Adrian Vermuele, Law’s Abnegation: From Law’s Empire to the Administrative State 160 (2016) (collecting cases since State Farm and finding that agencies often prevail under arbitrary and capricious review). A more explicit approach that is specifically targeted at policy reversals, such as the approach advocated by Justice Breyer’s dissent in that case, would be an improvement.

In brief, this position starts from a simple requirement: “To explain a change requires more than setting forth reasons why the new policy is a good one. It also requires the agency to answer the question, ‘Why did you change?’” 289Fox Television Stations, 556 U.S. at 549. This inquiry is necessarily more demanding than review of the initial decision, written on a blank slate. 290 Justice Breyer illustrates the point: “An (imaginary) administrator explaining why he chose a policy that requires driving on the right side, rather than the left side, of the road might say, ‘Well, one side seemed as good as the other, so I flipped a coin.’ But even assuming the rationality of that explanation for an initial choice, that explanation is not at all rational if offered to explain why the administrator changed driving practice, from right side to left side, 25 years later.” Id. Under this approach to arbitrary and capricious review, the costs of change must be addressed on their own terms. An agency that sought to significantly modify or repeal an existing rule would need to engage in additional fact-finding or analysis to support this conclusion. It may also need to account for uncertainty costs created by the change itself, as well as the likelihood that the new proposed rule would face opposition in a relatively short period of time. When taken into account, these considerations could have two effects. First, they would lead regulators to make more incremental changes to existing policy, given that more substantial deviations would require additional expense and likely impose greater costs. Second, the regulatory agencies, constrained by limited resources, would have incentives to focus on new issues, rather than revisiting prior regulations.

The recent experience with net neutrality provides an example of where a more rigorous review of policy reversals may matter. In 2015, the FCC finalized the Open Internet Order. 291 Protecting and Promoting the Open Internet, 80 Fed. Reg. 19,737 (Apr. 13, 2015) (to be codified at 47 C.F.R. pts. 1, 8, 20). This Order followed from prior unsuccessful efforts to require “net neutrality,” which prevents blocking and discrimination by broadband internet providers. 292 The initial rule was invalidated by the D.C. Circuit, which found that the FCC could not proceed without reclassifying broadband providers under Title II of the Communications Act of 1934. Verizon v. FCC, 740 F.3d 623, 659 (D.C. Cir. 2014). The 2015 Order imposed a range of new requirements of broadband internet providers, including “bright-line bans on blocking, throttling, and paid prioritization.” 293 Protecting and Promoting the Open Internet, 80 Fed. Reg. at 19,740. In supporting these regulations, the agency pointed to the prior decision of the D.C. Circuit, which had found its conclusions of fact supported by the record at the time. 294Verizon, 740 F.3d at 643–44. The 2015 Order itself involved a policy reversal, changing the classification of broadband providers from internet services to telecommunications services, which allowed for regulating this industry as common carriers. 295 Protecting and Promoting the Open Internet, 80 Fed. Reg. at 19,742–43. This change, however, was based on more than a decade of technological change, reflected in the new facts set forth in the agency record. 296Id. at 19,743. The policy was upheld by the D.C. Circuit in 2016, which cited to Fox Televisions Stations for the applicable standard. 297 U.S. Telecomm. Ass’n v. FCC, 825 F.3d 674, 706–07 (D.C. Cir. 2016). Judge Williams dissented on this point, placing more importance on the “reliance interests” language in Fox Television Stations298Id. at 745–46 (Williams, J., concurring in part and dissenting in part). On this view, the FCC failed to adequately consider whether the new facts overcame the reliance interests created by the old rule. 299Id. at 748.

Despite this authority for the new rule’s legality, the FCC changed course again only a year later. Following the election of Donald Trump and the appointment of Ajit Pai to head the FCC, the agency released its Restoring Internet Freedom Order on December 14, 2017. 300 Restoring Internet Freedom, FCC 17-166, (2017) (declaratory ruling, report and order, and order). The Order reversed the Title II reclassification for internet providers and removed the additional disclosure requirements of the Open Internet Order of 2015, along with other changes. 301Id. at ¶¶ 2–3. In the separate cost-benefit analysis that accompanied the 2017 Order, the FCC found that many of the prior regulations had no benefits at all, while finding that the rule led to significant costs for innovation and created uncertainty in the affected industries, directly refuting the agency’s earlier expert assessment. 302Id. at ¶¶ 304–12. The agency did not, however, consider the uncertainty costs associated with the new rule itself, or the broader political climate concerning the issue. 303Id. The 2017 Order generated considerable opposition, including from Senate Democrats in advance of the 2018 midterms, 304 John Hendel, Senate Democrats Score Net Neutrality Win in Bid for Midterm Momentum, Politico (May 16, 2018, 9:53 AM), https://www.politico.com/story/2018/05/16/senate-democrats-net-neutrality-539275. and it very well may find itself undermined if party balance swings back to Democrats.

The result of this back and forth is that, in a period of less than three years, both parties have pushed diametrically opposed policies, affecting an innovative and sizable industry, through unilateral executive action, which has been swiftly challenged in Congress and the courts. The first policy was reversed in a matter of years, and the new policy finds itself under attack. Despite these predictable political outcomes, present since 1992, the agency did not make explicit findings, at either stage, regarding the cost associated with political instability.

This single example demonstrates a much broader trend. Agencies are led by political leadership focused on pursuing increasingly ideological policy. Their successors, reliably a member of the other party, are incentivized to undercut prior efforts. As currently understood, arbitrary and capricious review does not account for how likely this fact pattern is at present.

Greater scrutiny of reversals would mitigate the costs of political instability, leading to more incremental policymaking. This scrutiny would also incentivize a more forward-looking regulatory agenda. 305 A failure to appropriately consider reliance interests was part of the basis for a recent decision finding that the Trump Administration’s decision to end the DACA program was unsupported. NAACP v. Trump, 315 F. Supp. 3d 457, 473 (D.D.C. 2018). The opinion adopts a more robust reading of the Supreme Court’s current case law regarding policy reversals, suggesting the law may be ripe for an explicit adoption of the standard we advocate. Id. (quoting Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125–26 (2016)) (“[I]t is not up to Secretary Nielsen—or even to this Court—to decide what she should or should not consider when reversing agency policy. Rather, the requirements are set by the APA, as interpreted by the Supreme Court: ‘When an agency changes its existing position, it … must … be cognizant that longstanding policies may have engendered serious reliance interests that must be taken into account.’”). If the courts looked with more skepticism on agency decisions to revisit their earlier policy choices, the agency will have greater reason to look in new directions. For instance, an FCC chair may be less willing to spend his first year revisiting net neutrality and may instead look to an area in need of attention, such as increasing broadband access. This is not to say that all deregulation should be more difficult than imposing new regulations. Adopting the position advanced by Justice Breyer in Fox Television Stations would instead prevent deregulation—or new regulation—that reflects a naked policy choice, rather than a change premised on a new factual record. 306 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 549 (2009) (Breyer, J., dissenting). Moreover, the rule is not absolute. Agencies are allowed to deviate from prior practice on the basis of policy, they just must provide a more thorough explanation than if operating from a blank slate. 307Id. To paraphrase Justice Breyer, a decision to change which side of the road we drive on needs a greater explanation than picking the right side in the first place, especially when the road signs will just be changed back in a few years. 308Id.

Of course, Justice Breyer was in dissent in Fox Television Stations, and his more rigorous approach to policy reversals has not carried the day. It may be too soon to call the matter closed, however. Even if the unique costs of reversals are not explicitly part of the doctrine at present, the instruction to consider reliance costs may allow courts to require a similar level of scrutiny under “traditional” arbitrary and capricious review. The Trump Administration, for instance, has had a difficult time in the lower courts trying to justify policy reversals. 309Roundup: Trump-Era Deregulation in the Courts, Inst. for Pol’y Integrity, https://policyintegrity.org/deregulation-roundup (last updated Aug. 6, 2019) (documenting historically low win rate for the Trump Administration in legal challenges to agency decisions). A doctrine that makes these costs explicit would obviously be preferable to one that does not, but it may be that the courts are becoming more cognizant of the unique costs sudden policy reversals pose.

B. Agency Independence

Another area of recent attention by the Supreme Court is agency independence. In setting up agencies, Congress makes a number of choices about how its leaders will be selected, how its funds will be allocated, and how its relationship with the White House will be structured. In defining constitutional limits on these choices, the Court has recently shown a willingness to impose new constraints on the use of institutional design to insulate agencies from political control, increasing agency exposure to the volatile influence of political oversight. 310 Free Enter. Fund v. PCAOB, 561 U.S. 477, 484 (2010).

Many recent constitutional challenges seek to restrict agencies’ statutory independence from the President. The first of these challenges involves the independence of agency heads while in office. The second involves selection, asking whether certain officials are “officers” at all. The resolution of each question has the potential to place agencies under greater political control, with a corresponding loss of control for employment-protected career civil servants.

First, the Supreme Court has long recognized that there may be some limit to Congress’ authority to restrict the President’s control over Executive Branch officers. 311 For a general discussion of the structural conditions that make agencies more independent from political influence, see Rachel E. Barkow, Insulating Agencies: Avoiding Capture Through Institutional Design, 89 Tex. L. Rev. 15 (2010). Since the New Deal era, the Supreme Court was willing to uphold all manner of restrictions on presidential control. 312See, e.g., Nixon v. Gen. Servs. Admin., 433 U.S. 425 (1977) (upholding a statute that allowed for agencies to manage presidential records over a claim of privilege); Humphrey’s Ex’r v. United States, 295 U.S. 602 (1935) (upholding a for-cause removal restriction on a federal officer). This trend ended in 2010 with the Court’s decision in Free Enterprise Fund v. PCAOB313Free Enter. Fund, 561 U.S. at 484. In severing a provision providing a for-cause removal provision for inferior officers within the SEC, the Court found that the Vesting Clause of Article II operates as an independent bar on removal restrictions, echoing the earlier, more restrictive rationale of Myers v. United States314Id. at 492–93 (citing Myers v. United States, 272 U.S. 52, 164 (1926)). The reach of this decision is uncertain, with much of the opinion grounded in the perception that the PCAOB structure was unlike any the Court had previously considered. 315See Richard H. Pildes, Free Enterprise Fund, Boundary-Enforcing Decisions, and the Unitary Executive Branch Theory of Government Administration, 6 Duke J. Const. L. & Pub. Pol’y 1, 3–4 (2011) (“In Free Enterprise Fund, the Court thus had something of a free pass in invalidating the Board’s structure: it could hold this peculiar design to be an unconstitutional limitation on the president’s Article II powers without putting much else about the structure of administrative government up for grabs.”). In fact, the Supreme Court was careful to note that the decision did not disrupt the prevailing doctrine for agencies with only one layer of removal protection. 316Free Enter. Fund, 561 U.S. at 514. Some scholars, however, (including the former Administrator of OIRA, Neomi Rao) have seized on the more expansive language of the opinion to argue for the invalidation of all similar provisions in federal law. 317 Neomi Rao, A Modest Proposal: Abolishing Agency Independence in Free Enterprise Fund v. PCAOB, 79 Fordham L. Rev. 2541, 2559 (2011) (“Despite claims to minimalism by the Court, Free Enterprise Fund logically implicates the constitutionality of agency independence.”); see also Neomi Rao, Removal: Necessary and Sufficient for Presidential Control, 65 Ala. L. Rev. 1205, 1208 (2014).

Even more recently, challengers made a similar argument against the Consumer Financial Protection Bureau, established as part of the Dodd-Frank financial reforms. 318See generally PHH Corp. v. Consumer Fin. Protection Bureau, 881 F.3d 75, 92–101 (D.C. Cir. 2018) (en banc). These challenges are premised on the unique combination of features found in the CFPB, which insulate the agency from some of the political controls of both the President and Congress. While the agency’s structure was recently upheld by the D.C. Circuit, the issue led to a fractured court, with numerous concurrences and dissents on the constitutional question. 319Id. at 110-111. This dispute, moreover, is not limited to an application of an agreed upon test for questions of agency independence. Instead, the judges disagreed about how to reconcile the Supreme Court’s precedents in this area. 320See generally id. Even more recently, a divided panel of the Fifth Circuit held that the Federal Housing Finance Agency was unconstitutionally structured. 321 Collins v. Mnuchin, 869 F.3d 640, 674 (5th Cir. 2018). As a result, the issue is likely to arrive at the Court in due time, when the authority of the President to control agencies under the Vesting Clause will be revisited. When it does, the Court may be more inclined to strike down statutes designed to insulate agencies than it was before. Its newest member, Justice Kavanaugh, was on the D.C. Circuit panel that heard the challenge to PCAOB’s structure, and dissented from his colleagues’ view that there was no constitutional problem, 322 Free Enter. Fund v. PCAOB, 537 F.3d 667, 685, 715 (D.C. Cir. 2008) (Kavanaugh, J., dissenting). and also dissented from the en banc decision upholding the removal restriction for the head of the CFPB. 323PHH Corp., 881 F.3d at 110–11.

The second recent fight has been fought on much more familiar terrain: the Appointments Clause. Since Buckley v. Valeo, the Supreme Court has been distinguishing between principal officers, inferior officers, and mere employees. Rather than leaving these determinations to Congress, at least in situations where the Constitution does not explicitly say otherwise, the Buckley Court found that all government personnel who “[exercise] significant authority pursuant to the laws of the United States” are officers. 324 Buckley v. Valeo, 424 U.S. 1, 126 (1976). Even though the current Court appears to be uniformly committed to this project, the actual line-drawing has created much disagreement and confusion. The decisions differentiating principal officers from their inferior counterparts have generated significant controversy, with the Court applying different formulations in two prominent cases, Morrison v. Olson and Edmond v. United States325Compare Morrison v. Olson, 487 U.S. 654 at 671–73 (1988), with Edmond v. United States, 520 U.S. 651, 662–63 (1997). Likewise, the line between employees and officers has proven difficult to discern. The Court’s most thorough consideration of the topic, Freytag v. Commissioner326See 501 U.S. 868 (1991). adopted a factors-based approach that divided the lower courts. 327 Intercoll. Broad. Sys., Inc. v. Copyright Royalty Bd., 684 F.3d 1332, 1342 (D.C. Cir. 2012).

Because of Free Enterprise Fund, where the Court found that two layers of for-cause removal protection was impermissible for officers, the importance of these issues has only grown. 328Free Enter. Fund, 561 U.S. at 514. The intersection between the Court’s appointment and removal jurisprudence was on display in Intercollegiate Broadcasting System, Inc. v. Copyright Royalty Board, decided by the D.C. Circuit in 2012. 329Id. Faced with a statute that provided for-cause removal for the Copyright Royalty Board, the Court reasoned that if the Board members were principal officers (based on the removal protection) then their appointment by the Librarian of Congress would be unconstitutional. 330Id. Under these two doctrines, the Court has created a link between the political control exercised by the President and the government personnel that must be selected through the political mechanism of the Appointments Clause.

Lucia v. SEC is the most recent foray into this area, and yet another win for those who favor a more expansive view of “officer” under the Appointments Clause. 331 Lucia v. SEC, 138 S. Ct. 2044, 2047 (2018). Adhering to the test set forth in Freytag, the majority found that SEC administrative law judges (ALJs) are officers of the United States. 332Id. at 2055. Reasoning by analogy, the Court found that the duration of the ALJ office, the discretion ALJs exercise, and the significance of the decisions ALJs make either mirrored or went beyond the judges at issue in Freytag333Id. at 2053. The majority opinion did not go beyond this analogy, 334Id. (“Freytag says everything necessary to decide this case.”). and the scope of the opinion is an open question for other ALJs and non-ALJ adjudicators, including the hundreds handling immigration and social security cases. 335See generally Kent Barnett et al., Administrative Conference of the United States, Non-ALJ Adjudicators in Federal Agencies: Status, Selection, Oversight, And Removal (Draft Report Feb. 14, 2018) in 53 Ga. L. Rev. 1 (2019) (examining role of non-ALJ adjudicators). Justices Thomas and Gorsuch, however, are prepared to go much further, finding a constitutional officer in any employee who has “an ongoing statutory duty.” 336 Lucia v. SEC, 138 S. Ct. at 2056 (Thomas, J., concurring) (quoting NLRB v. SW Gen., Inc. 137 S. Ct. 929, 946 (2017)). In declining to reach the statutory question, Justice Breyer noted the broader implications of the Court’s position. Noting the connection between Free Enterprise Fund and the Appointments Clause jurisprudence, he suggested that a finding that ALJs are officers could imperil not only their selection process, but also their removal protections. 337Id. at 2061–62 (Breyer, J., concurring in the judgment). Lucia does not provide an answer to these questions, but it did signal that other ALJs may soon be at risk, either under the Freytag analysis or the more ambitious approach of Justice Thomas’s concurrence.

Taken together, these changes in the doctrine suggest that two things will become more likely in the future. First, agency leadership will be subject to greater political control, 338 Greater political control does not necessarily mean a more politically responsive government. Even if elections reward candidates who reflect the views of the median voter in the electorate, elected officials will still reliably deviate from those views. In that situation, bureaucratic insulation may impose a “compensatory inertia” on policymaking process and actually bring policy outcomes closer to the views of the electorate. See Matthew C. Stephenson, Optimal Political Control of the Bureaucracy, 107 Mich. L. Rev. 53, 55 (2008) (“Up to a point, the benefit to a majority of voters from a reduction in outcome variance [that results from bureaucratic insulation] outweighs the costs associated with biasing the expected outcome away from the median voter’s ideal outcome.”). especially in light of the enthusiasm shown by the Trump Administration to use these decisions as an opportunity to go even further. 339See Lubbers, supra note 6, at 6–7. Second, the Executive Branch will require more personnel confirmed by the Senate to sustain current operations. These developments bring the hyper-partisanship and volatility of the political climate further into the administrative state. Additional positions subject to Senate confirmation create additional forums for partisan conflict. In unified government, these positions will become more extreme. In divided government, they will likely go vacant for sustained periods of time. Once appointed, a shift toward greater political control means they will pursue more ideological policy and bring further confrontation to the regulatory agenda. By constitutionalizing these administrative structures, a legal regime that emboldens the Vesting Clause and adopts a more expansive definition of “officers” prevents Congress from taking steps to allow agencies to better weather more intense political winds. 340 These doctrinal developments also put more pressure on the internal constraints placed on agencies, which are largely beyond the reach of courts. Internal agency processes have become a greater focus of attention in recent years and could be challenged by greater political influence on agency leadership. See Jennifer Nou, Intra-Agency Coordination, 129 Harv. L. Rev. 421, 424 (2015) (providing a general theory of how these structures are developed); see, e.g., Rebecca Ingber, Interpretation Catalysts and Executive Branch Legal Decisionmaking, 38 Yale J. Int’l L. 359, 404–12 (2013) (describing how internal processes altered outcomes in the national security context).

C. Deference

Chevron and Auer deference have become perennial topics of debate in administrative law, and the Court has shown increased appetite toward doctrinal reform in the area of judicial deference to agencies’ legal interpretations. Accounting for the current party systems helps clarify some of the stakes to this reform effort. The dominant approach by the Court, and one embraced by at least some academic commentators, has been to reduce the level of deference given to agency interpretations that implicate “major questions” of law and policy. 341See King v. Burwell, 135 S. Ct. 2480, 2483 (2015); FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 157 (2000); see also Lawson, supra note 11, at 673. But see Nathan Richardson, Keeping Big Cases From Making Bad Law: The Resurgent “Major Questions” Doctrine, 49 Conn. L. Rev. 355, 358 (2016) (defending the major questions doctrine as a means of keeping judges from overturning Chevron in its entirety on the basis of important cases). At a time of increasing polarization, this represents a risky approach because it inserts the judiciary exactly where it is least welcome: into the most controversial and contested policy issues, with heated positions on both sides. To the degree that the major questions doctrine increases the perception of judges as carrying out a partisan agenda, it risks undermining the legitimacy of the judiciary and increasing partisan influence over the nomination and confirmation of judges.

However, there is a countervailing pressure that provides some reason to endorse a decline of deference, at least under certain circumstances. Under Chevron, agencies can change their minds about the law (within reason) without any especially high bar to doing so. 342See Nat’l Cable & Telecomm. Ass’n v. Brand X Int’l Servs., 545 U.S. 967, 982–83 (2005). If Chevron were weakened, legal change would be constrained by normal principles of stare decisis. As discussed above, the era of partisan volatility makes this tradeoff between stability and flexibility especially important as reversals on questions of law become more and more common. 343See, e.g., Alan Feuer, Justice Department Says Rights Law Doesn’t Protect Gays, N.Y. Times (July 27, 2017), https://www.nytimes.com/2017/07/27/nyregion/justice-department-gays-workplace.html; see also Coglianese, supra note 272, at 58–59. Were courts to increase their scrutiny of changes to legal interpretations (mirroring the heightened arbitrary and capricious review discussed above) that could help mitigate the problem of policy oscillation and regulatory uncertainty inherent in an era of partisan volatility.

Under the doctrine of Chevron deference, courts will defer to agency interpretations of federal statutes in certain situations. Auer deference provides agencies with similar interpretive latitude for interpretations of their own regulations. 344 Auer v. Robbins, 519 U.S. 452, 462–63 (1997). Few legal topics have received more sustained attention in recent years than the continued vitality of these doctrines. 345See generally Christopher J. Walker, Attacking Auer and Chevron Deference: A Literature Review, 16 Geo. J.L. & Pub. Pol’y 103, 104 (2018) (“In recent years, we have seen a growing call from the federal bench, on the Hill, and within the legal academy to rethink administrative law’s deference doctrines to federal agency interpretations of law.”). The confirmation of Justice Gorsuch to the Supreme Court in 2017 gave additional momentum to Chevron’s detractors, who rightfully saw him as a skeptic of the doctrine. 346See Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1149 (10th Cir. 2016) (Gorsuch, J., concurring); see also Garco Constr., Inc. v. Speer, 138 S. Ct. 1052, 1052 (2018) (Thomas, J., joined by Gorsuch, J., dissenting from denial of certiorari). While the debate surrounding Chevron has largely focused on internal legal objections, such as the proper constitutional role of the courts in construing the

law, 347 For instance, Justice Thomas has written forcefully that both Chevron and Auer deference are inconsistent with Article III, particularly the judiciary’s obligation to “say what the law is.” See Michigan v. EPA, 135 S. Ct. 2699, 2712 (2015) (Thomas, J., concurring) (criticizing Chevron); Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1222–23 (2015) (criticizing Auer). the existing political climate suggests that there could be significant adverse consequences to the Court’s approach to abandoning the doctrine.

The scope of Chevron deference expanded in the first decade after the case that gave the doctrine its name. Shortly after the decision in Natural Resources Defense Council, Inc. v. Chevron, the D.C. Circuit interpreted the doctrine to allow for broad agency discretion in defining ambiguous law, even in statutes where the agency was not expressly empowered to engage in legal interpretation. 348See Rettig v. Pension Benefit Guar. Corp., 744 F.2d 133, 151 (D.C. Cir. 1984). In the following years, the doctrine grew as the court took a more expansive view of what provisions of law are “administered” by the agencies. 349See, e.g., Wagner Seed Co. v. Bush, 946 F.2d 918, 925 (D.C. Cir. 1991) (holding that the EPA is entitled to deference on reimbursement provision of CERCLA over dissent by Judge Williams). Courts also extended the deference afforded to agency constructions of statutory text to agency interpretations of their own regulations, a doctrine known as Auer deference. 350See Auer v. Robbins, 519 U.S. 452 (1997). This trend toward greater agency policymaking was consistent with the political climate, placing power in the hands of reliably conservative presidents. 351See supra Part II.C.

Since 2000, however, Chevron deference has been eroded, both as a doctrine and in popularity with academic commentators. Doctrinally, the Court in Brown & Williamson found that Chevron deference was inapplicable to certain “major questions.” 352 FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 159 (2000). Despite considerable uncertainty about the scope of that holding at the time, 353See Lawson, supra note 11, at 673. it was applied more recently in the context of regulations to enforce the ACA. 354 King v. Burwell, 135 S. Ct. 2480, 2489 (2015). In 2001, Chevron deference was further limited to agency interpretations that had the force and effect of law, which narrowed the substantive lawmaking power of agencies. 355See United States v. Mead Corp., 533 U.S. 218, 226–27 (2001). For Auer deference, the Court similarly circumscribed the doctrine’s scope, finding that deference may be inappropriate when there are not sufficient indicators of reasoned judgment by the agency. 356See Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 155 (2012) (“Although Auer ordinarily calls for deference to an agency’s interpretation of its own ambiguous regulation, even when that interpretation is advanced in a legal brief, this general rule does not apply in all cases. Deference is undoubtedly inappropriate, for example, when the agency’s interpretation is plainly erroneous or inconsistent with the regulation. And deference is likewise unwarranted when there is reason to suspect that the agency’s interpretation does not reflect the agency’s fair and considered judgment on the matter in question. This might occur when the agency’s interpretation conflicts with a prior interpretation, or when it appears that the interpretation is nothing more than a convenient litigating position, or a post hoc rationalization advanced by an agency seeking to defend past agency action against attack.” (citations omitted)). Despite these erosions, the core of the Chevron doctrine—that agencies have power to resolve ambiguous legal questions—has more or less remained untouched. 357See Nat’l Cable & Telecomm. Ass’n v. Brand X Int’l Servs., 545 U.S. 967, 982–83 (2005).

The academic debate surrounding Chevron has accelerated in recent years. Many of the arguments for moving away from Chevron and Auer are transcendental and internal, starting primarily from formal constitutional premises. 358See, e.g., Philip Hamburger, Chevron Bias, 84 Geo. Wash. L. Rev. 1187, 1189 (2016). Defenders of Chevron tend to focus more closely on pragmatic concerns. See, e.g., Richard J. Pierce, Jr., Chevron and Its Aftermath: Judicial Review of Agency Interpretations of Statutory Provisions, 41 Vand. L. Rev. 301, 303 (1988) (arguing that Chevron respects the distinction between law and policy). While Chevron deference can be understood in formal terms as a division of power to interpret the law between the Judicial and Executive Branches, a functional understanding highlights a delicate set of tradeoffs generated by the doctrine in an era of partisan volatility.

One feature of this era is the increasingly partisan character of the judiciary. Recent research suggests that Chevron deference constrains partisanship in judicial decision-making. 359 Kent H. Barnett, Christina L. Boyd & Christopher J. Walker, Administrative Law’s Political Dynamics, 71 Vand. L. Rev. 1463, 1467 (2018). Shifting the more policy-tinged parts of legal interpretation to judges might exacerbate the degree of partisan political attention paid to the regulatory views of potential judges. This is especially the case under the Court’s current doctrinal framework, with its emphasis on increasing review of “major questions.” If political actors can expect that the major policy issues of the day—even those delegated to administrative agencies—will ultimately be decided de novo in the courts, then they will expand their efforts to influence the nomination and confirmation process. Given the degree to which judicial appointments are already politicized, further incentives for a broader range of interests groups to wade into the process seem likely to pose real risks to the legitimacy and independence of the judiciary.

On the other hand, greater political accountability leads to greater disruption as ideological swings become more common. This is true for legal interpretation, just like it is for the questions of policy discussed above. The Trump Administration has taken a variety of legal positions different from its predecessor, 360See, e.g., Alan Feuer, Justice Department Says Rights Law Doesn’t Protect Gays, N.Y. Times (July 27, 2017), https://www.nytimes.com/2017/07/27/nyregion/justice-department-gays-workplace.html. just as the Obama Administration took positions at odds with those of the Bush-era regulators. 361See supra note 278 and accompanying text.Chevron facilitates volatility for the subset of questions that are entitled to deference. 362 For example, questions of statutory interpretation are at the heart of the dispute over the Clean Power Plan, leading to the potential for substantial volatility if those interpretations swing wildly between administrations. See supra notes 225–227 and accompanying text. Under Chevron, an agency can change its mind over time, so long as the new interpretation is still a reasonable resolution of an ambiguous question. 363See United States v. Home Concrete & Supply, LLC, 132 S. Ct. 1836, 1843–44 (2012). The same logic applied to Auer deference. When an Article III court is the one deciding the law, doctrines of stare decisis provide a much greater degree of stability. Scaling back Chevron or Auer would limit the ability of agencies to shift views in response to partisan volatility and may instead encourage them to spend their time on undecided legal questions.

Understanding the political context of the debate over Chevron and Auer helps bring the relevant tradeoffs into focus. Partisan volatility and polarization increase the cost of allocating decisions to agencies, because they are likely to oscillate between wildly conflicting views, with all of the associated undesirable social consequences that oscillation entails. Greater assertion of authority over statutory and regulatory interpretation could help stabilize the situation, by placing control in a relatively less volatile institution. However, abandoning Chevron and Auer risks politicizing the judiciary even further, as courts find themselves inserted into policy questions, without even the doctrines of deference to help shield them from political scrutiny and criticism. If courts become a major forum for policymaking on a greater number of issues, an expanded set of interest groups will feel compelled to participate in the process of judicial appointments, further politicizing an already deeply polarizing process.

Conclusion

Modern politics poses a very real threat to the U.S. administrative state. Ideological positions are further apart, and swings between them are increasingly common. Cooperative incentives are low, and national partisan conflict extends to the states and the courts. While administrative law has in the past adapted to match new political realities, such an adjustment is not necessary or inevitable, as current doctrinal trends demonstrate.

 

Confronting the realities of the current era of partisan volatility opens the possibility of transforming administrative law in ways that stabilize its capriciousness and curb its worst excesses. By doing so, law can facilitate administrative policymaking that is more stable and more forward-looking—grounding administration in the values of neutrality, expertise, and impartiality without sacrificing responsiveness to broad majoritarian desires. During a period of increased volatility, public policy requires more than bare reactivity to the most recent electoral happenstance, where each new electoral winner is roughly as out of step as the loser. The goal of administrative law should be to facilitate the flexibility that allows policy to respond to new circumstances, while also insulating the administrative state from partisan whim. Historical experience provides some hope that accommodation of this sort is at least possible. The growth of the administrative state has not been a straight line, but rather a series of specific adjustments to meet the demands of a political system. Once again, new political realities present new challenges. Frankly acknowledging and accounting for these challenges is the most basic precondition for meeting them.

In the future, new challenges will undoubtedly arise. The era of partisan volatility will not be with us forever. After recent electoral victories, both major political parties have boasted that they were on the cusp of breaking through the period of volatility by building a political coalition capable of sustained power. The slogan “demographic destiny” captured this optimism for Obama-era Democrats, 364See, e.g., Lisa García Bedolla & Kerry L. Haynie, The Obama Coalition and the Future of American Politics, 1 Pol. Groups & Identities 128, 132 (2013). just as Karl Rove’s invocation of the “permanent majority” buoyed Republicans years earlier. 365 Paul R. Abramson, John H. Aldrich & David W. Rohde, The 2004 Presidential Election: The Emergence of a Permanent Majority?, 120 Pol. Sci. Q. 33, 33–34 (2005). Obviously, neither came to pass. It may be, however, that eventually a similar prediction will come true. If other features hold, then American politics could soon be shaped by one dominant party, pursuing a relatively extreme ideological agenda for a sustained period, while being vigorously opposed by a substantial minority of the country. Unlike the earlier periods of unified government, such as the New Deal, this new governing majority may lack the constraining forces that were once imposed by party elites, state and local politics, or a strong Congress. The challenges that such a system would pose for effective administration would no doubt be very different than the ones created by the current environment. What remains constant, however, is the basic notion that these challenges can only be confronted once they are understood.

Footnotes

*Michael A. Livermore is a professor at the University of Virginia School of Law. Daniel Richardson, University of Virginia School of Law, J.D., 2018. The views expressed herein are the authors’ alone. Thanks to Rachel Barkow, Jeffrey Lubbers, Nina Mendelson, Dick Pierce, Teddy Rave, Richard L. Revesz, Mark Tushnet, Christopher Walker, and workshop participants at the University of Houston Law Center for helpful feedback on earlier versions of this Article.

1See Max Weber, Bureaucracy, in From Max Weber: Essays in Sociology 196, 232 (H. H. Gerthand & C. Wright Mills eds., 1946); Woodrow Wilson, The Study of Administration, 2 Pol. Sci. Q. 197, 197 (1887). See generally Dwight Waldo, Politics and Administration: On Thinking about a Complex Relationship, in A Centennial History of the American Administrative State 89 (Ralph C. Chandler ed., 1987).

2See Paul H. Appleby, Policy and Administration (1949); Frank J. Goodnow, Politics and Administration: A Study in Government (1900); Dwight Waldo, The Administrative State: A Study of the Political Theory of American Public Administration xvii (2d ed. 1984); Gerald E. Caiden, In Search of an Apolitical Science of American Public Administration, in Politics and Administration: Woodrow Wilson and American Public Administration 51 (Jack Rabin & James S. Bowman eds., 1984) (discussing mid-century heyday and eventual challenges to more traditional notions of a politics/administration dichotomy); Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J. L. Econ & Org. 81 (1985) (introducing “Law of Conservation of Administrative Discretion”); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667 (1975).

3See Josh Chafetz, Congress’s Constitution 28–35 (2018) (surveying literature on congressional oversight during periods of unified government and finding that Congress uses its power more sparingly during those periods); Jessica Bulman-Pozen, Executive Federalism Comes to America, 102 Va. L. Rev. 953 (2016); Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077 (2014) (arguing that national polarized politics interacts with federalism); Neal Devins, Party Polarization and Congressional Committee Consideration of Constitutional Questions, 105 Nw. U. L. Rev. 737, 758 (2011) (discussing “blood-sport” battles brought on by political incentives); Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245 (2001); Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2311 (2006) (arguing that unified control undermines inter-branch competition essential to oversight); Michael A. Livermore, Political Parties and Presidential Oversight, 67 Ala. L. Rev. 45 (2015); Thomas O. McGarity, Administrative Law as Blood Sport: Policy Erosion in a Highly Partisan Age, 61 Duke L.J. 1671, 1711–12 (2012) (examining consequence of polarization); Anne Joseph O’Connell, Shortening Agency and Judicial Vacancies Through Filibuster Reform? An Examination of Confirmation Rates and Delays from 1981 to 2014, 64 Duke L. J. 1645 (2015).

4 Lucia v. SEC, 138 S. Ct. 2044, 2049 (2018).

5See Exec. Order No. 13,843, 83 Fed. Reg. 32,755 (July 13, 2018) (excepting ALJs from the competitive service).

6 Opposition to the move has been heated, with the President of the American Bar Association referring to the order as “ill-considered and legally vulnerable.” Letter from Hilarie Bass to Congressman Pete Sessions and Congressman James McGovern (July 16, 2018), https://www.americanbar.org/content/dam/aba/uncategorized/GAO/ALJlettertorulescommitteeaboutEO.authcheckdam.pdf; see also Jeffrey S. Lubbers, The RAA Loses Steam But the Trump Executive Order on ALJ Selection Upturned 71 Years of Practice, 94 Chi.-Kent L. Rev. (forthcoming 2019).

7See infra Part III.

8See generally Jerry Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012).

9Id. at 175.

10 For other foundational work, see Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920 (1982).

11See Chloe Breyer et al., Administrative Law and Regulatory Policy 14–29 (6th ed. 2006); see, e.g., Richard J. Pierce, Jr., Administrative Law (5th ed. 2017); Gary Lawson, Federal Administrative Law 99–123 (7th ed. 2016).

12 Some recent scholarship fails to account for this evolution, treating the current terms of the debate as if they were universal. See, e.g., Philip Hamburger, Is Administrative Law Unlawful? (2014).

13See Byron E. Shafer, The American Political Pattern: Stability and Change, 1932–2016 (2016).

14Cass R. Sunstein, The Cost-Benefit State: The Future of Regulatory Protection ix (2002).

15See Richard H. Pildes, Institutional Formalism and Realism in Constitutional and Public Law, 2013 S. Ct. Rev. 1, 2 (2014) (“[Institutional] realism would entail constitutional and public law doctrines that penetrate the institutional black box and adapt legal doctrine to take account of how these institutions actually function in, and over, time.”). The current Supreme Court may be disposed toward the alternative “institutional formalism,” tinged with an ideological disposition to rein in the administrative state. See Gillian E. Metzger, 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1, 17–33 (2016) (describing the “current judicial challenges to national administrative government”).

16 Throughout this Article we use the term “party systems” to refer to stable periods of partisan political alignment. The concept is explored in greater detail below. See infra Part II.A.1.

17 For perspectives on the use of complexity theory to understand system-level dynamics in politics, see generally Political Complexity: Nonlinear Models of Politics (Diana Richards ed., 2000). See also J.B. Ruhl, The Fitness of Law: Using Complexity Theory to Describe the Evolution of Law and Society and Its Practical Meaning for Democracy, 49 Vand. L. Rev. 1407 (1996) (discussing intersection of complexity theory and law).

18Compare Wilson, supra note 1, at 201 (Wilson advocated for a separate science of administration, apart from politics: “This is why there should be a science of administration which shall seek to straighten the paths of government, to make its business less unbusinesslike, to strengthen and purify its organization, and to crown its duties with dutifulness.”), with Waldo, supra note 2, at 91 (“Nothing is more central in thinking about public administration than the nature and interrelations of politics and administration. Nor are the nature and interrelations of politics and administration matters only for academic theorizing. What is more important in the day-to-day, year-to-year, decade-to-decade operation of government than the ways in which politics and administration are conceptualized, rationalized, and related one to the other.”).

19See, e.g., Hamburger, supra note 12 (arguing that modern administrative law violates the common law Anglo-American tradition); Christopher Yoo & Steven Calabresi, The Unitary Executive: Presidential Power from Washington to Bush (2008); Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231 (1994) (arguing that the administrative state was unconstitutional from its creation as an originalist matter). There is also a substantially functionalist understanding of administrative law doctrine that focuses on how doctrines work in practice. See generally Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573 (1984) (arguing in favor of function-based understanding of separation of powers principles).

20 Matthew D. McCubbins, Roger G. Noll & Barry R. Weingast, Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 Va. L. Rev. 431, 432 (1989) (“[L]egislators see the choice of administrative structures and processes as important in assuring that agencies produce policy outcomes that legislators deem satisfactory.”).

21Id.

22See Randall L. Calvert, Mathew D. McCubbins & Barry R. Weingast, A Theory of Political Control and Agency Discretion, 33 Am. J. Pol. Sci. 588, 604 (1989); see also Brian D. Feinstein, Designing Executive Agencies for Congressional Influence, 69 Admin. L. Rev. 259, 266 (2017) (conducting an empirical assessment of how agency design features support ongoing congressional oversight).

23See Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 Colum. L. Rev. 1749, 1753 (2009) (“Rather, I credit the Court with an awareness of certain political facts in developing administrative law: Congress creates agencies with few constraints on their power and then seeks to control their decisionmaking, just as the President does. In essence, agencies are subject to two political principals. Furthermore, these principals may have divergent policy preferences and may seek agency decisions that move in different, even conflicting, directions. Under these circumstances, the Court might see its role as mediating the needs of both political branches for control of agency decisionmaking, consistent with separation of powers. In sincerely attending to procedural issues in connection with agency decisionmaking, the Court has attended to the needs of the legislative branch. It has rendered agency action more susceptible to ongoing congressional oversight ....”).

24 Terry M. Moe & Scott A. Wilson, Presidents and the Politics of Structure, 57 Law & Contemp. Probs. 1, 4 (1994).

25Id. at 3–4.

26See, e.g., Daphna Renan, The Law Presidents Make, 103 Va. L. Rev. 805, 813–14 (2017) (describing how different models of legal advice in the executive branch empower different actors).

27See David W. Rohde & Harold J. Spaeth, Supreme Court Decision Making 140–145 (1976); Jeffrey A. Segal & Harold J. Spaeth, The Supreme Court and the Attitudinal Model 65 (1993); Harold J. Spaeth, The Attitudinal Model, in Contemplating Courts 296 (Lee Epstein ed., 1995); see also Charles Grove Haines, General Observations on the Effects of Personal, Political, and Economic Influences in the Decisions of Judges, 17 Ill. L. Rev. 96, 114 (1922). See generally Lee Esptein, William M. Landes & Richard A. Posner, The Behavior of Federal Judges 27 (2013).

28 Chief Justice John Roberts famously said of his position: “my job is to call balls and strikes ....” See Chief Justice Roberts Statement—Nomination Process, U.S. Courts, http://www.uscourts.gov/educational-resources/educational-activities/chief-justice-roberts-statement-nomination-process [https://perma.cc/E63D-AYTG].

29 Antonin Scalia, Regulatory Reform—The Game Has Changed, 4 Regulation 13, 13–14 (1981).

30Id. at 13.

31See United States v. Mead Corp., 553 U.S. 218, 256–57 (2001) (Scalia, J., dissenting) (“To decide the present case, I would adhere to the original formulation of Chevron.... We accordingly presume—and our precedents have made clear to Congress that we presume—that, absent some clear textual indication to the contrary, Congress, when it left ambiguity in a statute meant for implementation by an agency, understood that the ambiguity would be resolved, first and foremost, by the agency, and desired the agency (rather than the courts) to possess whatever degree of discretion the ambiguity allows. Chevron sets forth an across-the-board presumption, which operates as a background rule of law against which Congress legislates: Ambiguity means Congress intended agency discretion. Any resolution of the ambiguity by the administering agency that is authoritative—that represents the official position of the agency—must be accepted by the courts if it is reasonable.” (citations omitted)).

32See, e.g., Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke L.J. 511, 516 (1989).

33 Levinson & Pildes, supra note 3, at 2339.

34 Bulman-Pozen, supra note 3, at 1080 (arguing that national polarized politics interacts with federalism).

35Id.; see also Jessica Bulman-Pozen & Heather Gerken, Uncooperative Federalism, 118 Yale L.J. 1256, 1267 (2009) (citing Larry Kramer, Understanding Federalism, 47 Vand. L. Rev. 1485, 1544 (1994)).

36 Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. Rev. 459, 462 (2008).

37 McGarity, supra note 3, at 1678.

38 Eric A. Posner & Cass R. Sunstein, Institutional Flip-Flops, 94 Tex. L. Rev. 485, 486 (2016).

39 Matthew S. Broker & Michael A. Livermore, Centralizing Congressional Oversight, 32 J.L. & Pol. 261, 261 (2017); Livermore supra note 3, at 47.

40 Daniel A. Farber & Anne Joseph O’Connell, The Lost World of Administrative Law, 92 Tex. L. Rev. 1137, 1160 (2014).

41See, e.g., Nicholas R. Parrillo, The Endgame of Administrative Law: Governmental Disobedience and the Judicial Contempt Power, 131 Harv. L. Rev. 685, 687 (2018).

42See, e.g., Shafer, supra note 13, at xii. The transitions between party systems are often accompanied by “realignment elections” that demarcate when an old order has given way to a new. See, e.g., Alan I. Abramowitz & Kyle L. Saunders, Ideological Realignment in the U.S. Electorate, 60 J. Pol. 634, 636 (1998) (applying the theory to the 1994 and 1996 elections and finding that they resulted a realignment of party loyalties). See generally James L. Sundquist, Dynamics of the Party System: Alignment and Realignment of Political Parties in the United States 35–36 (1983); V.O. Key, Jr., A Theory of Critical Elections, 17 J. Pol. 3, 3–4 (1955). Roosevelt’s 1932 victory is the paradigmatic case of a realignment election. When political shifts (and accompanying legal changes) are sufficiently tectonic, they might even be referred to as “constitutional moments.” Bruce Ackerman, We the People: Foundations 58 (1991) (describing Reconstruction and the New Deal as the second and third constitutional moments). The paradigm of party systems and realignment elections has been challenged as failing to capture the nuances of partisan change. See David R. Mayhew, Electoral Realignments: A Critique of an American Genre 12 (2002). Although these critiques may have merit, we stick with the language of party systems and realignments for expositional reasons, understanding that the reality may not be as pat.

43 Parties can also be understood beyond their impact on the electoral process, such as through their associational features. See, e.g., Tashjian v. Republican Party of Conn., 479 U.S. 208, 225 (1986) (finding that political parties have associational rights under the First Amendment that limit state regulation). Parties can also be understood as purely endogenous institutions that serve as tools for elected officials. See generally John Aldrich, Why Parties? The Origin and Transformation of Political Parties in America 28–29 (1995). For our purposes, however, political parties are important for their electoral consequences and the subsequent substantive conflicts that emerge from their representation in office. For a broader study of the many roles played by the modern American political party, see generally Marjorie Randon Hershey, Party Politics in America 6–10 (17th ed. 2017) (defining parties as including “the party organization,” “the party in government” and “the party in the electorate”). As Justice Reed explained, “... political parties ... were created by necessity, by the need to organize the rapidly increasing population, scattered over our Land, so as to coordinate efforts to secure needed legislation and oppose that deemed undesirable.” Ray v. Blair, 343 U.S. 214, 220–21 (1952).

44See also Kenneth Benoit, Duverger’s Law and the Study of Electoral Systems, 4 French Pol. 69, 70 (2006) (stating Duverger’s Law as: “The simple-majority single-ballot system favours the two-party system”). See generally Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State 207–08 (Barbara North & Robert North trans., 1954).

45 William Nisbet Chambers, Party Development and the American Mainstream, in The American Party Systems: Stages of Political Development, 3-4 ( William Nisbet Chambers & Walter Dean Burnham eds., 2d ed. 1967).

46See Shafer, supra note 13, at xii.

47Id.

48 Chambers, supra note 45, at 3.

49Id.

50See, e.g., Bulman-Pozen, supra note 3, at 1080, 1108 (arguing that both the state and federal governments are “sites of partisan affiliation,” such that voters are likely to identify more with a state government when the federal government is controlled by an opposing party).

51David R. Mayhew, Divided We Govern: Party Control, Lawmaking, and Investigations, 1946-2002, 4-5 (2005).

52 The concept of realignment elections has played a prominent role in political science since the publication of V.O. Key, A Theory of Critical Elections, 17 J. Pol. 3 (1955). See also Walter Dean Burnham, Critical Elections and the Mainsprings of American Politics 2–3 (1970) (providing an early account of critical realignment elections). See generally Mayhew, supra note 42, at 7–12 (summarizing the subsequent literature in this area). A similar dynamic is also observed in the policy process, where “political processes are characterized by stability and incrementalism, but occasionally ... produce large-scale departures from the past.” James L. True, Bryan D. Jones & Frank R. Baumgartner, Punctuated Equilibrium Theory: Explaining Stability and Change in Policymaking, in Theories of the Policy Process 155 (Paul A. Sabatier ed., 2007). Mayhew is critical of the concepts of realigning elections and critical elections as oversimplified and inaccurate. To Mayhew, these ideas fail because specific empirical claims that result from the theory concerning how long party systems last, whether critical elections look different from other elections, or whether specific policy outcomes result from critical elections, sometimes lack support. Mayhew, supra note 42, at 142. Although interesting, Mayhew’s critiques are largely orthogonal to how party systems are understood in this Article, as sustained periods of stable electoral coalitions, which in turn coincide with stable periods of administration. The empirical claims rejected by Mayhew are not required for this more modest use of the theory.

53 The cyclical patterns of stability and realignment that characterize party systems can exist alongside other patterns of change, cyclical and otherwise. Political scientist Stephen Skowronek, for example, has argued for a cyclical notion of “political time” that is tied to the presidential leadership. Stephen Skowronek, Presidential Leadership in Political Time: Reprise and Reappraisal x (2011). There are also well-recognized secular trends that have endured across party systems, such as the growing reach and complexity of the administrative state. American politics is also, of course, embedded within broader historical realities involving, for example, race, gender, culture, globalization, and technological development. See, e.g., Theda Skocpol, Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (1992) (examining intersection of gender and politics during origin period of U.S. welfare state). Our examination of the relationships between party systems and administrative law is not meant to endorse any particular cleavage between law and society, but rather to emphasize an underappreciated set of interactions within the broader interpenetration of law and society. See William J. Novak, Response: The People’s Welfare Redux, 57 Am. J. Legal Hist. 248, 249 (2017) (“[T]he separation of law from politics and both from society and/or economy [is] an obstacle to historical understanding.”).

54 The election of 1968, for instance, has led to scholarly disagreement. See generally Arthur C. Paulson, Electoral Realignment and the Outlook for American Democracy 6–9 (2007) (surveying the academic disagreement on whether or not the 1968 election constituted a realignment).

55 Chambers, supra note 45, at 7.

56 Periodization raises its own set of issues concerning where to draw lines, and more generally, the criteria to use when engage in the line-drawing exercise. For an alternative to the party system approach, see Joel H. Silbey, The American Political Nation, 1838–1893, 6 (1991) (offering alternative periodization of political eras based on “other constituents of the political nation besides electoral behavior—the nature of political leadership, the importance of particular political institutions, the strength of government, popular attitudes toward politics, and the like”). We maintain the party system approach in part because of its familiarity and, more so, because it tracks features of the political environment that are particularly important for administrative law. That said, we also recognize that the party system framework might miss elements of political context—such as changes popular attitudes toward the role of experts—that are important to administrative law and may not be reflected in the structure of organized politics. For the origins of the “one damn thing after another” quote, see Quote Investigator, https://quoteinvestigator.com/2015/09/02/life-one/ (last visited July 6, 2019).

57See Shafer, supra note 13, at xii.

58 For an account of the demand of the rule of law on administrative agencies, see Kevin M. Stack, An Administrative Jurisprudence, 115 Colum. L. Rev. 1985 (2015).

59See Charles L. Barzun, Three Forms of Legal Pragmatism, 95 Wash. U. L. Rev. 1003, 1004 (2018). Accounting for the organization of politics when shaping administrative law doctrine is not the same, and should not be understood as, judges using their position to construct law that favors a particular party program (Justice Scalia’s example notwithstanding). Rather, judges can take a pragmatic perspective that encompasses both the deep values embedded in the law as well as the political and administrative landscape in which their decisions will be given effect.

60Shafer, supra note 13, at 122–23 (“Neither of these intendedly major changes, a new kind of party balance fostered by growing partisan independence or revised partisan majorities built on moving against ideological polarization, ever came close to realization. Someone had to win, and Bill Clinton did. Yet none of these putative strategic theories were of any use in predicting the veritable kaleidoscope of electoral outcomes that would follow.... The presence of all possible combinations of election outcomes and, even more strikingly, their rapid circulation and replacement, was what distinguished this period electorally from its predecessors.”).

61See Mark Tushnet, Politics as Rational Deliberation or Theater: A Response to Institutional Flip-Flops, 94 Tex. L. Rev. 82, 83 (2016) (arguing that “charges of flip-flopping [over inconsistent positions on questions such as presidential power] result from the specific circumstances of contemporary U.S. politics, with a hyperpartisan and ideologically polarized party system”).

62See, generally Richard Hofstadter. The Idea of a Party System: The Rise of Legitimate Opposition in the United States, 1780–1840 (1969).

63See generally Jerry Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012).

64See infra notes 110–115 and accompanying text.

65See generally Silbey, supra note 56.

66 The post-New Deal period generated a considerable amount of administrative law doctrine that is sufficiently vital that it continues to be taught in introductory administrative law courses. See Lawson, supra note 11 (categorizing the development of law as “Before the New Deal,” “The New Deal,” and “After the New Deal”).

67See generally Richard P. McCormick, The Second American Party System: Party Formation in the Jacksonian Era 3 (1966).

68See Mashaw, supra note 8, at 58–59 (describing the Washington Administration’s focus on finding administrators with “fitness of character” as the primary means of exercising internal discipline); id. at 175 (explaining the transition to a more bureaucratized approach to internal control during the Second Party System, when there was a much broader electorate).

69See generally George Dangerfield, The Era of Good Feelings (1989).

70See Mashaw, supra note 8, at 98, 126 (discussing embargo and public lands management, respectively).

71 Federal sovereign immunity still bars some claims against the government directly, and suits against officers are still common. Nonetheless, statutes like the Federal Tort Claims Act, the Tucker Act, and the Administrative Procedure Act now waive sovereign immunity in many contexts for which relief was barred in Early America. See generally Gregory C. Sisk, A Primer on the Doctrine of Federal Sovereign Immunity, 58 Okla. L. Rev. 439, 458 (2005) (documenting “the proliferation of statutory waivers of sovereign immunity”).

72 The legality of government conduct would often be tested by determining the validity of the authorization defense. This was the posture of classic cases like Little v. Barreme, where contests about inherent presidential authority and statutory interpretation arise in the context of personal actions against officers. 6 U.S. (2 Cranch) 170, 177 (1804). These cases could arise in state as well as federal courts. See, e.g., Olney v. Arnold, 3 U.S. (3 Dall.) 308, 308 (1796).

73Mashaw, supra note 8, at 114.

74Id. at 74.

75See generally William J. Novak, The People’s Welfare: Law and Regulation in Nineteenth-Century America 1 (1996).

76 Randy Beck, Promoting Executive Accountability Through Qui Tam Legislation, 21 Chap. L. Rev. 41, 44–45 (2018) (“Regulation of government officials through qui tam legislation was widely practiced in the American colonies and early states. Qui tam monitoring was used to promote statutory compliance by an enormous variety of state officials, particularly those performing decentralized functions such as road construction and maintenance, judicial administration, and regulation of commercial activities. It was common for early states to rely on qui tam oversight to ensure lawful conduct by officials performing functions critical to public confidence in government, such as conducting elections and collecting taxes.”).

77See Randy Beck, Qui Tam Litigation Against Federal Officials: A Neglected History, 93 Notre Dame L. Rev. 1235, 1294 (2018) (“Why did Congress authorize qui tam enforcement of two forfeitures directed at customs officials under the Collection Act when it opted not to use qui tam enforcement for other aspects of the legislation? The decision likely flowed from the difficult management challenge created by a large and dispersed federal workforce.”).

78See generally David N. Mayer, The Constitutional Thought of Thomas Jefferson 123 (1994) (examining Jefferson’s politics in comparison to Federalist and Whig viewpoints).

79See, e.g., Robert Remini, Martin Van Buren and Making of the Democratic Party 123 (1959) (detailing Van Buren’s attempt to build the Democratic Party during the period of increased suffrage before the election of Andrew Jackson).

80See Richard P. McCormick, The Second American Party System: Party Formation in the Jacksonian Era 14 (1966) (describing how the Democratic Party competed in different regions of the country).

81Silbey, supra note 56, at 46.

82See generally id. (discussing embrace of parties during this time).

83 There was a threefold increase in the absolute number of votes cast in the 1824 and 1828 presidential elections, primarily driven by a growth of voter turnout. Popular vote numbers are collected by the American Presidency Project, housed at U.C. Santa Barbara. See John Woolley & Gerhard Peters, The American Presidency Project, http://www.presidency.ucsb.edu. Population statistics are available at Bureau of the Census, U.S. Dep’t of Commerce, Historical Statistics of the United States: Colonial Times to 1970 (1975), https://www.census.gov/history/pdf/histstats-colonial-1970.pdf.

84See Mashaw, supra note 8, at 177–78.

85 Note, The History and Development of Qui Tam, 1972 Wash. U. L.Q. 81, 101 (1972) (“What emerges from this historical evolution is recognition that in America, as earlier in England, qui tam proceedings began as a useful and perhaps necessary supplement to the efforts at law enforcement of inadequate public agencies. As the public agencies became more effective [during the latter part of the nineteenth century], the need for qui tam actions diminished.”).

86 The intervening Civil War and Reconstruction were perhaps the most important remaking of American politics, and they both had transformative effects on the administrative state. See generally Bruce Ackerman, We the People: Foundations 81 (1991) (presenting the Civil War and the subsequent work of the Reconstruction Republicans as a constitutional moment). Control of the White House changed four times during this period. In addition, there were two elections in which the popular vote winner did not prevail in the Electoral College. See Woolley, supra note 83. This was driven by Democrats’ ability to attract new immigrants to their party. See Mashaw, supra note 8, at 229.

87Mashaw, supra note 8, at 233, 241–42. This period is also the starting point for Skowronek’s review of administrative state building. See generally Skowronek, supra note 10, at 39–45.

88Mashaw, supra note 8, at 239.

89Id. at 243, 257–58 (providing an example involving military pensions).

90 See James E. Campbell, Party Systems and Realignments in the United States, 1868–2004, 30 Soc. Sci. Hist. 359, 380 (2006) (finding that 1896 reflected a realignment of political power towards Republicans at the presidential and congressional levels).

91Republican Party, Encyclopedia Britannica, https://www.britannica.com/topic/Republican-Party (last visited Aug. 12, 2019). The only deviation during this period was the election of Democrat Woodrow Wilson, who won with 42% over a fractured Republican field featuring a third-party challenge from Teddy Roosevelt’s Progressive Party, which finished second. United States Presidential Election Results, Encyclopedia Britannica, https://www.britannica.com/topic/United-States-Presidential-Election-Results-1788863 (last visited Aug. 17, 2019). The only real interruption of congressional control came during World War I, after which Republicans once again gained control of Congress in the fourteen years leading up to the Great Depression. Party Divisions of the House of Representatives, 1789 to Present, History, Art & Archives: U.S. House of Representatives, https://history.house.gov/Institution/Party-Divisions/Party-Divisions/ (last visited Aug. 17, 2019).

92See R. Hal Williams, Realigning America: Williams, Bryan, and the Remarkable Election of 1896, 11 (2010).

93See Robert L. Rabin, Federal Regulation in Historic Perspective, 38 Stan. L. Rev. 1189, 1216–29 (1986).

94 For an important overview of bureaucrat development of the time, see Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies 1862–1928, 7 (2001). For other foundational work, see Skowronek, supra note 10, at 39–45.

95 The passage of the Interstate Commerce Act is often seen as the start of “traditional” administrative law. See Stephen Skowronek, supra note 10, at 121 (beginning his analysis with 1877 with a discussion of the Interstate Commerce Act); see also Rabin, supra note 93, at 1189 (“[W]hen Congress established the Interstate Commerce Commission, it initiated a new epoch in responsibilities of federal government. For the first time, a national legislative scheme was enacted that provided for wide-ranging regulatory controls over an industry that was vital to the nation’s economy—the railroads.”). The Act itself was new in how it “defines the structural position of administrative agencies within the governmental system, specifies the decisional procedures those agencies must follow, and determines the availability and scope of review of their actions by the independent judiciary.” Richard B. Stewart, Administrative Law in the Twenty-First Century, 78 N.Y.U. L. Rev. 437, 438 (2003).

96 27 Stat. 403 (1883). The growing concerns regarding excess by public servants led to a shift away from bounties and towards official salaries as well. See Nicholas Parrillo, Against the Profit Motive: The Salary Revolution in American Government 1780–1940, 17 (2013).

97 26 Stat. 209 (1890).

98Skowronek, supra note 10, at 175.

99Ronald N. Johnson & Gary D. Libecap, The Federal Civil Service System and the Problems of Bureaucracy 12 (1994) (“[I]n the late nineteenth century, with the enthusiastic support of the president Congress voted to restrict the number of patronage positions that were available. With the enactment of the Pendleton Act (22 Stat. 403) on 16 January 1883, the process was established by which patronage was to give way gradually to merit-based employment. By 1904, only twenty-one years after the Pendleton Act was passed, over 50 percent of the total federal civilian labor force was under merit provisions.”).

100 Rabin, supra note 93, at 1216–18.

101Id. at 1224–26.

102Skowronek, supra note 10, at 285–87 (describing the period as one of “Modern American state-building”). To Skowronek, the system that emerged in this period was in direct contrast to the prior “state of courts and parties,” supported by “an intellectual vanguard” that “articulated the limitations” of the old system. Id. at 286.

103See Mashaw, supra note 8, at 4–9.

104 Rabin, supra note 93, at 1219–20 (summarizing the competing view of Progressive Era and concluding that “neither position anticipated government intervention of the kind that was to result from the New Deal perception of wholesale market malfunction”).

105Shafer, supra note 13, at 7.

106Id. at 6 (noting that the “elections of 1934 and 1936 were to be the only ones in American history in which Northern Democrats constituted a majority of the House of Representatives”).

107 Legislation from the first hundred days included the Federal Securities Act, Pub. L. No. 73-22, 48 Stat. 74 (1933); Agricultural Adjustment Act, Pub. L. No. 73-10, 48 Stat. 31 (1933); Glass-Steagall Act, Pub. L. No 73-66, 48 Stat. 612 (1933) (Glass-Steagall is part of the broader U.S.A. Banking Act); and the National Industrial Recovery Act, Pub. L. No. 73-66, 48 Stat. 195 (1933). The second wave of legislation included the Banking Act of 1935 and the Social Security Act, Pub. L. No. 74-271, 49 Stat. 620 (1935). For a general discussion of these two waves, see Shafer, supra note 13, at 12, 16 (collecting legislation).

108 Almost all of these legislative initiatives began in the White House itself, rather than in the congressional committees. Id. at 16.

109 Agencies like the Tennessee Valley Authority (TVA) exemplify how these two aims coexisted. Created as a public corporation at the start of the New Deal program in 1933, the TVA pursued an agenda of improving resource management along the Tennessee River and Mississippi River Basins, while also serving as a source of short-term jobs for millions of unemployed Americans. Tennessee Valley Authority Act, Pub. L. No. 73-17 (1933). Much like Jackson’s spoils system or civil service system of the 1870s, federal administrative employment itself once again became a forum for policymaking, not just a means to an end.

110See Mistretta v. United States, 488 U.S. 361, 371–72 (1989) (discussing the history of the non-delegation doctrine before and after 1935). In 1935, the Court had invalidated two provisions on non-delegation grounds. See A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495, 529, 542 (1935) (invalidating a provision of the National Industrial Recovery Act allowing the president to designate “codes of fair competition”); Panama Refining Co. v. Ryan, 293 U.S. 388, 414–20 (1935) (invalidating a provision of the National Industrial Recovery Act that authorized the President to prohibit petroleum commerce).

111See Wickard v. Filburn, 317 U.S. 111 (1942); NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1, 43 (1937).

112 Kagan, supra note 3, at 2365–66 (noting that the two statutes invalidated on non-delegation grounds involved delegations directly to the President).

113 Stewart, supra note 2, at 1677.

114See W. Coast Hotel Co. v. Parrish, 300 U.S. 379, 391 (1937). While this moment is frequently framed in “external terms,” as the Court responding to political pressure, Barry Cushman has offered an “internal” account, relying on the “legal intellectual dimension” of the shift in doctrine. See Barry Cushman, Rethinking the New Deal Court, 80 Va. L. Rev. 201, 207 (1994).

115 These changes were supported in a new scholarly movement in favor of expert regulation, reflected in James Landis’ Administrative Process, published in 1938. James M. Landis, The Administrative Process (1938). Defending the administrative state against its detractors, Landis argued that the new administrative process was a superior means of exercising both judicial and legislative functions and offered a way to “modernize” the separation of powers. This view was particularly skeptical of the courts, with Landis noting that “judicial review over administrative action gives a sense of battle” leading courts to “thwart the effects of legislative judgments ... under the guise of constitutional and statutory interpretation.” James Landis, Administrative Policies and the Courts, 47 Yale L.J. 519, 519 (1937). His explanation for why administrative agencies were superior to courts not only included the familiar description of expertise and practicality, but also an assault on the “judicial conservatism” that was, in his account, ill-suited to the demands of the modern economy. James Landis, The Place of Administrative Law, 13 Conn. Bar J. 71, 72–77 (1939). The new administrative state was also buoyed by legal doctrine, developed during the New Deal period, that allowed for statutory independence from presidential control. See Wiener v. United States, 357 U.S. 349 (1958); Humphrey’s Ex’r v. United States, 295 U.S. 602, 631–32 (1935).

116Sidney M. Milkis, The President and the Parties: The Transformation of the American Party System Since the New Deal 149–161 (1993).

117See Harry J. Enten, Were Republicans Really the Party of Civil Rights in the 1960s?, Guardian (Aug. 28, 2013, 9:15 AM), https://www.theguardian.com/commentisfree/2013/aug/28/republicans-party-of-civil-rights (documenting the regional, rather than partisan, difference in the vote).

118 Pub. L. No. 79-404, 60 Stat. 237 (1946).

119 Stewart, supra note 2, at 1684.

120Id. at 1761–76.

121Shafer, supra note 13, at 72–74. Writing at the time, Morris Fiorina argued that this pattern of consistent divided government signaled that the idea of realignment was “a dead concept.” See Morris P. Fiorina, An Era of Divided Government, in Developments in American Politics 324 (G. Peele et al. eds., 1992). He also noted that the era from 1968 to 1988 was a historical outlier for the prevalence of divided government. Id. at 325–26.

122See Lawrence C. Dodd & Bruce I. Oppenheimer, Congress Reconsidered 94–97 figs.4-4 to 4-8 (11th ed. 2016) (empirically demonstrating increased polarization throughout the period and noting that “survey data confirm that the congressional parties represent increasingly distinctive sets of voters.”); Shafer, supra note 13, at 94–98 (describing the increased ideological polarization of party activists).

123See Mark Seidenfeld, A Big Picture Approach to Presidential Influence on Agency Policy-Making, 80 Iowa L. Rev. 1 (1994) (discussing justifications for growth of presidential power over agencies); Sidney A. Shapiro, Administrative Law After the Counter-Reformation: Restoring Faith in Pragmatic Government, 48 U. Kan. L. Rev. 689, 707 (2000) (“Reformers ... seek[ing] to combat government failure ... initially chose presidential oversight as the principal procedural vehicle for reform, and the Reagan and Bush Administrations put these plans into effect.”).

124 For example, the Occupational Safety and Health Act, Pub. L. No. 91-596 (1970), passed with 83 votes in the Senate (35 of them Republican) and 310 votes in the House (138 of them Republican). See Votes to Pass S. 2193 (91st Congress), GovTrack, https://www.govtrack.us/congress/votes/91-1970/s599 (last visited Aug. 12, 2019) (showing the vote count in the Senate); Votes to Adopt the Conference Report on S. 2193 (91st Congress), GovTrack, https://www.govtrack.us/congress/votes/91-1970/h419 (last visited Aug. 12, 2019). The Clean Water Act, Pub. L. No. 92-500 (1972) was passed by bipartisan majorities following a veto by President Nixon. See Clean Water: Congress Overrides Presidential Veto, in CQ Almanac1972, at 11–17 (28th ed. 1973). Scholars of the period were struck with how divided government did not seem to undermine effectiveness. See generally Mayhew, supra note 51 (arguing that divided government was an effective system for developing public policy). The effectiveness of divided government during this period came from its stability. As Shafer put it, “cross-partisan and cross-institutional negotiations ... characterize[d] the policymaking process of the era of divided government [because the] particular and recurrent institutional split implied that the two parties could not hope to be rescued electorally, at least in the short run.” Shafer, supra note 13, at 104. The era of partisan volatility created very different incentives, a point we return to below.

125See supra Schafer note 13, at 72–74.

126 See Thomas W. Merrill, Capture Theory and the Courts: 1967-1983, 72 Chi.-Kent L. Rev. 1039, 1043 (1996) (“There was during the period [1967 to 1983] no loss of faith in activist government. But a key instrumentality of activist government—the administrative agency—came to be regarded as suffering from pathologies not shared by other governmental institutions such as legislatures or courts. The principal pathology emphasized during these years was ‘capture,’ meaning that agencies were regarded as being uniquely susceptible to domination by the industry they were charged with regulating.”); George Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3, 3 (1971).

127U.S. Dep’t of Justice, Attorney General’s Manual on the Administrative Procedure Act (1947). The manual was understood as embodying a hands-off role for the federal courts. Professor John Duffy described the manual as “a highly political document designed to minimize the impact of the new statute on executive agencies, shrewdly characterized the APA provisions governing judicial review as merely a ‘restatement’ and thereby invited courts and the bar to treat the Act as something less than a statute, as subservient to judge-made doctrine.” John F. Duffy, Administrative Common Law in Judicial Review, 77 Tex. L. Rev. 113, 119 (1998).

128 Stewart, supra note 2, at 1669.

129 This rethinking of the judicial role often played out in disagreements among the justices. In Abbott Laboratories v. Gardner, for instance, a majority of Court adopted a view of ripeness that allowed more private challenges to agency action. See 387 U.S. 136, 141 (1967) (finding that agency action could be ripe for judicial review prior to enforcement). Justice Fortas, long a political supporter of the New Deal Democratic Party, disagreed, advocating for adherence to the traditional view. See Toilet Goods Ass’n v. Gardner, 387 U.S. 167, 175 (1967) (Fortas, J., concurring in part and dissenting in part) (disagreeing with the holding in Abbott Laboratories and noting “that established principles of jurisprudence, solidly rooted in the constitutional structure of our Government, require that the courts should not intervene in the administrative process at this stage ....”).

130 Stewart, supra note 2, at 1683 (“[T]he exercise of agency discretion is inevitably seen as the essentially legislative process of adjusting the competing claims of various private interests affected by agency policy.”).

131Id. at 1682.

132See, e.g., Greater Bos. Television Corp. v. FCC, 444 F.2d 841, 851 (D.C. Cir. 1970) (“Its supervisory function calls on the court to intervene not merely in case of procedural inadequacies, or bypassing of the mandate in the legislative charter, but more broadly if the court becomes aware, especially from a combination of danger signals, that the agency has not really taken a ‘hard look’ at the salient problems, and has not genuinely engaged in reasons decision-making.”). Gary Lawson’s administrative law casebook provides a colorful account of this period on the D.C. Circuit; see Gary Lawson, Federal Administrative Law 357 & nn.28–31 (7th ed. 2016).

133See, e.g., Portland Cement Ass’n v. Ruckelshaus, 486 F.2d 375, 394 (D.C. Cir. 1975) (requiring that agencies make available information that was relied on during the rulemaking as part of “hard look” review); Auto. Parts & Accessories Ass’n v. Boyd, 407 F.2d 330, 338 (D.C. Cir. 1968) (noting that judicial review ensures “major issues of policy were ventilated” during rulemaking process); Scenic Hudson Pres. Conference v. FPC, 354 F.2d 608, 616 (2d Cir. 1965) (finding that parties had standing to sue because of their recreational interest in “the conservation of natural resources”). For a discussion of the expansion of standing throughout this period, see generally Stewart, supra note 2, at 1723–48. For a thorough description of the debate on the D.C. Circuit at the time, see generally Reuel Schiller, Rulemaking’s Promise: Administrative Law and Legal Culture in the 1960s and 1970s, 53 Admin. L. Rev. 1139 (2001). Whether expanding judicial review along these lines actually helped regulatory beneficiaries is a separate empirical question. See, e.g., Wendy Wagner, Revisiting the Impact of Judicial Review of Agency Rulemakings: An Empirical Investigation, 53 Wm. & Mary L. Rev. 1717, 1789 (2012) (concluding that judicial review is less influential than generally expected in the administrative law context); Melissa F. Wasserman, Deference Asymmetries: Distortions in the Evolution of Regulatory Law, 93 Tex. L. Rev. 625, 645–49 (2015) (discussing ways in which deference regimes may favor regulated industry).

134See Paul Verkuil, Judicial Review and Informal Rulemaking, 60 Va. L. Rev. 185, 186 (1974) (“Rulemaking is unique to the administrative process. And properly applied it represents the ultimate form of that process. Even though some adjudication occurs in the administrative decisionmaking model, the role assigned to administrative agencies is in many ways different from that assigned the judicial. Administrative agencies frequently must act prospectively to plan, organize, and set standards. The legislative branch recognizes these functions by accompanying most agencies’ mandates with grants of rulemaking power.”).

135See Richard P. Nathan, The Administrative Presidency 34–42 (1983) (describing the Nixon Administration’s increase of staff within the White House); see also Jim Tozzi, OIRA’s Formative Years: The Historical Record of Centralized Regulatory Review Preceding OIRA’s Founding, 63 Admin. L. Rev. 37 (2011) (chronicling the evolution of regulatory review).

136 Freedom of Information Act, 5 U.S.C. § 552 (2012); Federal Advisory Committee Act, 5 U.S.C. app. (2012); National Environmental Policy Act, 42 U.S.C. § 4321 (2012).

137 Donald P. Moynihan & Alasdair S. Roberts, The Triumph of Loyalty Over Competence: The Bush Administration and the Exhaustion of the Politicized Presidency, 70 Pub. Admin. Rev. 572, 573 (2010).

138See Thomas J. Weko, The Politicizing Presidency: The White House Personnel Office, 1948–1994, 96–97 (1995) (noting the importance of the screening process and of the appointees’ political beliefs).

139 Some form of regulatory review had been in place since Nixon, but Reagan substantially expanded the substantive and institutional strength of this requirement. Office of Mgm’t & Budget, Office of Information and Regulatory Affairs (OIRA) Q&A’s, White House (Nov. 2009), https://obamawhitehouse.archives.gov/omb/OIRA_QsandAs (“The issuance of Presidential regulatory principles, and the centralized review of draft regulations, has been an accepted part of regulatory development for 30 years in one form or another. This began with President Nixon’s ‘Quality of Life’ program, and continued in the 1970s with President Ford’s requirement in Executive Orders 11821 and 11949 for agencies to prepare inflation/economic impact statements and with President Carter’s Executive Order 12044 on ‘Improving Government Regulations.’”).

140 Cynthia R. Farina, Undoing the New Deal Through the New Presidentialism, 22 Harv. J.L. & Pub. Pol’y 227, 235 (1998).

141See James F. Blumstein, Regulatory Review by the Executive Office of the President: An Overview and Policy Analysis of Current Issues, 51 Duke L.J. 851, 855 (2001).

142See generally Patricia M. Wald, The D.C. Circuit: Here and Now, 55 Geo. Wash. L. Rev. 718 (1986) (discussing the changing makeup of the court and its emerging practices and jurisprudence).

143See Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 557 (1978).

144 Richard J. Pierce, Jr., Waiting for Vermont Yankee II, 57 Admin. L. Rev. 669, 673, 683 (2005) (noting practice of the First Circuit to maintain precedents eroded by Vermont Yankee).

145See, e.g., Air Transp. Ass’n of Am. v. Fed. Aviation Ass’n, 169 F.3d 1, 7 n.5 (D.C. Cir. 1999) (suggesting that prior ex-parte limitation was undermined by Vermont Yankee); Sierra Club v. Costle, 657 F.2d 298, 402 (D.C. Cir. 1981) (declining to extend prior decision limiting ex-parte communications in agency decision making);

146 Heckler v. Chaney, 470 U.S. 821, 831–32 (1985). See generally Cary Coglianese & Christopher S. Yoo, Introduction: The Bounds of Executive Discretion in the Regulatory State, 164 U. Pa. L. Rev. 1587, 1594 (2016) (discussing importance of the “power to defer”).

147 462 U.S. 919, 955–59 (1983).

148See E. Donald Elliott, INS v. Chadha: The Administrative Constitution, the Constitution, and the Legislative Veto, 1983 Sup. Ct. L. Rev. 125, 126 (1983) (describing the decision in Chadha as a “shock” following the courts’ prior balancing approach to the separation of powers); cf. Richard I. Goldsmith, INS. v. Chadha and the Nondelegation Doctrine: A Speculation, 35 Syracuse L. Rev. 749, 750–51 (1984) (wrongly predicting that the decision in Chadha would lead to subsequent development of the non-delegation doctrine). Although the initial decision shifted power to the executive, dynamically the elimination of the legislative veto may, paradoxically, have led to an increasingly assertive legislature, ultimately canceling out the executive-enhancing nature of the decision. See Edward H. Stiglitz, Unitary Innovations and Political Accountability, 99 Cornell L. Rev. 1133, 1154 (2014) (drawing lessons from history of the legislative veto at the state level).

149 Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837 (1984); Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mutual Auto. Ins. Co., 463 U.S. 29 (1983). By way of illustration, State Farm has been cited more than 700 times by the D.C. Circuit alone, while Chevron has nearly 1,400 citations by the Circuit.

150Chevron, 467 U.S. 837.

151See Jon F. Hale, The Making of the New Democrats, 110 Pol. Sci. Q. 207, 207 (1995) (“Clinton did not want to be regarded as a Democrat in the line of Michael Dukakis, Walter Mondale, Jimmy Carter, and George McGovern. The New Democrat rhetoric, however, was not simply campaign rhetoric concocted by Clinton’s coterie of talented campaign strategists. It stood for a substantively new Democratic approach to active government that had been developing since the early years of the Reagan administration.”).

152Compare Everett Carll Ladd, The 1992 Vote for President Clinton: Another Brittle Mandate?, 108 Pol. Sci. Q. 1, 1–2 (1993) (arguing that the 1992 election was a deviation, but not a realignment, noting that “little happen[ed] in the mix of party identification”), with Robin Toner, 1992 Elections: At Dawn of New Politics, Challenges for Both Parties, N.Y. Times, Nov. 5, 1992, at B1 (citing democratic sources who believed that the Republicans “‘electoral lock’ on the White House was ‘broken forever’”).

153See, e.g., Linda J. Blumberg et al., How Would Coverage, Federal Spending, and Private Premiums Change if the Federal Government Stopped Reimbursing Insurers for the ACA’s Cost-Sharing Reductions? 1–3 (2017) (discussing the implications of the Trump Administration’s decision regarding reimbursements for ACA exchange insurers, one of the many issues over which the new Republican administration departed from the prior administration in its implementation of the ACA).

154Shafer, supra note 13, at 121.

155Id. at 122.

1562018 Midterm Election Results, N.Y. Times (Nov. 27, 2018), https://www.nytimes.com/interactive/2018/us/elections/calendar-primary-results.html.

157Shafer, supra note 13, at 122.

158See, e.g., Robin Toner, Southern Democrats’ Decline is Eroding the Political Center, N.Y. Times (Nov. 15, 2004), https://www.nytimes.com/2004/11/15/politics/southern-democrats-decline-is-eroding-the-political-center.html.

159Shafer, supra note 13, at 141 (“[I]n … the Era of Partisan Volatility, the ideological dynamic that had captured the activist stratum in both parties broke through the previous firewall, capturing elected officialdom and fostering generalized polarization.”).

160Id. at 159. For instance, the ACA received zero Republican votes in either chamber, while the Trump Administration’s signature tax legislation received no Democratic votes.

161Id. at 160–61; see American Gridlock: The Sources, Character, and Impact of Political Polarization xxii (James E. Thurber & Antoine Yoshinaka eds., 2015) (listing as consensus views the proposition that “[t]he parties in Congress are as polarized … as at any time in history,” “[t]he fit between ideology and party is unusually strong,” and “[u]nder divided government and split chamber control, the current Congress has ceased to operate as an effective legislative body.”). The pieces in this collection go on to explore particular factors reinforcing this polarization, like changes to the media and political redistricting. See, e.g., Micah Altman & Michael McDonald, Redistricting and Polarization, in American Gridlock: The Sources, Character, and Impact of Political Polarization 45 (James E. Thurber & Antoine Yoshinaka eds., 2015).

162See Hershey, supra note 43, at 6; see also Tashjian v. Republican Party of Conn., 479 U.S. 208, 224–25 (1986) (recognizing associational rights of political parties as private associations); Aldrich, supra note 43, at 4–6 (adopting a more elite-driven and insider-driven understanding of political parties; Donald Green et al., Partisan Hearts and Minds: Political Parties and the Social Identities of Voters 4 (2002) (“To be sure, party issue positions have something to do with the attractiveness of partisan labels …. But causality also flows in the other direction: When people feel a sense of belonging to a given social group, they absorb the doctrinal positions the group advocates.”

163See Julia Azari, Weak Parties and Strong Partisanship are a Bad Combination, Vox (Nov. 3, 2016, 4:40 PM), https://www.vox.com/mischiefs-of-faction/2016/11/3/13512362/weak-parties-strong-partisanship-bad-combination.

164Marty Cohen et al., The Party Decides: Presidential Nominations Before and After Reform 7 (2008).

165Id. This work defines political parties in a broad way, arguing that the highly involved activists function as a single political party, even without formal affiliations with the national organization.

166Id. at 5–6.

167See id. at 5, 187–88.

168See Harry Enten, The Four Things I Learned from the Donald Trump Primary, FiveThirtyEight (May 6, 2016, 4:55 PM), https://fivethirtyeight.com/features/the-four-things-i-learned-from-the-donald-trump-primary/.

169 Samuel Issacharoff, Outsourcing Politics: The Hostile Takeover of Our Hollowed-Out Political Parties, 54 Hous. L. Rev. 845, 862–64 (2017) (summarizing the Court’s decisions in FEC v. Colo. Republican Fed. Campaign Comm. (Colorado Republican II), 533 U.S. 431, 455 (2001) and Colo. Republican Fed. Campaign Comm. v. FEC (Colorado Republican I), 518 U.S. 604, 614 (1996)).

170 Examples include Christine O’Donnell, Ken Buck, Roy Moore, and Todd Akin.

171See, e.g., Edward-Isaac Dovere, Why Tom Steyer Doesn’t Care What Nancy Pelosi Thinks, Politico (May 29, 2018), https://www.politico.com/magazine/story/2018/05/29/trump-impeachment-democrats-steyer-pelosi-218548.

172See Julia Azari, Weak Parties and Strong Partisanship Are a Bad Combination, Vox (Nov. 3, 2016, 4:40 PM), https://www.vox.com/mischiefs-of-faction/2016/11/3/13512362/weak-parties-strong-partisanship-bad-combination.

173See id. (focusing on the collapse of political norms that results from weak institutional parties); see also Issacharoff, supra note 169, at 879–80.

174Hershey, supra note 43, at 11–12. See generally Gary W. Cox & Mathew D. McCubbins, Legislative Leviathan: Party Government in the House (2d ed. 2007).

175Richard L. Revesz & Michael A. Livermore, Retaking Rationality: How Cost-Benefit Analysis Can Better Protect the Environment and Our Health 21–30 (2008).

176See, e.g., Devins & Lewis, supra note 36, at 480–83 (describing the growth in the nomination of party loyalists to independent agency positions since the 1980s).

177See Kagan, supra note 3, at 2249–50.

178See Livermore, supra note 3, at 73–80.

179Id. at 91–92.

180Cf. Lindsay McPherson, 2018 Republican Agenda Not What Lawmakers Envisioned, Roll Call (Feb. 5, 2018, 5:04 AM), https://www.rollcall.com/news/politics/2018-republican-agenda-not-lawmakers-initially-envisioned (noting the role of intraparty disputes in limiting the legislative agenda).

181 Max Boot, Trump Security Strategy a Study in Contrasts, CFR (Dec. 18, 2017), https://www.cfr.org/expert-brief/trump-security-strategy-study-contrasts.

182 The diffusion of authority to political appointees at agencies has been accompanied by a sidelining of civil service experts. Daniel A. Farber, Presidential Administration Under Trump 3–4 (Aug. 8, 2017) (unpublished research paper) (on file with U.C. Berkeley School of Law), https://ssrn.com/abstract=3015591 (“[R]ather than deferring to agency experts, the Administration has often cut them out of the loop and has shown itself hostile in important ways to traditional forms of expertise.”)

183 See Kathryn Dunn Tenpas, Record-Setting White House Staff Turnover Continues with News of Counsel’s Departure, Brookings Inst. (Oct. 19, 2018), https://www.brookings.edu/blog/fixgov/2018/10/19/record-setting-white-house-turnover-continues-with-news-of-counsels-departure/) (“Much like turnover within the larger sample of senior White House staff, President Trump is breaking records.”); Kathryn Dunn Tenpas, With the Revelation of Marc Short’s Impending Departure, President Trump Has Lost the Vast Majority of Tier 1 Staff Members, Brookings Inst. (June 27, 2018), https://www.brookings.edu/blog/fixgov/2018 /06/27/trump-has-lost-the-vast-majority-of-tier-one-staff-members/ (describing record levels of turnover in the Trump Administration for the twelve most senior positions within the White House).

184 Amber Phillips, That Confirmation Hearing, and Three Other Reasons Betsy DeVos’s Nomination Is on the Rocks, Wash. Post (Feb. 2, 2017), https://www.washingtonpost.com/news/the-fix/wp/2017/02/02/that-confirmation-hearing-and-3-other-reasons-betsy-devoss-nomination-is-on-the-rocks/?utm_term=.ae066a6face9.

185 For a description of how decentralization can lead to greater ideological separation in the context of congressional oversight, see Broker, supra note 39, at 270–73.

186 Leonard Leo, the Executive Vice President of the Federalist Society, served as President Trump’s judicial adviser during both Supreme Court confirmations. See Jay Michaelson, The Secrets of Leonard Leo, the Man Behind Trump’s Supreme Court Pick, Daily Beast (July 9, 2018, 5:16 AM), https://www.thedailybeast.com/the-secrets-of-leonard-leo-the-man-behind-trumps-supreme-court-pick.

187 Farber, supra note 182, at 19.

188Id. at 28.

189See generally Aldrich, supra note 43.

190See supra Part II.

191Frances E. Lee, Insecure Majorities: Congress and the Perpetual Campaign 1 (2016).

192See generally id. at 1–3 (arguing that the electoral parity between the two major parties contributes to gridlock because each party has a plausible chance of winning the next election, and accordingly has less of an incentive to cooperate).

193See, e.g., James L. Sundquist, Needed: A Political Theory for the New Era of Coalition Government in the United States, 103 Pol. Sci. Q. 613, 617 (1988) (“As soon as political science emerged as a scholarly discipline, its adherents began to pronounce and elaborate the theoretical foundation of the system of party government…. [T]he scholars concluded that [parties] were useful and necessary … in unifying a government of dispersed powers and thereby making it more effective.”).

194Mayhew, supra note 51, at 3–5.

195Id. at 79–80.

196Id. at 78, 85–87.

197Id. at 3-4.

198See Lee, supra note 191, at 1–2, 5 (describing the differing incentives for cooperation during the era of divided government and the current era).

199 Sundquist, supra note 193, at 613.

200Id.

201Id. at 627.

202 Alex Pareene, Political Power Never Lasts. Democrats Need To Use Theirs While They Have It, Wash. Post (Nov. 9, 2018), https://www.washingtonpost.com/outlook/political-power-never-lasts-democrats-need-to-use-theirs-while-they-have-it/2018/11/09/f5f050d2-e396-11e8-8f5f-a55347f48762_story.html?noredirect=on&utm_term=.753dbf326258 (summarizing legislation passed largely by Democrats with unified government in 2009 and 2010).

203 Even during periods of unified government, legislation may still be thwarted by the increasingly aggressive use of the filibuster. The willingness to use the filibuster to block legislation has grown over time. As a result, major legislation has often been passed through “reconciliation bills,” which only require a simple majority. Both the Affordable Care Act and the Tax Cuts and Jobs Act were passed this way. These reconciliation bills, however, are limited to budget-related legislation and are capped at three per year. See generally Megan S. Lynch & James V. Saturno, Cong. Research Serv., R44058, The Budget-Reconciliation Process: Stages of Consideration (2017).

204Barbara Sinclair, Unorthodox Policymaking: New Legislative Processes in the U.S. Congress 64 (1997).

205Shafer, supra note 13, at 163–66.

206 Nicholas Bagley, Legal Limits and the Implementation of the Affordable Care Act, 164 U. Pa. L. Rev. 1715, 1716 (2016).

207 Pareene, supra note 202.

208 Ryan Lizza, Why John McCain Killed Obamacare Repeal—Again, New Yorker (Sept. 22, 2017). Notably, no Democrat supported the Obamacare repeal efforts, even though many represented states or districts that lean heavily Republican. This highly partisan result further supports the recent trends for major legislation during periods of unified government. Supra notes 191–192 and accompanying text.

209See Jack Beermann, Midnight Rules: A Reform Agenda, 2 Mich. J. Envtl. & Admin. L. 285, 333–34 (2013). Not surprisingly, the study of midnight rules began to gain force during the Clinton Administration, which was the first presidency of the era of partisan volatility. Id. at 291. While midnight rules are frequently critiqued, they may not be all that different in kind from rules promulgated during other periods in a President’s term, and therefore do not get significant focus here. See generally Edward H. Stiglitz, Unaccountable Midnight Rulemaking?: A Normatively Informative Assessment, 17 N.Y.U. J. Legis. & Pub. Pol’y 137 (2014) (empirically examining the practice of midnight rulemaking).

210See Nancy LeTourneau, Democrats Should Start Talking About Comprehensive Immigration Reform, Wash. Monthly (June 1, 2018) (“When it became clear that their attempts would be successful in blocking comprehensive immigration reform, President Obama signed DAPA, granting deferred action to parents of children who were citizens or lawful permanent residents.”); Editorial Board, Don’t Like EPA’s Power-Plant Plan? Complain to Congress, Bloomberg (Sept. 29, 2016), https://www.bloomberg.com/opinion/articles/2016-09-29/don-t-like-epa-s-power-plant-plan-complain-to-congress (“Congress left the Obama administration no choice. The only way to avert potentially disastrous climate change is to lower emissions as soon as possible. Congress’s refusal since 2010 to consider any climate policy at all forced the EPA to look for alternatives.”).

211Frances E. Lee, Insecure Majorities: Congress and the Perpetual Campaign 4 (2016).

212 Restoring Internet Freedom, 83 Fed. Reg. 7852, 7852 (Apr. 23, 2018).

213 Memorandum from Elaine C. Duke, Acting Sec’y, U.S. Dep’t of Homeland Sec., to James W. McCament, Acting Dir., U.S. Citizenship & Immigration Servs., et al. (Sept. 5, 2017) (on file with U.S. Dep’t of Homeland Sec.).

214 Repeal of Carbon Pollution Emissions Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 82 Fed. Reg. 51787 (proposed Nov. 8, 2017).

215 Julie Terkowitz, Trump Slashes Size of Bears Ears and Grand Staircase Monuments, N.Y. Times (Dec. 4, 2017), https://www.nytimes.com/2017/12/04/us/trump-bears-ears.html.

216 Tony Romm, Here’s How the FCC Plans To Defend Its Net Neutrality Repeal in Federal Court, Wash. Post (Oct. 11, 2018), https://www.washingtonpost.com/technology/2018/10/12/heres-how-fcc-plans-defend-its-net-neutrality-repeal-federal-court/?utm_term=.7bf9402a311e.

217Maeve P. Carey et al., Cong. Research Serv., R43992, The Congressional Review Act: Frequently Asked Questions 1 (2016). The CRA governs legislative procedure only; it is unlike the general management laws imposed on the executive to impose procedural obligations on agencies directly.

218Id. at 14.

219Id. at 12–13. Holdover appointees can create a context where Congress and the President can use the CRA effectively well into a new period of unified government.

220Id. at 5.

221 Stream Protection Rule, Pub. L. No. 115-5, 131 Stat. 10 (2017) (disapproving the Department of the Interior’s Stream Protection Rule).

222 Department of Education—Teacher Preparation Issues, Pub. L. No. 115-14, 131 Stat. 78 (2017) (disapproving the Department of Education’s teacher preparation rules).

223 Clarification of Employer’s Continuing Obligation, Pub. L. No. 115-21, 131 Stat. 87 (2017) (disapproving the Department of Labor’s regulation of reporting employee injury and illness).

224See Nina A. Mendelson, The Uncertain Effects of Senate Confirmation Delays in the Agencies, 64 Duke L.J. 1571, 1576, 1603–04 (2015).

225 Carbon Pollution Emissions Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 80 Fed. Reg. 64,661, 64,664 (Oct. 23, 2015) (to be codified at 40 C.F.R. pt. 60).

226Id. at 64,679–82.

227See Blumberg et al., supra note 153.

228 Michael Stratford, Trump and DeVos Fuel a For-Profit College Comeback, Politico (Aug. 31, 2017, 4:58 AM), https://www.politico.com/story/2017/08/31/devos-trump-forprofit-college-education-242193.

229 Michael A. Livermore, Cost-Benefit Analysis and Agency Independence, 81 U. Chi. L. Rev. 609, 612 (2014).

230Id. at 614; Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 844 (1984) (“Such legislative regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to statute.”).

231See supra note 1 and accompanying text.

232See infra note 238.

233See Exec. Order No. 12,291, 3 C.F.R. § 127 (1981), reprinted in 5 U.S.C. § 601 (1988), revoked by Exec. Order No. 12,866, 3 C.F.R. § 638 (1993), reprinted in 5 U.S.C. § 601 (2000).

234See generally Richard L. Revesz & Michael A. Livermore, Retaking Rationality: How Cost-Benefit Analysis Can Better Protect the Environment and Our Health 21–30 (2008) (discussing use of cost-benefit analysis during the Reagan and George H. W. Bush Administrations). The cost-benefit analysis standard also provided a means for agencies to use their relative substantive sophistication to protect their independence. See Livermore, supra note 229, at 613.

235Revesz & Livermore, supra note 234, at 31–32.

236 Michael A. Livermore & Richard L. Revesz, Regulatory Review, Capture, and Agency Inaction, 101 Geo. L.J. 1337, 1339 (2013).

237Id. at 1361–62. OIRA’s stabilizing role naturally generates criticism when it dials back activities that might otherwise be pursued at the agency level. See generally Lisa Heinzerling, Statutory Interpretation in the Era of OIRA, 33 Fordham Urb. L.J. 1097, 1117 (2006) (arguing that OIRA has usurped the appropriate role of agencies in interpreting statutes).

238See Jason Schwarz & Jeffrey Shrader, Muddying the Waters: How the Trump Administration Is Obscuring the Value of Wetlands Protection from the Clean Water Rule 1 (2017) (discussing inconsistent treatment of prior studies of regulatory costs and benefits); Caroline Cecot & Michael A. Livermore, The One-In, Two-Out Executive Order Is a Zero, 166 U. Penn. L. Rev. Online 1, 16 (2017); Jeremy Berg et al., Joint Statement on EPA Proposed Rule and Public Availability of Data, Science (May 4, 2018), https://science.sciencemag.org/content/360/6388/eaau0116.full (discussing efforts to ignore relevant scientific findings on benefits of pollution reduction); Richard L. Revesz, Pruitt Would Like Us to Ignore the Indirect Benefits of Environmental Regulations, Slate (June 13, 2018, 12:38 PM), https://slate.com/technology/2018/06/scott-pruitt-is-trying-to-undermine-environmental-regulation-in-a-creative-way.html (quoting former EPA Administrator’s promise to alter method of treating secondary effects of regulation).

239 Neomi Rao, Curriculum Vitae (2016) (on file with author).

240Id.

241Id.

242See, e.g., Neomi Rao, Administrative Collusion: How Delegation Diminishes the Collective Congress, 90 N.Y.U. L. Rev. 1463, 1525 (2015); Neomi Rao, Removal: Necessary and Sufficient for Presidential Control, 65 Ala. L. Rev. 1205, 1276 (2014).

243 Steve Eder, Neomi Rao, The Scholar Who Will Help Lead Trump’s Regulatory Overhaul, N.Y. Times (July 9, 2017), https://www.nytimes.com/2017/07/09/business/the-scholar-who-will-help-lead-trumps-assault-on-rules.html. Administrator Rao’s tenure recently ended when she was confirmed to serve as a judge on the D.C. Circuit.

244 5 U.S.C. § 706 (2012).

245See supra Part I.B.

246 Joan Biskupic, FDA Can’t Regulate Tobacco, Supreme Court Rules 5 to 4, Wash. Post (Mar. 22, 2000), https://www.washingtonpost.com/archive/politics/2000/03/22/fda-cant-regulate-tobacco-supreme-court-rules-5-to-4/3e3fd725-a469-4261-b660-a3415d2f99c1/?utm_term=.8364a7b7ab17 (describing how FDA action on tobacco was in response to failed legislative efforts).

247 FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 161 (2000).

248 Massachusetts v. EPA, 549 U.S. 497, 534 (2007) (“EPA has offered no reasoned explanation for its refusal to decide whether greenhouse gases cause or contribute to climate change. Its action was therefore ‘arbitrary, capricious, … or otherwise not in accordance with law.’”).

249 Texas v. United States, 809 F.3d 134, 178 (5th Cir. 2015), aff’d by an equally divided court, 136 S. Ct. 2271, 2272 (2016).

250 NAACP v. Trump, 298 F. Supp. 3d 209, 249 (D.D.C. 2018).

251 This ultimately happened to revised versions of the travel ban in June of 2018. See Trump v. Hawaii, 138 S. Ct. 2392, 2423 (2018).

252Judicial Vacancy List for December 2016, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2016/12/vacancies (last visited July 5, 2019). This figure was much higher than when Bush left office (42) or Clinton left office (67). Judicial Vacancy List for December 2008, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2008/12/vacancies (last visited July 5, 2019); Judicial Vacancy List for December 2000, U.S. Cts., https://www.uscourts.gov/judges-judgeships/judicial-vacancies/archive-judicial-vacancies/2000/12/vacancies (last visited July 5, 2019);See also Anne Joseph O’Connell, Shortening Agency and Judicial Vacancies Through Filibuster Reform? An Examination of Confirmation Rates and Delays from 1981 to 2014, 64 Duke L.J. 1645, 1660–61 (2015) (describing failed nominations by administration).

253 Indeed, there appears to be widespread agreement that the judicial confirmation process is becoming more politicized. David Russell, Politicization in the Federal Judiciary and Its Effect on the Federal Judicial Function, N.Y.U. J. Legis. & Pub. Pol’y Quorum 21, 26 (2018). Many commentators ascribed this trend to the Robert Bork nomination. Nina Totenberg, The Confirmation Process and the Public: To Know or Not to Know, 101 Harv. L. Rev. 1213, 1213 (1988); John M. Walker, Jr., The Unfortunate Politicization of Judicial Confirmation Hearings, Atlantic (July 9, 2012), https://www.theatlantic.com/national/archive/2012/07/the-unfortunate-politicization-of-judicial-confirmation-hearings/259445/. There is an active debate regarding whether increased politicization is necessarily a bad thing. Benjamin Wittes, Confirmation Wars: Preserving Independent Courts in Angry Times 76–77 (Peter Berkowitz & Tod Lindberg eds., 1st ed. 2006); Henry Paul Monaghan, The Confirmation Process: Law or Politics, 101 Harv. L. Rev. 1202, 1203 (1988).

254See Lara E. Chausow, R. Eric Peterson & Amber Hope Wilhelm, Cong. Research Serv., R43947, House of Representatives Staff Levels in Member, Committee, Leadership, and Other Offices, 1977-2016 (2016) (describing the shift from committee staff to individual member office support over time). This decline in staff began right at the beginning of the era of partisan volatility, with the Republican takeover of Congress in 1994. See Russell W. Mills & Jennifer L. Selin, Congressional Committee Staff Have Shrunk. Here Is One Way Congress Makes up the Difference, Wash. Post (June 14, 2017), https://www.washingtonpost.com/news/monkey-cage/wp/2017/06/14/congressional-committee-staffs-have-shrunk-heres-one-way-congress-makes-up-the-difference/?noredirect=on. Many commentators have noted the declining role of congressional committees and called for a rebalancing of the committee-leadership relationship. See, e.g., Kevin R. Kosar & Adam Chan, R Street Policy Study No. 66: A Case for Stronger Congressional Committees, R St. Inst. 1 (Aug. 2016) (“[T]he clearest path ahead may be to go backward, away from a hierarchical, leadership-dominated model of operating the chambers to one that disperses more power to committees.”).

255Kevin R. Kosar & Adam Chan, R Street Policy Study No. 66: A Case for Stronger Congressional Committees 1 (2016).

256 Curtlyn Kramer, Vital Stats: Congress Has a Staffing Problem Too, Brookings Inst.(May 24, 2017), https://www.brookings.edu/blog/fixgov/2017/05/24/vital-stats-congress-has-a-staffing-problem-too/.

257 These deep cuts also included the elimination of another Legislative Branch institution, the Office of Technology Assessment, which provided “nonpartisan scientific studies of policy decisions.” Ed O’Keefe, When Congress Wiped an Agency Off the Map, Wash. Post (Nov. 29, 2011), https://www.washingtonpost.com/blogs/federal-eye/post/when-congress-wiped-an-agency-off-the-map/2011/11/29/gIQAIt0J9N_blog.html.

258 Vann R. Newkirk II, The GOP Escalates Its Battle with the CBO, Atlantic (July 22, 2017), https://www.theatlantic.com/politics/archive/2017/07/republicans-cbo-hhs-score-bcra/534480/.

259Id.

260See generally Daniel J. Hopkins, The Increasingly United States: How and Why American Political Behavior Nationalized 227–28 (2018).

261See Bulman-Pozen, supra note 3, at 995.

262See Bulman-Pozen & Gerken, supra note 35, at 1267; see also Deil S. Wright, Understanding Intergovernmental Relations 49 (1978) (presenting a model of federalism premised on overlapping authority, in which the federal government was not in hierarchical relationship with state and local actors).

263See Bulman-Pozen & Gerken, supra note 35.

264See id. at 1264 (“[W]e argue that contestation can and does take place in many areas of federalism where states lack policymaking autonomy…. We do not quibble with the idea that sovereignty confers one sort of power, but we think it is a mistake to neglect the possibilities associated with a different sort of power – the power of the servant.”).

265See id. at 1278 (documenting state opposition to the Patriot Act based on state interpretations of the federal Constitution).

266 This decision was made easier for states following the Supreme Court’s holding that all existing Medicaid funds could not be conditional on the expansion under the ACA. See NFIB v. Sebelius, 567 U.S. 519, 585 (2012).

267See Claire McCusker, The Federalism Challenges of Impact Litigation by State and Local Government Actors, 118 Yale L.J. 1557, 1557 (2009).

268 Brief for Utah et al. as Amici Curiae Supporting Petitioners, Lucia v. SEC, 138 S. Ct. 2044 (2018) (No. 17-130).

269Id. at 1.

270 556 U.S. 502, 514 (2009).

271 The major questions doctrine can be traced to FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000). See Kevin O. Leske, Major Questions About the “Major Questions” Doctrine, 5 Mich. J. Envtl. & Admin. L. 479 (2016) (tracing the history of the doctrine).

272See Cary Coglianese, Moving Forward with Regulatory Lookback, 30 Yale J. on Reg. 57, 58–59 (2013) (describing regulatory “look-back” initiatives of the Obama Administration).

273 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 69 Fed. Reg. 22,122, 22,123 (Apr. 23, 2004) (to be codified at 29 C.F.R. pt. 541).

274 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 81 Fed. Reg. 32,391, 32,393 (May 23, 2016) (to be codified at 29 C.F.R. pt. 541).

275Id. at 32,392.

276Id. at 32,396 (explaining that “the FLSA’s overtime protections are a linchpin of the middle class”).

277See Katharine Q. Seelye, Bush White House Reverses Clinton Decision on Mining, N.Y. Times (Oct. 26, 2001), https://www.nytimes.com/2001/10/26/us/bush-white-house-reverses-clinton-decision-on-mining.html.

278 The Bush Administration’s rule concerning hours of service for commercial truckers was promulgated in 2008, after years of litigation challenging earlier iterations. See Hours of Service of Drivers, 73 Fed. Reg. 69567, 69568 (Nov. 19, 2008) (to be codified at 47 C.F.R. pt. 76). The new Obama-era rule was promulgated in 2011. See Federal Motor Carrier Safety Administration, Hours of Service of Drivers, 76 Fed. Reg. 81134, 81134 (Dec. 27, 2011) (to be codified at 47 C.F.R. pts. 385, 386, 390, 395).

279 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 514 (2009).

280Id. at 515.

281 As in many areas of law, the Justices’ proclivity to produce numerous concurring and dissenting opinions muddies the waters of Fox Television Stations somewhat. In particular, Justice Kennedy provided the deciding vote and wrote separately to note that he agreed with some of the arguments found in Justice Breyer’s dissent, which did urge more probing review of agency policy reversals. See Jeffrey S. Lubbers, A Guide to Federal Agency Rulemaking 509–13 (6th ed. 2018).

282See Encino Motorcars LLC v. Navarro, 136 S. Ct. 2117, 2125–26 (2016) (quoting Fox Television Stations, 556 U.S. at 515–16).

283Fox Television Stations, 556 U.S. at 547 (Breyer, J., dissenting).

284Id. at 541 (Stevens, J., dissenting).

285See id. The actual difference between the majority and dissenting opinions in Fox Television Stations is uncertain. The majority references the importance of “reliance interests,” but the rule explicitly declines to impose a higher burden for situations in which a prior decision created such interests. Some lower courts, however, have read the majority to arrive at a position comparable to that of the dissent. Judge Williams’ opinion in U.S. Telecommunications is such an example. U.S. Telecomm. Ass’n v. FCC, 825 F.3d 674, 744-45 (D.C. Cir. 2016). Both sides seem to agree that the reliance created by prior regulations is important; the Breyer dissent, however, would require a higher justification to overcome these interests.

286See supra notes 225–229 and accompanying text.

287See, e.g., Bauer v. DeVos, 325 F. Supp. 3d 74, 101 (D.D.C. 2018). The district court in Bauer, applying the analysis of the majority in Fox Television Stations, found that the agency has insufficiently explained its reasons for changing its view of the legality of regulations related to student loan borrowers. Id.

288See also Adrian Vermuele, Law’s Abnegation: From Law’s Empire to the Administrative State 160 (2016) (collecting cases since State Farm and finding that agencies often prevail under arbitrary and capricious review).

289Fox Television Stations, 556 U.S. at 549.

290 Justice Breyer illustrates the point: “An (imaginary) administrator explaining why he chose a policy that requires driving on the right side, rather than the left side, of the road might say, ‘Well, one side seemed as good as the other, so I flipped a coin.’ But even assuming the rationality of that explanation for an initial choice, that explanation is not at all rational if offered to explain why the administrator changed driving practice, from right side to left side, 25 years later.” Id.

291 Protecting and Promoting the Open Internet, 80 Fed. Reg. 19,737 (Apr. 13, 2015) (to be codified at 47 C.F.R. pts. 1, 8, 20).

292 The initial rule was invalidated by the D.C. Circuit, which found that the FCC could not proceed without reclassifying broadband providers under Title II of the Communications Act of 1934. Verizon v. FCC, 740 F.3d 623, 659 (D.C. Cir. 2014).

293 Protecting and Promoting the Open Internet, 80 Fed. Reg. at 19,740.

294Verizon, 740 F.3d at 643–44.

295 Protecting and Promoting the Open Internet, 80 Fed. Reg. at 19,742–43.

296Id. at 19,743.

297 U.S. Telecomm. Ass’n v. FCC, 825 F.3d 674, 706–07 (D.C. Cir. 2016).

298Id. at 745–46 (Williams, J., concurring in part and dissenting in part).

299Id. at 748.

300 Restoring Internet Freedom, FCC 17-166, (2017) (declaratory ruling, report and order, and order).

301Id. at ¶¶ 2–3.

302Id. at ¶¶ 304–12.

303Id.

304 John Hendel, Senate Democrats Score Net Neutrality Win in Bid for Midterm Momentum, Politico (May 16, 2018, 9:53 AM), https://www.politico.com/story/2018/05/16/senate-democrats-net-neutrality-539275.

305 A failure to appropriately consider reliance interests was part of the basis for a recent decision finding that the Trump Administration’s decision to end the DACA program was unsupported. NAACP v. Trump, 315 F. Supp. 3d 457, 473 (D.D.C. 2018). The opinion adopts a more robust reading of the Supreme Court’s current case law regarding policy reversals, suggesting the law may be ripe for an explicit adoption of the standard we advocate. Id. (quoting Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125–26 (2016)) (“[I]t is not up to Secretary Nielsen—or even to this Court—to decide what she should or should not consider when reversing agency policy. Rather, the requirements are set by the APA, as interpreted by the Supreme Court: ‘When an agency changes its existing position, it … must … be cognizant that longstanding policies may have engendered serious reliance interests that must be taken into account.’”).

306 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 549 (2009) (Breyer, J., dissenting).

307Id.

308Id.

309Roundup: Trump-Era Deregulation in the Courts, Inst. for Pol’y Integrity, https://policyintegrity.org/deregulation-roundup (last updated Aug. 6, 2019) (documenting historically low win rate for the Trump Administration in legal challenges to agency decisions).

310 Free Enter. Fund v. PCAOB, 561 U.S. 477, 484 (2010).

311 For a general discussion of the structural conditions that make agencies more independent from political influence, see Rachel E. Barkow, Insulating Agencies: Avoiding Capture Through Institutional Design, 89 Tex. L. Rev. 15 (2010).

312See, e.g., Nixon v. Gen. Servs. Admin., 433 U.S. 425 (1977) (upholding a statute that allowed for agencies to manage presidential records over a claim of privilege); Humphrey’s Ex’r v. United States, 295 U.S. 602 (1935) (upholding a for-cause removal restriction on a federal officer).

313Free Enter. Fund, 561 U.S. at 484.

314Id. at 492–93 (citing Myers v. United States, 272 U.S. 52, 164 (1926)).

315See Richard H. Pildes, Free Enterprise Fund, Boundary-Enforcing Decisions, and the Unitary Executive Branch Theory of Government Administration, 6 Duke J. Const. L. & Pub. Pol’y 1, 3–4 (2011) (“In Free Enterprise Fund, the Court thus had something of a free pass in invalidating the Board’s structure: it could hold this peculiar design to be an unconstitutional limitation on the president’s Article II powers without putting much else about the structure of administrative government up for grabs.”).

316Free Enter. Fund, 561 U.S. at 514.

317 Neomi Rao, A Modest Proposal: Abolishing Agency Independence in Free Enterprise Fund v. PCAOB, 79 Fordham L. Rev. 2541, 2559 (2011) (“Despite claims to minimalism by the Court, Free Enterprise Fund logically implicates the constitutionality of agency independence.”); see also Neomi Rao, Removal: Necessary and Sufficient for Presidential Control, 65 Ala. L. Rev. 1205, 1208 (2014).

318See generally PHH Corp. v. Consumer Fin. Protection Bureau, 881 F.3d 75, 92–101 (D.C. Cir. 2018) (en banc).

319Id. at 110-111.

320See generally id.

321 Collins v. Mnuchin, 869 F.3d 640, 674 (5th Cir. 2018).

322 Free Enter. Fund v. PCAOB, 537 F.3d 667, 685, 715 (D.C. Cir. 2008) (Kavanaugh, J., dissenting).

323PHH Corp., 881 F.3d at 110–11.

324 Buckley v. Valeo, 424 U.S. 1, 126 (1976).

325Compare Morrison v. Olson, 487 U.S. 654 at 671–73 (1988), with Edmond v. United States, 520 U.S. 651, 662–63 (1997).

326See 501 U.S. 868 (1991).

327 Intercoll. Broad. Sys., Inc. v. Copyright Royalty Bd., 684 F.3d 1332, 1342 (D.C. Cir. 2012).

328Free Enter. Fund, 561 U.S. at 514.

329Id.

330Id.

331 Lucia v. SEC, 138 S. Ct. 2044, 2047 (2018).

332Id. at 2055.

333Id. at 2053.

334Id. (“Freytag says everything necessary to decide this case.”).

335See generally Kent Barnett et al., Administrative Conference of the United States, Non-ALJ Adjudicators in Federal Agencies: Status, Selection, Oversight, And Removal (Draft Report Feb. 14, 2018) in 53 Ga. L. Rev. 1 (2019) (examining role of non-ALJ adjudicators).

336 Lucia v. SEC, 138 S. Ct. at 2056 (Thomas, J., concurring) (quoting NLRB v. SW Gen., Inc. 137 S. Ct. 929, 946 (2017)).

337Id. at 2061–62 (Breyer, J., concurring in the judgment).

338 Greater political control does not necessarily mean a more politically responsive government. Even if elections reward candidates who reflect the views of the median voter in the electorate, elected officials will still reliably deviate from those views. In that situation, bureaucratic insulation may impose a “compensatory inertia” on policymaking process and actually bring policy outcomes closer to the views of the electorate. See Matthew C. Stephenson, Optimal Political Control of the Bureaucracy, 107 Mich. L. Rev. 53, 55 (2008) (“Up to a point, the benefit to a majority of voters from a reduction in outcome variance [that results from bureaucratic insulation] outweighs the costs associated with biasing the expected outcome away from the median voter’s ideal outcome.”).

339See Lubbers, supra note 6, at 6–7.

340 These doctrinal developments also put more pressure on the internal constraints placed on agencies, which are largely beyond the reach of courts. Internal agency processes have become a greater focus of attention in recent years and could be challenged by greater political influence on agency leadership. See Jennifer Nou, Intra-Agency Coordination, 129 Harv. L. Rev. 421, 424 (2015) (providing a general theory of how these structures are developed); see, e.g., Rebecca Ingber, Interpretation Catalysts and Executive Branch Legal Decisionmaking, 38 Yale J. Int’l L. 359, 404–12 (2013) (describing how internal processes altered outcomes in the national security context).

341See King v. Burwell, 135 S. Ct. 2480, 2483 (2015); FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 157 (2000); see also Lawson, supra note 11, at 673. But see Nathan Richardson, Keeping Big Cases From Making Bad Law: The Resurgent “Major Questions” Doctrine, 49 Conn. L. Rev. 355, 358 (2016) (defending the major questions doctrine as a means of keeping judges from overturning Chevron in its entirety on the basis of important cases).

342See Nat’l Cable & Telecomm. Ass’n v. Brand X Int’l Servs., 545 U.S. 967, 982–83 (2005).

343See, e.g., Alan Feuer, Justice Department Says Rights Law Doesn’t Protect Gays, N.Y. Times (July 27, 2017), https://www.nytimes.com/2017/07/27/nyregion/justice-department-gays-workplace.html; see also Coglianese, supra note 272, at 58–59.

344 Auer v. Robbins, 519 U.S. 452, 462–63 (1997).

345See generally Christopher J. Walker, Attacking Auer and Chevron Deference: A Literature Review, 16 Geo. J.L. & Pub. Pol’y 103, 104 (2018) (“In recent years, we have seen a growing call from the federal bench, on the Hill, and within the legal academy to rethink administrative law’s deference doctrines to federal agency interpretations of law.”).

346See Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1149 (10th Cir. 2016) (Gorsuch, J., concurring); see also Garco Constr., Inc. v. Speer, 138 S. Ct. 1052, 1052 (2018) (Thomas, J., joined by Gorsuch, J., dissenting from denial of certiorari).

347 For instance, Justice Thomas has written forcefully that both Chevron and Auer deference are inconsistent with Article III, particularly the judiciary’s obligation to “say what the law is.” See Michigan v. EPA, 135 S. Ct. 2699, 2712 (2015) (Thomas, J., concurring) (criticizing Chevron); Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1222–23 (2015) (criticizing Auer).

348See Rettig v. Pension Benefit Guar. Corp., 744 F.2d 133, 151 (D.C. Cir. 1984).

349See, e.g., Wagner Seed Co. v. Bush, 946 F.2d 918, 925 (D.C. Cir. 1991) (holding that the EPA is entitled to deference on reimbursement provision of CERCLA over dissent by Judge Williams).

350See Auer v. Robbins, 519 U.S. 452 (1997).

351See supra Part II.C.

352 FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 159 (2000).

353See Lawson, supra note 11, at 673.

354 King v. Burwell, 135 S. Ct. 2480, 2489 (2015).

355See United States v. Mead Corp., 533 U.S. 218, 226–27 (2001).

356See Christopher v. SmithKline Beecham Corp., 567 U.S. 142, 155 (2012) (“Although Auer ordinarily calls for deference to an agency’s interpretation of its own ambiguous regulation, even when that interpretation is advanced in a legal brief, this general rule does not apply in all cases. Deference is undoubtedly inappropriate, for example, when the agency’s interpretation is plainly erroneous or inconsistent with the regulation. And deference is likewise unwarranted when there is reason to suspect that the agency’s interpretation does not reflect the agency’s fair and considered judgment on the matter in question. This might occur when the agency’s interpretation conflicts with a prior interpretation, or when it appears that the interpretation is nothing more than a convenient litigating position, or a post hoc rationalization advanced by an agency seeking to defend past agency action against attack.” (citations omitted)).

357See Nat’l Cable & Telecomm. Ass’n v. Brand X Int’l Servs., 545 U.S. 967, 982–83 (2005).

358See, e.g., Philip Hamburger, Chevron Bias, 84 Geo. Wash. L. Rev. 1187, 1189 (2016). Defenders of Chevron tend to focus more closely on pragmatic concerns. See, e.g., Richard J. Pierce, Jr., Chevron and Its Aftermath: Judicial Review of Agency Interpretations of Statutory Provisions, 41 Vand. L. Rev. 301, 303 (1988) (arguing that Chevron respects the distinction between law and policy).

359 Kent H. Barnett, Christina L. Boyd & Christopher J. Walker, Administrative Law’s Political Dynamics, 71 Vand. L. Rev. 1463, 1467 (2018).

360See, e.g., Alan Feuer, Justice Department Says Rights Law Doesn’t Protect Gays, N.Y. Times (July 27, 2017), https://www.nytimes.com/2017/07/27/nyregion/justice-department-gays-workplace.html.

361See supra note 278 and accompanying text.

362 For example, questions of statutory interpretation are at the heart of the dispute over the Clean Power Plan, leading to the potential for substantial volatility if those interpretations swing wildly between administrations. See supra notes 225–227 and accompanying text.

363See United States v. Home Concrete & Supply, LLC, 132 S. Ct. 1836, 1843–44 (2012).

364See, e.g., Lisa García Bedolla & Kerry L. Haynie, The Obama Coalition and the Future of American Politics, 1 Pol. Groups & Identities 128, 132 (2013).

365 Paul R. Abramson, John H. Aldrich & David W. Rohde, The 2004 Presidential Election: The Emergence of a Permanent Majority?, 120 Pol. Sci. Q. 33, 33–34 (2005).