We need to recognize that the first step—rather than further government "medicine"—is to stop poisoning the patient, and return to capitalism.
Multiple scandals involving myriad federal agencies have placed the administrative state front-and-center in many Americans’ minds. Though the scandals involve overreach by specific agencies, they raise broader and more profound questions that extend to the entire federal bureaucracy because the institutional problems are systemic. Are administrative agencies with massive powers compatible with a free society? How can we hold them accountable? Can we have agencies without abandoning the rule of law, checks and balances, and other cherished institutional ideas embedded in our constitutionalism?
When the political process fails to promote limited government and individual rights, turning to the courts seems natural. Yet in recent decades conservatives have been reluctant to use the courts to limit the administrative state. In this forum I will argue that because agencies are not directly accountable to the President in a meaningful sense, and because the courts can and have historically served to limit administration, conservatives should reconsider this reluctance.
The Conservatives’ Turn to Deference
The scope of judicial review of administrative agencies has proven a highly controversial and intractable question, and progressives and conservatives have shifted their positions with the changing of political dynamics. When the administrative state was established most progressives favored judicial deference to agencies, arguing that administrative discretion could not be subjected to general legal principles and that the courts were too conservative anyway. By the 1960s and 1970s progressives had changed their position. Agencies were increasingly seen as “captured” by the economic interests they were tasked with regulating. In response progressives turned to courts as the solution rather than the problem. Various administrative law doctrines emerged to make the courts players in the administrative state. Typically, these courts used their newfound power to expand regulation, especially on environmental and civil rights matters.
Conservatives’ dedication to judicial deference to agencies is largely a function of the progressives’ turn to judicial oversight of the administrative state. Since judicial review was being used to expand regulation, conservatives ironically embraced the earlier progressives’ dedication to judicial restraint. The most controversial development in this context was the rise, under Justice Scalia’s leadership, of the Chevron doctrine, which declared that reviewing courts should defer to administrative agencies’ interpretations of law.
One question surrounding Chevron, namely whether deference extends to agencies interpreting the very scope of their power, led to the Supreme Court’s recent decision in City of Arlington v. FCC. In Arlington, the Court ruled that agencies receive deference from courts even when they interpret the extent of their own authority.
The ruling alarmed many commentators. It seems paradoxical to suggest that courts have no role in saying what the law is, particularly when there is an obvious conflict of interests, such as when agencies interpret the extent of their own authority. Even more interesting than the result, however, is the fact that the “conservative” judges on the Court were internally divided. Scalia, supported by Thomas, sided with the agencies. Chief Justice Roberts, joined by Alito and Kennedy, dissented.
Recent events and Arlington highlight the pressing need for conservatives and libertarians to devise sound principles for determining the judicial role in the administrative state. Arlington also suggests the difficulty conservatives and libertarians have experienced in devising coherent principles in this area.
Two challenges in particular have stood in the way. First, there is insufficient understanding of the administrative state’s political dynamics. Unitary executive theory is often cited as a justification for judicial deference to agencies. But this assumes that the President can be held accountable for the actions of administrative agencies, a proposition I will question in the next section. Second, there is inadequate understanding of courts’ historical and constitutional role in administration before the administrative state, which I will discuss in the third section.
Is the President the “Manager-in-Chief?”
Astonishingly, we still don’t fully understand how responsible the President is for the actions of administrative agencies (such as the IRS). Yet this question is highly important for devising a theory of the administrative state that will help determine the proper role for judicial review. This issue influenced the dispute between the majority and dissent in the Arlington case. Dissenting, Chief Justice Roberts argued that agencies not only possess “a potent brew of executive, legislative, and judicial power,” they also exercise these powers independently of the President. On the other hand, Chevron’s defenders often argue that the President is accountable to the people, and that keeping unelected judges out of the way will increase the agencies’ accountability to the President, and to the people.
Simply put, if the administrative state is really accountable, legally and practically, to the President, then it is not a “fourth branch” of government but merely part of the executive branch, accountable to the public. Alternatively, if the administrative state is not accountable to the President, one cannot defend Chevron on separation of powers or accountability grounds.
Scholars often assume that the President is responsible for the administrative state. Eric Posner and Adrian Vermeule, for example, argue that the administrative state has produced an executive branch on steroids, one that increasingly “relegate[s] legislatures and courts to the sidelines.” In this view, the administrative state is the President’s province, and Congress and the courts play insignificant roles. Citizens also generally assume that administrations belong to the President. The media and citizens refer to the “Obama Administration,” without asking whether the President was involved in the decision being discussed.
Yet this assumption has not been demonstrated and there are reasons for believing that Congress and the courts influence the administrative state as much as or more than the President. The most obvious tools the President could use to control and influence agency decision-making are appointment and removal of agency personnel. Yet the President’s power over personnel has been greatly curtailed over the past century. While there are millions of civilian civil servants in the administrative agencies, there are only a few thousand political appointees.
Although these appointees sit atop many agencies, career bureaucrats have ample means of undermining them – a fact Republican presidents often learn the hard way. Richard Nathan, for instance, chronicled the challenges the Nixon administration faced in establishing an “administrative presidency.” When Nixon realized that Congress would not enact his “New Federalism” agenda, he sought to make policy through the bureaucracy. Yet he found that directing the agencies was just as difficult as directing the Congress. Rather than serving as “spokesmen for the president” and his agenda, his appointees were drawn “into the orbit of the program interests of their agency.” The “[f]orces that come into play after an appointment has been made” turned Nixon’s own appointees against him. They had to rely upon the expertise of the career bureaucrats in the agency, who knew much more about program details. Furthermore, “strong outside forces,” particularly congressional committees and various interest groups, stymied efforts to impose the President’s objectives on unwilling agencies. In short, Nixon’s political appointees in the administration were coopted by the agencies and the various players that were oriented around them.
These forces continue to prevent the President from taking control of agencies through appointments. Exacerbating the challenge, most bureaucrats (including many political appointees) are removable only for “cause,” which ensures their political independence from the White House. In short, the President does not hire most bureaucrats, and is also incapable of firing most of them.
The President’s inability to control and oversee administrative personnel suggests the challenges presidents face in confronting bureaucratic resistance. The alternative that Nixon devised, namely circumventing political appointees in the agencies with personal appointees in the White House, was also unsuccessful due to the volume and complexity of the administrative state – a factor that has increased massively since Nixon. In other words, whether he seeks to use political appointees in the agencies, or personal and loyal members of the White House staff, the President lacks the ability to oversee the administrative state in any meaningful sense.
Seeking other options, Ronald Reagan (and subsequent Presidents) issued an executive order establishing White House review of “major” rules promulgated by executive agencies, undertaken by the Office of Information and Regulatory Affairs (OIRA). Though OIRA review has enhanced the President’s control over agency rulemaking to an extent, it has not been a game-changing innovation. Because the legal authority for OIRA review comes from an executive order, it applies only to executive agencies. Independent regulatory commissions are exempt. Furthermore, it is legally and practically unclear how conflicts between OIRA and administrative agencies are resolved. The Office of Legal Counsel’s opinion in 1981, when Reagan issued the original executive order establishing OIRA review, was that the President may “consult with those having statutory decision-making responsibilities…as long as the President does not divest the officer of ultimate statutory authority.” For a variety of reasons OIRA review’s effect has been diminished and it is still difficult to hold the President accountable for the administrative state.
Therefore, while centralizing regulatory review power in the White House gives the President the power to control some administrative activity in a targeted way, it does not render the President accountable for the administrative state as a whole, which is too vast for him to oversee given the limited tools at his disposal. This setup seems to give presidents the best of both worlds: the power to intervene in particular decisions, behind the scenes, and the ability to avoid responsibility for unpopular actions (such as the NLRB’s prosecution of Boeing). This is certainly not what Hamilton advocated when he asserted in Federalist 70 that a unified executive allow the public “to determine on whom the blame or the punishment of a pernicious measure, or series of pernicious measures, ought really to fall.”
Clearly we need a more sophisticated assessment of the institutional dynamics of the modern administrative state. Any account that ties the agencies simply to Congress, or to the President, or to the courts is overly simplistic. All three branches exert influence using the various tools at their disposal. Congress’s tools, particularly its appropriation and authorization powers, are considerable. The President’s powers, including those of appointment, removal, and review, are substantial. And in spite of Chevron and other seemingly deferential administrative law doctrines, agencies’ decisions are still profoundly affected by the prospect of judicial review. In short, the story of influence in the administrative state cannot easily be told. The administrative state is neither beholden to Congress, nor to the President, nor to the judiciary, but to all three, and to varying degrees depending on the occupants of the various branches.
As a result of this complex array of influences, the logic of the unitary executive and judicial deference to the political branches does not necessarily apply. Judicial deference to agencies is not deference to an accountable bureaucracy, at least in the constitutional respect. And the bureaucracy does not represent a unitary executive, but rather an administrative branch without direct accountability to any single political actor.
The Judicial Role in Administration
Today’s administrative state, therefore, makes a mess of the constitutional separation of powers and its careful adjustment of incentives, checks and balances. In such a system, what role can and should the courts play in reviewing agency decision-making? Here is where a deeper understanding of the courts’ historical role in administration is most needed.
If one examines the last several decades of judicial review of agency decisions, one might conclude that judicial review could only serve to expand and increase the intrusion of the administrative state. As Michael Greve has noted on this site, eliminating judicial deference is not guaranteed to constrain the administrative state: “David Tatel on the D.C. Circuit…decides more AdLaw cases in a month than the Supremes will see in a decade. You don’t want to arm him” with the ability to overturn agencies’ legal interpretations. Federal courts could use judicial review to require reluctant agencies to regulate rather than to restrain overactive agencies from regulating. Massachusetts v. EPA – where the Court imposed its own reading of the Clean Air Act on the EPA, prodding it to regulate greenhouse gas emissions – illustrates this possibility.
Certainly an active judiciary can cause great headaches for conservatives and libertarians seeking to limit the administrative state. But if one takes a broader view of the role that courts have historically played, and considers Madisonian constitutional theory, an alternative possibility emerges. In short, in the first decades of American political development the judicial power was used to limit administration and judges even served in place of administrative bodies.
As Jerry Mashaw and others have noted, the threat of judicial review hung constantly over revenue collectors and other officials in early America. Administrative officials were even held personally liable for injuring citizens in the course of their duties. Citizens had recourse to damages as a remedy for administrative injustice and juries could impose liability on office holders. Common law remedies were widely available for those who sought to challenge administrative power, in short. Judicial review of administrative action under the common law was robust. Moreover, it was hardly difficult to gain access to the courts. Qui Tam actions allowed private “informers” to bring legal action against those who violated the law – obviating in many cases the need for federal bureaucrats to prosecute citizens. These informers often gained part of the fine that the court would assess for violating the law.
Judicial review of administration in early America, therefore, was prevalent. But in many cases the courts were themselves the administrators. Regulation was administered, in many cases, by courts applying common law principles in particular cases. Nuisance regulations and the like were judicially crafted and enforced. The licensing of vessels by the national government during the early republic relied in part on judicial enforcement. When the terms of the license were violated, only a court and not an administrative body could revoke the license. In these and in many other cases, the Founding generation opted to lodge the power of applying the law not in administrative authorities, but in the courts.
Similar developments occurred in Europe in the late 19th Century, a phenomenon Friedrich Hayek discussed in The Constitution of Liberty. In Germany legal thinkers and reformers developed the idea of the Rechtstaat, or state of law. Central to the Rechtstaat was the idea of controlling administrative discretion and preserving the rule of law. The German reformers insisted upon subjecting “all disputes between the administrative authorities and private citizens…to the jurisdiction of ordinary courts” rather than administrative tribunals. An important outcome of this movement was the establishment of administrative courts – outside and independent of the agencies – that had ordinary jurisdiction over administrative issues.
In America, a few thinkers advanced similar proposals. The most prominent of these scholars was Roscoe Pound, who compared the rise of the American administrative state to similar developments in the contest between the common law and the administrative tribunals in 16th and 17th Century England. The solution for American law, Pound believed, was to develop specialized courts that would alleviate the need for administrative tribunals: “If…we meet the movement away from law by a modernizing of the legal and judicial machinery…we may be confident that now, as in Tudor and Stuart England, the law will prevail.” In Pound’s vision courts would eventually replace administrative adjudication. The development of these specialized courts “would enable the judicial department to do adequately the work which, in desperation of efficient legal disposition, we have been committing more and more to administrative boards and commissions, which are contrary to the genius of our institutions and often, at best, are mere experiments.”
How do these historical lessons apply to the modern administrative state? Consider why the Founders might have set up these arrangements, and what they saw as the virtues of judicial involvement in administration. First, judicial review of administration and judicial involvement in administration supported the separation of powers by limiting and checking the powers of administrators. They did not believe courts were usurping another branch’s powers by reviewing the legality of administrative action. Second, the 19th century approach promoted an independent enforcement power, since the courts were fully separate and independent from the administrators. Third, the 19th century approach ensured the protection of judicial procedure when life, liberty, or property was on the line. Administrative adjudication resembles a judicial proceeding, but lacks the protections of the judicial process. The end of adjudication is not justice for the parties, but the public interest. As a result, administrative adjudication does not require trial by jury and other due process protections. The deck is stacked because the nature of administrative adjudication fundamentally differs from the nature of a judicial proceeding.
Broadly, these considerations suggest that conservatives may gain more than they would lose by promoting legal doctrines that defend judicial review of administration. They also indicate the importance of putting some adjudicative powers back into the judiciary, perhaps in specialized courts, where judges can ensure that particular decisions are made consistent with the rule of law and due process.
More specifically, these considerations pose questions for the Chevron doctrine. The theory and practice of separation of powers in the early republic offer little support for this idea of judicial deference. Perhaps it is also worth considering whether, in certain contexts, allowing citizens to bring actions against administrative agencies, and enabling common law remedies in common law courts, would help to preserve accountability and the rule of law in the face of an increasingly arbitrary administration.
These are tentative ideas. History has demonstrated that the judicial power is not categorically disposed to increasing or reducing the scope of the administrative state. The effects of judicial review will probably depend on the quality of judges and lawyers: it is only conditionally (when judges and lawyers are poorly trained) that judicial review bolsters the administrative state. On the other hand, the administrative power is categorically inclined to expand the administrative state.In the final analysis, reintroducing judicial review of administration – and judicial administration itself – is worth considering when the alternative is our headless fourth branch of government.
 The standard history of these developments is Richard B. Stewart, “The Reformation of American Administrative Law,” Harvard Law Review 88 (1975): 1667-1813.
 R. Shep Melnick, “The Politics of Partnership,” Public Administration Review 45 (1985): 653-660.
 Jonathan Turley has pointed out that President Obama has insisted he had no knowledge of recent decisions made by the State Department, Justice Department, or the IRS. Turley, “The Rise of the Fourth Branch of Government,” Washington Post, May 24, 2013, available at http://articles.washingtonpost.com/2013-05-24/opinions/39495251_1_federal-agencies-federal-government-fourth-branch.
 For instance, the Court’s rationale for deference in Chevron itself relied in part on the argument that “While agencies are not directly accountable to the people, the Chief Executive is, and it is entirely appropriate for this political branch of Government to make such policy choices” as those entrusted to the EPA. Chevron v. Natural Resources Defense Council, 467 U.S. 837, 865 (1983). See also Jeremy Rabkin, Judicial Compulsions: How Public Law Distorts Public Policy (New York: Basic Books, 1989), 7-8: “[J]udges cannot be responsible for public policy because they are, quite literally, not responsible. The traditional constitutional view was that public policy must be left to officials who are responsible because they are accountable to the public….[O]fficials who can be dismissed at will by the president – in a system in which presidents face the voters every four years – clearly are more accountable than life-tenured judges in any meaningful sense of the word ‘accountable.’”
 Eric Posner and Adrian Vermeule, The Executive Unbound: After the Madisonian Republic (New York: Oxford University Press, 2011), 3.
 Richard Nathan, The Plot that Failed: Nixon and the Administrative Presidency (New York: Wiley & Sons, Inc., 1975), 39.
 Nathan, The Plot that Failed, 39.
 Nathan, The Plot that Failed, 41.
 This phenomenon prompted Nixon’s close aide John Erlichman to remark at a press briefing about the administration’s appointees: “We only see them at the annual White House Christmas party; they go off and marry the natives.”
 See Nathan, The Plot that Failed, 49-53. In summary, Nathan writes, “Even with significantly enlarged White House and Executive office staffs,” Nixon’s counter-bureaucracy “was limited in relation to the far greater numbers of agency personnel. They simply could not get a handle on the multitudinous field of domestic affairs.” (61)
 Major agency rules are those costing $100 million or more.
 By Reagan’s second term, roughly 30% of proposed rules were modified or abandoned altogether after undergoing OIRA review. During the presidency of George W. Bush, 150 regulations were withdrawn after OIRA review. Thus OIRA review has clearly given the President some influence, but that influence has been limited.
 As Christopher DeMuth explains, “Closer cases are typically the subject of vigorous internal disagreement; sometimes OMB prevails, sometimes the agencies prevail, sometimes OMB and the agencies compromise.” DeMuth, “The Regulatory State,” National Affairs (Summer 2012), 80.
 As Lisa Heinzerling summarizes, “The process is utterly opaque….[and] diffuses power to such an extent…that at the end of the day no one is accountable for the results it demands.” Heinzerling, “Inside EPA: A Former Insider’s Reflections on the Relationship Between the Obama EPA and the Obama White House,” Pace Environmental Law Review (forthcoming). Draft June 12, 2013. Available at SSRN: http://ssrn.com/abstract=2262337. Christopher DeMuth similarly explains that while “White House review has, with little ado, culled many clearly bad agency proposals and fortified many clearly good ones…these improvements, while real, have been marginal.” DeMuth, “The Regulatory State,” 80.
 For an example of an effective targeted intervention by the President, involving the termination of the Yucca Mountain repository, see Adam J. White, “Yucca Mountain: A Post-Mortem,” The New Atlantis (Fall 2012): 3-19. For the President’s denial of involvement in the NLRB’s action against Boeing, see White, “Reining in the Agencies,” National Affairs (Spring 2012): 42-58.
 Mashaw, Creating the Administrative Constitution, 66, 76.
 Mashaw, Creating the Administrative Constitution, 61;
 I elaborate on this theme in “Regulation, Administration, and the Rule of Law during the Early Republic,” in Anthony Peacock, ed., Freedom and the Rule of Law (Lanham, MD: Lexington Books, 2010), 48-52.
 F.A. Hayek, The Constitution of Liberty, paperback ed. (Chicago: University of Chicago Press, 1978), 198. Emphasis added.
 Hayek, Constitution of Liberty, 201. The rise of progressivism in Germany at about this time, however, ended the Germans’ brief experiment with the Rechtstaat, and in the Anglo-American legal world attention turned away from the idea of separate administrative courts and focused on the ability of courts to review administrative adjudication. See Hayek, Constitution of Liberty, 203-204.
 Pound, “The Organization of Courts” (Law Association of Philadelphia, 1913), 7.
 Pound, “Organization of Courts,” 21-22. Pound’s solution for turning some administrative powers over to the courts was later echoed by the law professor Bernard Schwartz, who argued that the best course “would be for [the administrative state] to follow the pattern of the executive tribunals of three centuries ago. The justice dispensed by the great federal agencies must become truly judicialized and administered by bodies possessing solely judicial authority. Such bodies will, in time, follow the example of Chancery and develop into courts.” Bernard Schwartz, “The Administrative Agency in Historical Perspective,” Indiana Law Journal 36, no. 3 (Spring, 1961): 278.
 In fact, the ability of courts to undermine administrative power in the 19th century approach is what progressives (and today’s historians) condemned most strongly. “Regulation, Administration, and the Rule of Law in the Early Republic,” 45-48, 59-63.
th branch of government.