The Hedgehog & the Fox in Administrative Law

Abstract This essay examines the constitutional muddle of the administrative state with reference to how agencies operate–it looks at a hedgehog's problem from the fox's perspective. Not only does the structure and delegated authority of administrative agencies often exist in substantial tension with the Constitution, but agencies regularly fail to act in a manner that promotes “constitutional values.” Drawing from my experience as regulatory czar, I explain that regulatory policy is frequently developed with little regard for separation of powers, political accountability, due process, or other values drawn from the Constitution. Proponents of the status quo thus cannot rely on such values to legitimize the ever-expanding activity of administrative agencies.

"The fox knows many things, but the hedgehog knows one big thing." 1 I saiah Berlin's famous dichotomy between the hedgehog and the fox posits a distinction between those who focus on big ideas and universal truths and those who focus on granular realities. In reading the essays in this volume, it struck me that the dichotomy sheds light on the fundamental debate in administrative law: namely, whether the administrative state is constitutional.
Favoring the hedgehog's approach, I have previously raised arguments against the constitutionality of the modern administrative state. 2 Such arguments have gained substantial ground in recent years. Scholars have advanced textual, structural, and historical explanations for how the administrative state exists in substantial tension with the Constitution, including the expansive delegations of legislative authority to the executive branch, the existence of independent agencies, 3 and the combination of lawmaking, execution, and judicial functions in agencies.
In response to the constitutional critiques, some modern defenders of the administrative status quo have claimed that it is consistent with "constitutional values." They have sought to shift the debate away from the Constitution and toward the mechanisms and structures of the administrative state they believe can replicate constitutional values and functions. 4 Unlike the arguments of the early Progressives, these claims depend not only on the necessity or desirability of expert T he most fundamental debate in administrative law has always concerned whether and how the administrative state can be reconciled with the Constitution. As Justice Robert H. Jackson noted in 1952, federal agencies "have become a veritable fourth branch of the Government, which has deranged our three-branch legal theories much as the concept of a fourth dimension unsettles our three-dimensional thinking." 6 The combination of powers within administrative agencies flies in the face of the separation of powers and threatens individual liberty, democratic accountability, and the fundamental protections of due process. Scholars and jurists who look at the original meaning of the Constitution find the administrative state incompatible with the Constitution's careful vesting of distinct powers in branches with distinct features. 7 Perhaps most fundamental, the vesting of all legislative power in Congress means that such power cannot be delegated to the executive or the courts. 8 But overly broad delegations of legislative power to administrative agencies allow for the exercise of a kind of lawmaking power by the executive branch, rather than by Congress. This flies in the face of the nondelegation principle, which provides perhaps the central protection for the republican form of government under a limited Constitution. 9 Moreover, the sheer size and reach of the executive branch makes it difficult for the president to retain control of administration. The creation of so-called independent agencies places substantial delegated authority outside the direct control of the president, in contravention of the creation of a unitary executive and the vesting of all executive power in the president. 10 Finally, the courts must exercise the judicial power to say what the law is, but the complexity of regulatory decisions and the lack of a judicially enforced nondelegation principle often results in courts deferring to administrative agencies. 11 It may come as a surprise that in their critiques of the administrative state, present-day originalists read the Constitution in essentially the same way as the early Progressives. Those Progressives forthrightly acknowledged that the creation of an expansive administrative state, operating under broad delegations and combining the powers of lawmaking, enforcement, and adjudication, would be fundamentally incompatible with the Constitution. 12 The Progressives celebrated this fact: rather than follow outmoded concerns for individual liberty and private property, the new agencies would focus on expertise and government control for the social good. Early proponents of the administrative state understood that the government they hoped to establish would stand in stark conflict with the text, structure, and purposes of the Constitution. 13 Modern proponents of the administrative state break with both originalists and the early Progressives. Against the background of the expansive modern ad-Neomi Rao ministrative state, some defenders of the administrative state now emphasize constitutional values, largely accepting existing judicial interpretations allowing open-ended delegations and the independence of agencies from political control. Bracketing arguments about the fundamental unconstitutionality of the administrative state, they would shift the focus away from the Constitution and to constitutional values. 14 They propose that the administrative state is arranged and structured to reflect constitutional standards and the functional equivalent of separation of powers. 15 This attempt to ground the existing administrative state in constitutional values has gained in popularity among constitutional and administrative law scholars. Gillian Metzger, one of the primary proponents of this view, has argued that "the administrative state is essential for actualizing constitutional separation of powers today, serving both to constrain executive power and to mitigate the dangers of presidential unilateralism while also enabling effective governance." 16 She explains that the "bureaucracy, expert and professional personnel, and internal institutional complexity" of the administrative state make "an accountable, constrained, and effective executive branch." 17 These features "carry constitutional significance, both in satisfying constitutional structural requirements and in ensuring that broader separation of powers principles retain force in the world of contemporary governance." 18 Similarly, Emily Bremer has suggested that administrative law can further the separation of powers through 1) the relationships among the three branches in controlling the administrative state; 2) the relationship between the administrative state and each of the other branches; and 3) in the separation of functions within each agency. 19 Metzger and Kevin Stack have also emphasized the legitimacy promoted by "internal administrative law," which they identify as the internal processes, guidelines, policies, management structures, and other procedures that serve as effective constraints on agency power. 20 They see internal administrative law as playing a "critical role in ensuring the legitimacy and accountability of the administrative state." 21 These are just some of the variants of a general project of defending the functional constitutionality of the administrative state. 22 Critics of the administrative state point primarily to law: the text, structure, and history of the Constitution. Certain arrangements are lawful or unlawful. By contrast, the broad claim that administration fits within a reconstructed range of constitutional values depends in large measure on how administration actually works. Does the expert bureaucracy provide accountability? Do the structural arrangements of administrative agencies provide checks and balances in a manner that mirrors the Constitution's separation of powers?
The modern defenders of administration are foxes, relying not on the Constitution, but on assertions about how particular agency arrangements can reflect and promote constitutional values. This view rests on an explicit empirical claim: the administrative state is "not just beneficial in a good government sense" but also "satisf[ies] constitutional structural requirements and . . . ensur[es] that broader separation of powers principles retain force in the world of contemporary governance." 23 For instance, Metzger asserts that the administrative state "yields important constitutional benefits" and that it "performs essential constitutional functions in supervising, constraining, and effectuating executive power." 24 Metzger and Stack postulate that at the "conceptual level," "internal structures" imposed by insulating agencies from presidential control can "implement basic commitments to legality and political accountability." 25 These arguments invoke the Constitution, and so have a formalist patina, but they are in fact functionalist claims that turn on how administrative law works in the real world. The constitutional values defense relies on a series of factual assertions that administrative agencies as presently structured can provide the type of accountability and constraint consistent with constitutional values.
Critically, proponents of the administrative status quo do not claim it is consistent with the Constitution, but rather maintain that administrative agencies nonetheless serve values reflected in the Constitution. I should note that I am not here addressing the difficult question of what "values" are reflected in the Constitution. The Constitution is not a hortatory document: there is no "accountability" clause or "legitimacy" clause or "separation of powers" clause. The Constitution reflects essential principles for our constitutional republic; however, it implements those principles through the creation of branches with particular features and the careful vesting of government powers in those branches. The administrative state reassigns and blends those powers in countless ways, which naturally raises the question of how constitutional values can be served outside of the Constitution's requirements. That question lies outside the scope of this essay and in the discussion that follows I simply take the constitutional values asserted by proponents on their own terms. F rom my experience as administrator of OIRA, overseeing the regulatory activity of agencies across the executive branch, I have found little evidence to support the claims that constitutional values are furthered in administrative structure or practice. In fact, many features of modern administration systematically subvert political accountability, separation of powers, expertise, and due process.
The new defenders of the administrative state make arguments that sound in constitutional theory, but they turn inexorably on facts about "constitutional benefits" and "constitutional functions." From this perspective, constitutional law is not treated as a binding and knowable constraint, and so the validity of the constitutional values defense depends on whether administrative agencies in fact possess the claimed properties. 26 The theory depends on empirical realities, but proponents are long on abstractions and short on details. Supporting the claim Neomi Rao that the administrative state reflects constitutional values requires more than conceptual generalizations.
While I can hardly claim a comprehensive study of the operation of administration in these pages, I share some fox-like observations about the regulatory process and whether and how it reflects constitutional values, as broadly defined by proponents of this view. I start by explaining how OIRA provides one of the most effective mechanisms for promoting constitutional values in administration by ensuring presidential and White House control over regulatory policy. It cannot cure all the pathologies of administration, but OIRA review can make regulatory policy more constitutional. I also highlight some examples of the nitty-gritty workings of regulatory practice, explaining some persistent, and sometimes overlooked, features of administration that run headlong into values of democratic accountability, separation of powers, and expertise. G iven the ever-expanding reach of regulatory policy, centralized review of significant regulations at OIRA provides an essential form of accountability, rationality, and coordination. For over forty years, the office has promoted fundamental principles of presidential control over administration and thereby democratic accountability for regulatory decisions. OIRA also advances other important principles of good government, such as public participation, coordination, and due process. In a variety of ways, the process of centralized regulatory review serves many of the constitutional values identified by defenders of the administrative state.
Because OIRA is often known as the most important office no one has ever heard of, I will briefly explain how it works. 27 OIRA originated with the Paperwork Reduction Act, which President Carter signed into law in 1980. President Reagan then set forth more detailed parameters for OIRA's regulatory review process in Executive Order (E.O.) 12291. 28 Essentially, OIRA coordinates and directs regulatory policy by reviewing economically and politically significant regulations from across the executive branch. The review process includes career experts at OIRA carefully reviewing the proposed regulation: its justifications, legal authority, and cost-benefit analysis. Just as important, the review process shares the proposed regulation with other affected agencies and White House offices, including the Counsel's Office, Domestic Policy Council, National Economic Council, and myriad other presidential advisors as appropriate. This centralized process allows political and career officials from across the executive branch and within the White House to weigh in on significant regulations from their different perspectives. Conflicts and differences of opinion are generally resolved by OIRA and, if necessary, with a meeting of agency heads and ultimately the president.
The The Hedgehog & the Fox in Administrative Law utive order. With unusual and thoroughgoing bipartisan support, E.O. 12866 has become foundational to the regulatory process. The Executive Order starts with a "regulatory philosophy" and articulates twelve regulatory principles. 29 These ideas have guided regulatory review across both Democratic and Republican administrations. They are nonpartisan principles for regulation and apply both to deregulatory and regulatory actions. E.O. 12866 is truly a constitutive document in that it does not speak to how much regulatory activity or what type of regulation an administration will pursue; instead, it sets forth a philosophy and basic principles of rationality, expertise, and public welfare. It creates mechanisms to implement these principles and promote these values.
Scholars who have very different perspectives on administrative law have advocated leaving E.O. 12866 in place, and for good reason. 30 Rooting White House review in this foundational document gives it a continuity and weight irrespective of the regulatory direction of an administration. Presidents invariably have their unique guiding principles for regulatory policy, but they have maintained E.O. 12866 and its fundamental principles of regulatory review. 31 President Trump, for example, set out to eliminate burdensome and ineffective regulations, with a focus on freeing individuals, families, and companies from unnecessary government control. It was in large measure a kind of populist deregulatory agenda, focused on promoting economic, social, and religious liberty. The goal was to make administration more constitutional and, at the same time, more effective. He maintained E.O. 12866 but issued a series of additional executive orders, including the creation of a regulatory budget and the requirement of eliminating two regulations for each new one. 32 Soon after taking office, President Biden repealed some of Trump's executive orders, but also "reaffirm[ed] the basic principles" of E.O. 12866 and called for "modernizing regulatory review" based on the values of "social welfare, racial justice, environmental stewardship, human dignity, equity, and the interests of future generations." 33 As an institution, OIRA avoids conflicts about the substance of regulatory policy, instead focusing on good regulatory practices that can improve decision-making, reduce arbitrariness, and ultimately promote better outcomes for the American people. Presidents with very different regulatory approaches have remained committed to OIRA and its regulatory review function.
OIRA's process of centralized regulatory review promotes a number of constitutional principles. First and foremost, it operationalizes the unitary executive. The Constitution vests all executive power in the president, which means that the president must be able to control and direct execution of the laws. 34 Such control involves superintending administration: the president serves not only as the commander in chief, but as the administrator in chief. 35 Although disagreement continues over the extent of such control, the very idea of "presidential administration" has widespread purchase. 36 In a vast administrative state, the president can-Neomi Rao not possibly track even major regulatory initiatives. OIRA ensures that important policies are reviewed by senior White House officials who are closest to the president and his policy agenda. 37 This provides essential democratic accountability because regulations will follow the election, particularly with respect to discretionary policy choices.
Second, OIRA review provides internal checks on regulatory policy. It creates a mechanism for different White House offices and agencies to review regulatory policies, providing a wider base of participation, expertise, and judgment. A regulatory problem will be vetted from a variety of different perspectives, thus checking and balancing the particular and narrow interests of a single agency and improving the legitimacy of the ultimate regulatory decision. 38 Third, OIRA reduces the arbitrariness of regulatory decisions. At the outset, OIRA makes agencies answer the question of why a particular regulatory action is necessary and how it fits into the existing regulatory landscape: what is the problem to be solved and, if regulation is not required by a statute, is it a problem susceptible to a regulatory solution? Agencies must also demonstrate that their policy produces net benefits for the American people: namely, that the benefits of the regulation outweigh the costs. While debates will always exist about which costs and benefits should count, it is difficult to justify a regulation that imposes greater costs than benefits on society. 39 Because OIRA passes a proposed regulation through other agencies and White House offices, the process can be used to avoid duplication or to resolve conflicts, such as when agencies adopt different standards to deal with the same problem.
The primary limitation on OIRA review is its reach. Notably exempt from the OIRA review process are the regulatory actions of the historically independent agencies, despite the long-standing understanding that such review would be constitutional. 40 In addition, OIRA review extends to economically and politically significant regulatory actions, which includes only a subset of all regulatory activity; however, OIRA determines which regulations are significant, and so could review more regulations with additional resources.
OIRA review provides a powerful mechanism for implementing political control over the bureaucracy. In practice, OIRA and the process of regulatory review it oversees is one of the most effective institutional mechanisms to ensure constitutional administration. Initiation myopia: regulatory policy without political supervision. While an ongoing debate continues about whether and with what specificity Congress may delegate authority to agencies, expansive and numerous delegations have a consequence that is overlooked: namely, that regulations are often initiated at a very low level of government. The conventional view assumes that regulatory policy originates with an agency head or senior official, or at times with a White House directive, and that therefore the president asserts political control, at least indirectly, over delegated authority. Yet a sea of regulatory activity occurs outside of such accountability structures. Regulations, guidance documents, and policy statements sometimes find their origination and completion with a single government employee, despite the fact that Congress in most instances delegates authority to the heads of agencies. 42 Regulatory actions can be radically decentralized, not only away from presidential control, but without control or supervision by any accountable political official.
Faced with a significant volume of regulatory responsibility, agency heads sometimes subdelegate their statutory authority, with varying degrees of residual oversight. 43 Agency staff can thus seize the opportunity to identify a problem and write up an advanced notice of proposed rulemaking, then a notice of proposed rulemaking, and ultimately a final rule. Depending on the agency and its organization, and the importance of the regulation, such activity might be reviewed by senior officials; but once the regulatory ball is rolling it is very difficult to change direction, much less stop it altogether.
The problem expands when we take into account subregulatory activity, such as guidance documents and policy statements. As OIRA administrator, I asked agencies to review their guidance documents, which involved identifying them, eliminating outdated or conflicting guidance, and making the documents publicly accessible. 44 In many instances, this proved to be an overwhelming task. We found instances of extant guidance documents that existed nowhere but the drawer of a single employee. Agencies such as Health and Human Services frankly acknowledged that it would be impossible to identify and catalog all guidance documents. While the government binds the public with regulations and then interprets those regulations through guidance documents, some agencies could not even identify, much less make public and available on a website, all of their guidance documents. And although guidance documents are not formally binding, the reality is that guidance may have coercive power, not dissimilar from a statute or regulation. Agencies have significant enforcement powers, as well as control over billions of dollars of grant money, and so regulated entities frequently attempt to take shelter in guidance. 45 With significant opportunities for regulatory action, a single bureaucrat can at times exercise an authority that exceeds that of a member of Congress. Consider that hundreds of bills are proposed each year by individual representatives and senators, or small groups of lawmakers. Most of these, irrespective of their Neomi Rao merits, get not so much as a committee hearing, much less a vote. The agency employee, however, may not only initiate but complete a regulation that affects the rights and obligations of private parties, or pen a guidance document that influences how those rights and obligations are understood and enforced.
Some agencies have a greater degree of centralized review of regulations and, of course, economically and politically significant actions go through OIRA's centralized regulatory review process, which helps "to rein in bureaucratic freelancing." 46 Such review, however, reaches only a small fraction of regulatory activity. 47 Meaningful burdens can be imposed by regulations that do not reach the threshold for OIRA review or even consideration by an agency head or other political official.
The practical reality of how regulatory discretion and power are exercised undercuts the claim that administration reflects constitutional accountability. The Constitution creates a particular type of accountability that depends on direction and supervision by politically accountable actors. In agencies, however, many decisions are made without such direction and supervision. Initiation of policy by lone, politically unaccountable employees fractures the unitary structure of execution of the laws: a single official might not know what is happening elsewhere in the agency (much less in other agencies) and is less likely to be aware of conflicting regulations or policies. It is unrealistic to assume that a person trained in a narrow area, and without involvement in her agency's broader strategic decisionmaking, would be able to see the big picture and whether a regulation is necessary or effective. In the absence of political oversight and direction, agency staff may, through inadvertence or design, undermine the policies of the president, the democratically elected head of the executive branch.
Moreover, fractured decision-making has only a tenuous claim to "expertise." True regulatory expertise requires not just the specialized or granular knowledge that a few officials may possess, but also a broader understanding of the existing regulatory landscape, legal requirements, and economic and social needs. 48 Every regulatory choice involves a series of trade-offs between various public interests, policy goals, and costs. One could hardly expect such expertise to exist in a few government officials who are unaware of the wider regulatory picture. Decisions that seem rational in isolation may in fact be unnecessary, duplicative, or arbitrary when considered in light of additional information.
Regulatory myopia is magnified when decision-making is pushed to lower levels of government. Progressives sometimes point to professional norms of the bureaucracy as providing important constraints in addition to expertise, and I was fortunate at OIRA to work with an exceptionally talented and professional career staff. Nonetheless, the incorporation of professional norms varies across agencies and also from individual to individual and so cannot adequately or consistently stand in for expertise and accountability. Regulatory authority is often exercised in dispersed silos, a fact that challenges the claims that internal or functional separation of powers operates to check and balance administration. Administrative structures fail consistently to ensure the necessary political accountability is brought to bear on the wide range of regulatory decisions made by career staff.
The pernicious and pervasive problem of regulatory carve-outs. In order to avoid regulatory burdens, individuals, companies, and members of Congress acting on their behalf frequently seek exemptions. The process of creating and granting regulatory exemptions undermines the accountability, legitimacy, and expertise claims for administration.
As administrative activity expands, so too does the use of exemptions. 49 Exemptions, like regulations, are often secured through rent-seeking and tend to benefit those with the greatest ability to sway agency officials. Getting out from under onerous and expensive regulations can mean big business and is thus pursued by special interest groups as well as members of Congress representing industries within their districts and states. Regulatory exemptions and waivers are an insider's game, often turning on access and influence and providing little visibility and accountability. Targeted exemptions thus tend to benefit the wellheeled and connected. The disparate availability of exemptions runs against our egalitarian and democratic values, which affirm that no man (or company or congressman) should be above the law.
Exemptions can also distort incentives, resulting in less beneficial regulation and, in some cases, unnecessary and overly burdensome regulation. 50 While some might cheer poking holes in an otherwise onerous regulatory regime, exemptions provide short-term benefits to a few well-connected groups, which in turn only make it more likely that onerous regulations will be placed on other parties. If the primary opponents to a regulation secure an exemption before the regulation is enacted, they may in fact support the imposition of regulatory burdens on their competitors and barriers to entry for future competitors. The granting of exemptions eliminates the constituency most likely to fight against or to moderate a regulation, which in turn may result in less socially beneficial regulatory policy. Moreover, regulators often have little to lose by granting exemptions: they can be a relatively low-cost way of buying off vocal opposition and allowing the agency to move forward with an otherwise controversial policy.
Exemptions and nonenforcement practices vary across agencies and come in different shapes and sizes, more than I can canvass in this essay. 51 Some exemptions may be socially beneficial, such as those that tailor regulations to generate the greatest benefits at the lowest costs by, for example, exempting small entities. 52 Other exemptions may seek to protect important constitutional liberties, such as freedom of religious exercise. 53 Nonetheless, exemption practices often reflect some of the worst problems with administration. For instance, the avail-Neomi Rao ability of exemptions and who benefits from them is often entirely hidden from the public and therefore from political accountability. Moreover, because exemptions frequently turn on the political influence of a favored member of Congress, company, or individual, the granting of exemptions is often unconnected with expertise or good regulatory outcomes.
Agencies often have statutory authority for waivers. 54 Although the explicit grant from Congress may increase the legal legitimacy of exemptions, it does not necessarily improve regulatory outcomes. As Mila Sohoni has explained, waivers and delay can undermine the "administrative constitution." She identifies problems in a number of areas, including immigration policy and the Deferred Action for Childhood Arrivals waiver program, health care and Affordable Care Act waivers, and education and the No Child Left Behind waiver program. 55 Another example is the Dodd-Frank Wall Street Reform and Consumer Protection Act, which creates substantial discretion within several financial regulatory agencies, expanding waiver authority. 56 Richard Epstein has argued that agency discretion "can end up blurring the line between coercive power and waiver power in a way that grants these agencies an immense amount of informal authority-authority that extends well beyond the powers they are granted by Congress." He identifies the Food and Drug Administration's process for new drug approval as a prime example. 57 Others have focused on renewable fuel standard credits, which involve ongoing, intense rent-seeking, and are the subject of litigation in the courts of appeals and the Supreme Court. 58 Moreover, waivers are distinct from the executive branch's traditional nonenforcement power. When the government declines to enforce the law, the law remains the same and could be enforced if circumstances change. On the other hand, waivers purport to change the law, granting a specific exemption or reprieve from certain legal requirements. This distinction may particularly matter in regulatory areas, such as environmental law, where Congress has authorized citizen suit enforcement. Agencies sometimes argue against judicial review of waivers, maintaining that those who are subject to regulatory requirements lack standing to challenge a waiver given to a different person. 59 The process of regulatory exemptions highlights another institutional and constitutional difficulty. The executive power includes a discretionary authority not to prosecute or not to enforce administrative requirements. Congress has set general laws through the legislative process and the executive branch can exercise a discretionary nonenforcement power, consistent with a system of checks and balances. In the regulatory space, however, the agencies both write the "law" through regulation and then determine who is exempt from it. Administrative rulemaking thus blends general lawmaking power with the execution of those laws. 60 The collapsing of these functions further undercuts the claims that agencies effectively embody constitutional values and internal separation of powers. Social values and administration. While the discussion of initiation myopia and exemptions focuses on procedural or structural problems with administration, the substance of regulatory policy increasingly raises constitutional concerns. Agency regulation on hot-button moral, ethical, and social issues challenges the democratic legitimacy of administration. One of the most important constitutional principles is that separation of powers serves individual liberty and protects against government intrusions on individual rights. The Article I, Section 7 requirements of bicameralism and presentment make it difficult for Congress to act on issues about which Americans are divided, and action on such matters is usually possible only with compromise and a minimalist approach. The administrative state unravels many of these fundamental protections.
There is a substantial and important literature on the economic impacts of regulation and how it infringes individual liberty by tangling individuals and businesses in red tape. There is scant discussion, however, on how the administrative state-regulations as well as welfare transfers with conditions-distorts not just the marketplace, but also family life, community, and religious practice. Regulatory approaches to hot-button cultural issues demonstrate that agencies lack the restraints incorporated into constitutional checks and balances. We live in a pluralist society in which Americans have diverse, and sometimes incommensurate, religious, cultural, and social values. Divisions among Americans make a uniform federal approach difficult to enact, and so it is hardly surprising that Congress virtually never legislates on matters such as abortion, contraception, or affirmative action.
Instead, Congress has delegated substantial authority to agencies, authority that agencies increasingly use to impose federal mandates that implicate matters of life and death, religious practice, marriage, and the family. For example, the Equal Employment Opportunity Commission sought to regulate church hiring decisions, a regulatory action found unconstitutional by the Supreme Court. 61 Whereas the far-reaching Affordable Care Act was silent on contraception, Health and Human Services imposed a regulation mandating the provision of contraception by employers. 62 Agencies also use regulatory action and federal funding to condition whether domestic and foreign entities provide abortion, an issue that whipsaws from administration to administration. It is difficult to imagine Congress passing any of these regulations through the ordinary legislative process.
The involvement of agencies on such matters is a relatively new development. For most of U.S. history, the federal administrative state had nothing whatsoever to say about religious and moral questions. The expansion of federal programs, grants, and transfer programs has provided agencies with numerous levers to impose social policy in a way that takes sides in the culture wars. The Constitution largely left these issues to local and state governments, but federal agencies increasingly issue sweeping regulations that leave little room for disagreement and accommodation of different viewpoints and beliefs.

Neomi Rao
Many Americans consider such intrusions deeply illegitimate and unconstitutional, and oblivious to differences in religious and community norms. In particular, Americans with sincerely held religious beliefs increasingly find their views under siege by administrative agencies. The reality is that the federal bureaucracy largely (though of course not exclusively) reflects a particular class of workers that is not representative of Americans as a whole. For example, although recent elections reflect a country fairly divided between Democrats and Republicans, political donations from agency workers skew overwhelmingly to Democrats. 63 Unlike the legislative process, which brings together representatives from around the country who reflect their communities' diverse beliefs and mores, agency workers tend to represent a narrower political class centered in Washington, D.C.
Problems of legitimacy and accountability do not run exclusively in one direction. Presidents pursuing conservative regulatory policy will no doubt frustrate progressive Americans. Administrations are directed, quite appropriately, by the president, and on controversial issues, administrations will follow the president's policies, though not always with the moderating influence of legislation. The difficulty of enacting legislation means that presidents will seek to capitalize on their control over administrative agencies. On disputed matters, about which agencies often have substantial discretion, internal checks and balances may fail to provide legitimacy and accountability for those on the losing side of regulatory policy.
Congress rarely legislates on cultural issues because members cannot reach consensus or compromise on what are often contentious questions. In part, this reflects our Constitution at work: when a common federal approach cannot be reached, individuals are left free to follow their beliefs and work within their communities to resolve problems through state and local political processes. By contrast, the ever-expanding administrative state is not content to leave such matters to individuals, families, and their local communities. Sweeping regulatory approaches to cultural issues demonstrate how the administrative state fails to promote the legitimacy, accountability, and protection for individual liberty at the heart of our Constitution. A dministration often falls short of constitutional values because it often falls short of the Constitution. Restoring Congress as the central lawmaking body in our federal government would go a long way to making administration more constitutional. Delegations to the executive branch have upended our system of government, distorting not just the lawmaking power but also the executive and judicial powers. Holding that hedgehog's idea, however, will not cure the pathologies of administration, at least not right away. The relationship between big ideas and more ordinary facts is complex, in administrative law no less than in political philosophy. Absent a substantial realignment of the administrative state, important work remains for the fox. As I learned at OIRA, faithful execution of the laws means ensuring agencies stay within their delegated authority, follow processes that encourage political accountability, and promote due process in the creation and enforcement of regulatory policy. The exercise of the judicial power reflects a different institutional balance between hedgehog and fox, but that is a topic for another day.
author's note  : 223 (explaining that while Americans previously took it as "an axiom that the people themselves were subject to certain fundamental limitations, running back of all constitutions and inherent in the very nature of free government," now "a great change has gone forward . . . the present generation seems eager to reject the idea of a fundamental law . . . [and] is eager to unshackle administration, to take away judicial review of administrative action"). 13  ) ("OIRA's regulatory review process, which draws on the expertise of the entire government, has come to provide an essential mechanism to ensure unity and coherence in execution of the law"). 38 John McGinnis has suggested that under broad delegations, OIRA serves the same pur poses as bicameralism by putting "an additional screen between government decrees and the citizen by filtering out special interest regulations from those in the public in terest," and that "the substance of the costbenefit analysis mandated by the regulatory review order helps screen public interest regulations from those sought by special in terests-an implicit objective of the original Constitution's requirements for legislative enactments.  40 See "Extending Regulatory Review Under Executive Order 12866," 5-7 (concluding the president may direct independent agencies to comply with the OIRA regulatory review process and discussing the views of the Reagan and Clinton administrations that OIRA review of independent regulatory agencies would be constitutional). 41 The full exploration of examples is outside the scope of this essay, so I focus on illustra tive features that are discussed infrequently, if at all. 42 Congress usually vests rulemaking authority in constitutional officers, most common ly the head of an agency, a principal officer who must be appointed by the president with the advice and consent of the Senate. Expertise might be the guiding light of ad