Cited: 51 Duke L. J. 851
[*pg 851]




   A. The Origins of Executive Order 12,291: An Anecdote
   B. Disclosure of Predecisional Deliberations and the Wolfe Case

   A. Executive Order 12,291: Reagan Administration Regulatory Review
   B. Executive Order 12,498: Reagan Administration Regulatory Planning
   C. Executive Order 12,866: Clinton Administration Regulatory Review and Planning

   A. A Cynic's Question: Are the Structural/Analytical Issues Real or Strategic?
   B. The Unitarian Accord: The Bright Line Between Independent Agencies and Other Executive Agencies
   C. The Technocratic Premises
   D. The Substantive Policy Issues

   A. The Use of Prospective Directives: Implementing a Proactive Agenda
   B. The Regulatory Planning Process





After about twenty years of political and intellectual Sturm und Drang on the issue of centralized presidential regulatory review1 -- [*pg 852] about its desirability, its legality, and the methods of its implementation -- it appears that we are all2 (or nearly all)3 Unitarians4 now.

This is not a statement about religion, although some of the intellectual combatants have approached the issue of centralized regulatory review through the Executive Office of the President (EOP) -- via the Office of Management and Budget (OMB) and its Office of Information and Regulatory Affairs (OIRA) -- with something like religious zeal.5 Nor is it a statement of constitutional interpretation. Some unitarians derive their position from Article II, under which the president embodies the executive powers of the federal government.6 Other unitarians, however, arrive at their unitarianism not from the [*pg 853] Constitution but prudentially, as a matter of policy-driven statutory construction.7 Whatever the justification, the much-maligned institutionalized process of centralized presidential regulatory review, adopted during the early days of the Reagan administration,8 has survived and even thrived under a successor Republican administration and two terms of a Democratic administration that one might have expected to dismantle it (given the hostility to OIRA and to regulatory review generally evinced by Democratic constituencies during the Reagan-Bush era).9

The Clinton administration's retention of Reagan-era institutions of centralized presidential regulatory review has been described as "a dramatic and in many ways quite surprising step" that both "maintains the basic process [of OMB regulatory review] inaugurated by President Reagan . . . . [and] also maintains much of the substantive focus of the Reagan orders, including the emphasis on cost-benefit analysis as the basic foundation of decision."10 The Clinton administration's Executive Order 12,866 actually and explicitly expanded the scope of centralized presidential regulatory review by including independent regulatory agencies in the agency planning process.11 And the Clinton administration's executive order made explicit what had been left implicit in the Reagan and Bush executive orders -- that centralized presidential regulatory review is aimed at making agency regulations "consistent with . . . the President's priorities."12 The emphasis on imposing the primacy of the president's priorities in agency decisionmaking permeates Executive Order 12,866; the phrase "the President's priorities" appears in at least seven places.13

[*pg 854]

In 1995, Professors Richard Pildes and Cass Sunstein observed that "[f]rom the recent evidence, it seems clear that presidential oversight of the regulatory process, though relatively new, has become a permanent part of the institutional design of American government."14 Six years later, at the conclusion of the second Clinton term, Professor Elena Kagan, formerly a senior member of President Clinton's White House domestic policy staff, asserted flatly that "[w]e live today in an era of presidential administration," an "assertion" that, she acknowledged, might be "jarring" or "puzzling" to some because of its reaffirmation of controversial policies that had their origin in the Reagan years.15 According to Professor Kagan, "presidential control of administration, in critical respects, expanded dramatically during the Clinton years, making the regulatory activity of the executive branch agencies more and more an extension of the President's own policy and political agenda."16 Evident from the terms of Executive Order 12,866 itself, this was a "self-conscious and central object of the White House" that "in large measure set the administrative agenda for key agencies."17 Early evidence indicates that the George W. Bush administration will continue the policy of centralized presidential regulatory review.18 On September 20, 2001, John D. Graham, the administrator of OIRA, issued a memorandum to the President's Management Council calling attention to the Clinton administration's executive order and noting that the president's chief of staff directed that Graham "work with the agencies to implement vigorously the principles and procedures in E.O. 12866 until a modified or new Executive order is issued."19

From controversial fringe to mainstream in twenty years, centralized presidential regulatory review has now taken center stage as [*pg 855] an institutionalized part of the modern American presidency. The Clinton administration, without fanfare, helped institutionalize fundamental policies and values of the Reagan administration, normalizing many of President Reagan's ideas and reforms much as the Eisenhower administration -- the first post-New Deal Republican administration -- normalized the New Deal values and structural reforms of President Franklin Roosevelt.

In this Essay, I argue that centralized presidential review of agency regulatory activity is an understandable and salutary development. At the outset, I offer a brief historical and personal note about the implementation of centralized presidential regulatory review through the issuance of an executive order during the Reagan administration. That personal experience provides some of the basis for what follows. It also contributes two important insights. First, in something of a flashback, it shows how President Reagan's executive order successfully dealt with the built-in inertia of an existing bureaucracy by implementing centralized presidential regulatory review before opponents had an opportunity to mobilize. That success came at a price, however -- the controversy and resistance the regulatory review process engendered remained a fact of life even nine years later when I was nominated to serve as the administrator of the agency charged with administering the regulatory review process.

Second, my personal perspective attests to the importance that the proponents of centralized presidential regulatory review assigned to the unitary executive principle -- the principle that the Executive Office of the President should either control or influence (based on the president's policy goals and agenda) the exercise of agency discretion, except for independent agencies. Intensive negotiations about the scope of disclosure of executive branch decisionmaking delayed my confirmation for months, and that delay eventually led to the nomination's demise on other grounds.20 The result was that the Bush administration refused to nominate anyone else to fill the position of OIRA administrator, and the vacancy went unfilled for the remainder of the Bush administration. This transferred substantial authority to [*pg 856] the Office of the Vice President, which (through the Competitiveness Council) took on additional responsibility for overseeing the centralized presidential regulatory review process.

In Part II, I set forth the essential provisions of the centralized presidential regulatory review executive orders that were adopted during the Reagan administration, followed during the Bush administration, and adhered to, albeit in modified form, by the Clinton administration.

Part III focuses on centralized presidential regulatory review, concluding that such review is warranted on policy grounds and, although it is and should be legally authorized, that the core issues are political rather than legal in nature. Section A asks whether the often sharp debate about the legal and policy aspects of centralized presidential regulatory review should be taken seriously in an analytical sense, or whether it is merely a smokescreen for underlying political arguments dressed up in structural garb. I conclude that, although the political agendas may often trump principled arguments about the proper extent of executive review, the issue is nevertheless worthy of being taken seriously on its own structural terms. Section B notes that a remarkable (if not uniform) consensus has emerged on a form of the unitary executive principle. That consensus draws a bright-line distinction between the role of presidential oversight of regulatory decisions made by independent agencies (where the president must respect their regulatory autonomy) and decisions made by other executive agencies (where the president can exercise greater influence on agency regulatory policymaking). In Section C, I contend that critics of centralized presidential regulatory review concede the technocratic claims for such review -- that such review can improve government-wide rulemaking in the traditional (noncontroversial) technocratic sense. But, as Section D explains, consensus traditionally has broken down with regard to the policy-based claims for centralized presidential regulatory review. I contend that important values are served by centralized presidential regulatory review, and that those values are consistent with our traditional understanding of how the agency rulemaking process works both in theory and in practice. Traditional constraints on centralized presidential regulatory review -- for instance, the rules that no exercise of discretion by the executive branch can violate the terms of a statute and that an agency's decision must be supported by the rulemaking record -- prevent excessive politicization of agency rulemaking. Finally, in Part IV, I consider implementation issues that can affect the degree of receptivity or hostility [*pg 857] to centralized presidential regulatory review as an institutional adaptation.



In April 1990, President Bush nominated me21 -- ill-fatedly as it turned out22 -- to serve as administrator of OIRA. During the nomination and confirmation process, I saw how contentious centralized presidential regulatory review continued to be nearly a decade after its introduction at the outset of the Reagan administration.

Although nominally I was to report to Richard Darman, then director of the OMB, I learned that C. Boyden Gray, the White House counsel, had a special interest in centralized presidential regulatory review and that I would, in effect, have two bosses. Gray had been counsel to Vice President Bush for eight years and had been a major architect of the regulatory review process.23 He wanted to be in the loop, in a general way, on OIRA issues, and I very much wanted to take advantage of his expertise and commitment. Darman was otherwise occupied at the time, negotiating a budget deal with the Demo- [*pg 858] cratic leadership of Congress,24 and Gray's active interest in the centralized presidential regulatory review process assured priority attention at the White House to issues on which Darman could not fully focus. Before accepting the nomination, I cleared with Darman this understanding of Gray's active oversight role. Darman graciously and willingly agreed to share some of his formal authority.

During my nomination and the confirmation process, Gray provided much good advice and institutional sophistication about the workings of centralized presidential regulatory review. I learned, for example, that President Bush, who as vice president had initiated centralized regulatory review, was very concerned with allegations that his administration was presiding over a re-regulatory process.25 The president was committed to a rational, analytical regulatory review process and wanted presidential regulatory review to retain the vigorous role it had during the Reagan administration. I signed on to the enterprise with the stipulation to Darman (also communicated to Gray) that I believed in and would enforce a tough-minded, intellectually rigorous, and analytically fair review of agency rulemaking, but that I would not be a mindless deregulator or a shill for industry. Both Darman and Gray readily agreed with that stance.

A. The Origins of Executive Order 12,291: An Anecdote

A flashback to the origin of the Reagan administration's executive order supplies historical context for the contentious debate about the wisdom, legality, and implementation of centralized presidential regulatory review. President Reagan issued Executive Order 12,291 within a month of taking office in 1981.26 That executive order established a process of centralized presidential regulatory oversight of agency27 rulemaking, promulgated a set of cri- [*pg 859] teria28 to govern agency conduct (such as requiring cost-benefit analysis before agencies issued new regulations),29 and, to enforce both the substantive standards and the regulatory review process, required agencies to prepare a Regulatory Impact Analysis.30

There had been a long tradition of attempts to expand presidential influence over the regulatory activity of federal agencies.31 The Carter administration, partly in response to a report and recommendation emanating from the American Bar Association,32 had recently supported enactment of the Paperwork Reduction Act,33 which gave centralized review of agency paperwork collection requests to the OMB.34 Nothing as far-reaching as Executive Order 12,291, however, had been implemented previously.

Because deregulation had been a centerpiece of his campaign, President Reagan was eager to begin the process; Executive Order 12,291 represented his first move. The order evolved from a working group that started before President Reagan took office. Vice President Bush's office took a major role in developing and coordinating the initiative. In the vice president's office, C. Boyden Gray, counsel to the vice president, took the lead.35

When I met with Gray around the time of my nomination, he spoke mischievously of a meeting of sundry agency officials convened about the new executive order. Gray described a roomful of agency officials and lawyers furiously reading through the order to see what was being proposed. Heads were shaking vigorously, sidebar conversations were expressing disagreement and skepticism. Those in attendance would let the new administration know that this idea was a bad one and impractical to boot.

[*pg 860]

Gray described how the agitated body language and cacophony turned to stunned silence as the officials in attendance, one by one, reached the end of a document they had assumed to be a draft, only to find the signature of Ronald Reagan on the last page. The far-reaching executive order already had been executed by the new president; it was a done deal. Like Athena, born from the head of Zeus, the executive order sprung full-grown from the senior members of President Reagan's team. The rapid-fire implementation of the executive order demonstrated its support at the most senior level of the Reagan administration. The centralized presidential regulatory review process would be given the respect and would have the clout in the bureaucratic implementation process that that high-level support suggested.

But there are often costs to such a striking success, and that was the case with the ultimate implementation of the executive order. Career agency officials are not without resources. Agency officials find recourse in their own expertise and longevity; carefully nurtured political allies, especially in congressional oversight and appropriation committees, subcommittees, and their staffs; and in special constituencies (such as labor unions in matters of occupational safety or environmental advocacy groups in matters of clean air and clean water initiatives). Although the sudden birth of Executive Order 12,291 effectively stymied bureaucratic resistance and trench warfare at the outset, that blitzkrieg approach rechanneled bureaucratic infighting to the implementation phase, helping to trigger a period of remarkable polarization and acrimony that has been described as an "oversight arms race."36 The political atmosphere remained contentious nearly a decade later during my nomination and confirmation process.

B. Disclosure of Predecisional Deliberations and the Wolfe Case

My confirmation was delayed by negotiations between Gray's office and congressional leaders about the reauthorization of the Paperwork Reduction Act.37 The negotiations, in which I was not an ac- [*pg 861] tive participant (on advice of those helping in the confirmation process), focused on issues involving the unitary executive concept.

Some members of Congress, led in this regard by Senator Carl Levin,38 sought the ability to track contemporaneously whether and when proposed rules went from the agency to OIRA for review.39 The goal, apparently, was legislatively to overrule the District of Columbia Circuit's en banc decision in Wolfe v. Department of Health and Human Services,40 which had denied access under the Freedom of Information Act (FOIA) to records that would indicate "what actions have been completed by the Food and Drug Administration ('FDA') but which await final decision or approval by the Secretary of Health and Human Services ('HHS') or [the OMB]."41 Advocacy groups in Wolfe had wanted access to information about the deliberative process "in order to influence decision-makers more efficiently during predecisional deliberations."42 They wanted to identify which proposals actually had been formulated and sent forth for approval by the FDA and how far along the decisional chain each proposal had traveled. That would allow advocates an opportunity to identify and target decisionmakers outside the FDA and to contest and challenge those decisionmakers' input.43 The District of Columbia Circuit held that the information sought by plaintiffs under FOIA was protected by the deliberative process privilege, which protected "inter-agency or intra-agency memorandums or letters which would not be available to a party other than an agency in litigation with the agency."44

In protecting the executive branch's deliberative process from disclosure under FOIA, the Wolfe court was quite aware that the issues were not just technical ones under FOIA but arose from "dissatisfaction with the results of the development of formal presidential oversight of executive branch rulemaking"45 and represented a fundamental and substantive challenge to that process. While adjudicat- [*pg 862] ing a FOIA claim, the court also was making a statement about the nature of executive branch decisionmaking, the deliberative process, and the nature of accountability.

The court noted that the deliberative process exemption in FOIA was "manifestly not meant to isolate agency decision-makers from public opinion or to silence public voices."46 As a unified whole, the executive branch is accountable for the end result of its deliberative process, both at the proposed and final rulemaking stages.47 However, agencies need a "space within which they may deliberate"48 free from public scrutiny so that, with appropriate public input and participation, they can reach a judgment and formulate a policy. The deliberative process privilege promotes "subordinates' willingness to provide decision-makers with frank opinions and recommendations" and prevents the "premature disclosure of proposed policies before they have been finally formulated or adopted."49 Premature disclosure "could chill discussion at a time when agency opinions are fluid and tentative" and could "force officials to punch a public time clock," allowing advocates "to attribute delay" to a particular bureaucratic player and "to identify and publicize the office or even the person they deem responsible."50

In a strong endorsement of at least one form of the unitary executive principle, the Wolfe court expressed concern that the advocates sought "to issue themselves an invitation to agency deliberations."51 For the court, "[i]t strains credulity to believe" that such an invitation "would not lead to hasty and precipitous decision-making. . . . It is just such a fishbowl" that the deliberative process privilege is designed to avoid.52 Yet, it was that type of intensified scrutiny that the advocates sought. And they sought it precisely because it allowed the advocates to use the political process -- strategic disclosures to the media, arousing a constituent base, and so on -- at interstitial stages of the unified deliberative process to rally support [*pg 863] for or opposition to a position being taken by a particular bureaucratic player. To the Bush administration, it seemed the height of irony -- bordering on hypocrisy -- that critics of centralized presidential regulatory review purported to object to the process on the ground that it politicized administrative decisionmaking when the reason that advocates wanted an invitation to all phases of agency deliberations was fundamentally political: to publicize and otherwise place pressure on elements of the bureaucracy, such as OIRA, outside the advocates' sphere of influence.

Through C. Boyden Gray, the Bush administration strongly resisted legislatively overturning the Wolfe decision. Although a nonlegislative compromise on disclosure was reached,53 it did not satisfy Senator Levin, who used my confirmation hearings as an (unsuccessful) attempt to have me repudiate the deal struck between Congress and the Bush administration.54



Presidents as early as Theodore Roosevelt55 "have struggled to assert more centralized control over the regulatory state."56 Securing presidential control of the budgeting process was an obvious place to start. Early legislation enacted in 1921 allowed the president to develop and seek congressional approval of a unified federal budget.57 Subsequent legislation58 established the Executive Office of the President (of which OMB is a part), and in the Nixon administration, the [*pg 864] "most direct precursor to the current structure of executive oversight of regulation" was developed, as "agencies were required to submit significant rules to OMB in advance of publication in the Federal Register."59 OMB's role was to "circulate the agency draft to other agencies for review and comment," but OMB's role was "rarely substantive." Rather, it served "a coordinating function."60

The process of centralized presidential regulatory review adopted by the Reagan administration, on the other hand, was substantive and, although it did serve a "coordinating function," it had a much more far-reaching objective and impact. This Part explains what the Reagan administration's two regulatory review executive orders did and how the Clinton administration adopted the core values and overall structure of the Reagan-Bush regulatory review process, with certain ameliorative dimensions (such as imposing time limits on the process and providing for more transparency).

A. Executive Order 12,291: Reagan Administration Regulatory Review

The Reagan administration established its far-reaching centralized presidential regulatory review process in Executive Order 12,291, designed to "reduce the burdens of existing and future regulations, increase agency accountability for regulatory actions, provide presidential oversight of the regulatory process, minimize duplication and conflict of regulations, and insure well-reasoned regulations."61 To achieve these goals, the executive order required all agencies to follow requirements set forth in the executive order in "promulgating new regulations, reviewing existing regulations, and developing legislative proposals concerning regulation."62 The requirements focused on rationalizing agency decisionmaking by requiring that agency "decisions . . . be based on adequate information concerning the need for and the consequences of proposed government action"63 and that agencies engage in cost-benefit analysis to determine the propriety of agency regulatory activity. Regulatory action was not to be taken by [*pg 865] an agency "unless the potential benefits to society for the regulation outweigh the potential costs to society;"64 regulatory objectives were to be "chosen to maximize the net benefits to society;"65 and among "alternative approaches to any given regulatory objective," agencies were required to select the alternative "involving the least net cost to society."66

To implement its substantive requirements, the executive order required, with respect to each "major rule," that an agency prepare a Regulatory Impact Analysis (RIA), which would identify and describe the costs and benefits of the rule, including those that cannot be "quantified in monetary terms," along with a description of comparable lower-cost approaches.67

OMB, through OIRA, was "authorized to review any preliminary or final Regulatory Impact Analysis, notice of proposed rulemaking, or final rule based on the requirements of" the executive order.68 Until OMB review was concluded, an agency could not publish its RIA or notice of proposed rulemaking.69 If OMB declined to clear a proposal as initially presented from an agency, it would provide comments. No agency could publish its final RIA or a final rule "until the agency ha[d] responded to [OMB's] views, and incorporated those views and the agency's response in the rulemaking file."70

Under Executive Order 12,291, OMB had authority not only to review proposed regulatory initiatives by an agency, but also to initiate review of existing regulations inconsistent with Executive Order 12,29171 and to "[m]onitor agency compliance with the requirements of [Executive Order 12,291] and advise the President with respect to such compliance."72

[*pg 866]

B. Executive Order 12,498: Reagan Administration Regulatory Planning

In January 1985, just prior to the start of President Reagan's second term, he signed Executive Order 12,498, which was designed to "create a coordinated process for developing on an annual basis the Administration's Regulatory Program, establish Administration regulatory priorities, increase the accountability of agency heads for the regulatory actions of their agencies, provide for Presidential oversight of the regulatory process, [and] reduce the burdens of existing and future regulations."73 Executive Order 12,498 established a "regulatory planning process" for developing and annually publishing a regulatory program.74 Whereas the centralized regulatory review process set up under Executive Order 12,291 was retrospective, the regulatory planning process envisioned by Executive Order 12,498 was prospective.

Executive Order 12,498 required each agency to report annually to OMB on its plans and objectives for the upcoming year75 and to plan consistently with agency and administration goals and principles.76 In its review of an agency's proposed regulatory agenda, OMB would consider whether the agency's plans were consistent with those of the administration and other agencies,77 proactively identifying any action the agency should (in OMB's view) take to make its plans consistent.78 Any disagreements between OMB and the agency regarding the agency's draft regulatory program could be reviewed by the president or his designees at the request of either the agency head or OMB.79

The regulatory planning process created by Executive Order 12,498 was designed to complement the regulatory review procedures established under Executive Order 12,291.80 The regulatory review procedures under Executive Order 12,291 typically reflected after-the-fact oversight by OMB. That type of ex post review maximized [*pg 867] interagency conflict and, at the same time, constrained the ability of OMB to influence agency regulatory and deregulatory agendas "[t]o assure consistency with the goals of the Administration."81 In contrast, the regulatory planning process contemplated by Executive Order 12,498 allowed OMB to exert its influence earlier in agency decisionmaking. It also provided an important vehicle for necessarily involving political appointees at the agencies in the process of developing a regulatory or deregulatory agenda within the agencies themselves. Since the agency head was specifically charged with ensuring that the agency's program was "consistent with the goals of the . . . Administration,"82 and with "adher[ing] to the regulatory principles stated in . . . Executive Order No. 12291,"83 Executive Order 12,498 pressured politically accountable officials in the agency to participate actively in the early agenda-setting stages of agency decisionmaking.

C. Executive Order 12,866: Clinton Administration Regulatory Review and Planning

The Clinton administration's Executive Order 12,86684 retained both the basic process85 of the Reagan-Bush centralized presidential regulatory review structure and the substantive emphasis on cost-benefit analysis.86

In some ways, though, Clinton's executive order assumed even more presidential authority over agencies than Reagan's orders had.87 For example, "the Reagan orders . . . exempted the independent agencies," while "the Clinton order included them to a certain extent," reflecting a "strong commitment to presidential oversight of administration, extending even beyond the executive branch to the [*pg 868] independent agencies."88 Moreover, beyond the regulatory review process, "presidential control of administration . . . expanded significantly during the Clinton Presidency, moving . . . to the center of the regulatory landscape."89 As Professor Kagan, a Clinton administration insider, describes, President Clinton "came to view administration as perhaps the single most critical . . . vehicle to achieve his domestic policy goals," developing "a set of practices that enhance[d] his ability to influence or even dictate the content of administrative initiatives."90

[*pg 869]

Despite this more "expansive" scope, the Clinton executive order "produced fewer battles between OMB and the agencies."91 In part, this resulted because the pro-regulatory initiatives of the Clinton administration better matched the proclivities of the agency's regulatory-oriented rulemaking staff 92 (exactly the problem the Reagan-Bush orders were designed to address). But this decreased level of interagency rancor also may have resulted from some aspects of the Clinton executive order that were seen by agency staff, outside special-interest advocates, and some commentators as ameliorative.93

These ameliorative provisions set an explicit timetable for OIRA's review of agencies' proposed actions,94 "thus preventing OMB from using delay as a tool for pressuring agencies to revise proposed rules."95 In addition, Executive Order 12,866 seemed to afford [*pg 870] greater transparency in regulatory conduct. The Regulatory Working Group offered an expanded opportunity to participate in "identifying and analyzing important regulatory issues."96 The order's expanded accountability and disclosure provisions contributed to transparency.97 For example, when OIRA returned a regulatory action to an agency for further consideration, it had to "provide the issuing agency a written explanation for such return."98 An elaborate set of disclosure requirements was set forth, ostensibly aimed at "ensur[ing] greater openness, accessibility, and accountability in the regulatory review process."99 More substantively, pro-regulationists found the Clinton administration's executive order to mandate a more nuanced cost-benefit approach because it "rhetorically acknowledge[d] the frequent legitimacy of regulation" and "caution[ed] against an overreliance on 'hard variables.'"100



This Part moves from the descriptive to the analytical. The Reagan, Bush, and Clinton executive orders clearly established centralized presidential regulatory oversight in practice, but debate continues about its desirability, legality, and methods of implementation. This Part examines that debate, beginning with an inquiry into whether the debate, which nominally concentrates on structural questions relating to decisionmaking authority, focuses on real or strategic issues.

[*pg 871]

A. A Cynic's Question: Are the Structural/Analytical Issues Real or Strategic?

In reviewing the debate over centralized presidential control of agency rulemaking, one wonders how much criticism is about principle and how much is about substantive outcomes. Is the critique of centralized presidential regulatory review a strategic concern about political influence and policy outcomes, rather than a principled structural objection to increased presidential oversight of the regulatory process?

Some critics of the Reagan and Bush regulatory review executive orders have stated candidly that the centralized review process was "biased against regulation,"101 a "profoundly anti-regulatory phenomenon" that "threatens the legacy of the New Deal."102 Others have acknowledged their hostility to a "myopic vision of the regulatory process which places the elimination of cost to industry above all other considerations,"103 but have couched their critique in more generic terms "[q]uite apart from whether one agrees with the anti-regulatory bias of [the Reagan] Administration."104 And still others have supported the principle of centralized presidential regulatory review, while criticizing its implementation in certain respects.105

Criticisms of centralized presidential regulatory review, so pronounced in the deregulatory Reagan era, quieted under the regulation-friendly Clinton administration.106 Legal academics have been slow to recognize, much less criticize, the Clinton administration's more aggressively centralized presidential involvement in agency decisionmaking.107 A cynic would think that the lack of conflict during the Clinton administration most likely emanated from the relatively pro-regulatory stance of OMB-OIRA.108 Professor Kagan readily acknowledges, for example, that the "Clinton OMB chose to implement" regulatory review "in a way generally sympathetic to regula- [*pg 872] tory efforts," that "objections to OMB review in the Reagan and Bush era arose in large part from its deregulatory tendencies," and that the "reversal of substantive direction [during the Clinton administration] contributed to the waning of interest in, and even recognition of, the involvement of the President and his [Executive Office of the President] staff in administration."109

Support for the cynical view also could be mustered from early experience in the George W. Bush administration. Thomas A. Scully, administrator of the Centers for Medicare and Medicaid Services (CMMS) in the Department of Health and Human Services (HHS), apparently initiated within CMMS a process of rethinking nursing home regulatory enforcement policy.110 The goal was to "refocus enforcement efforts to reduce the frequency of inspections at nursing homes that had good records of compliance with federal health and safety standards."111 When confronted with this idea, Bush administration spokesman Ari Fleischer "denounced" the CMMS plan, which was being characterized in the press as a plan "to ease regulatory requirements on nursing homes" and asserted that the administration intended to "beef up," rather than ease, nursing home regulations.112 Mr. Fleischer averred that the White House had "rejected out of hand" two weeks earlier "[a]ny options dealing with any changes in the nursing home procedures that could lead to less inspections."113 Scully moved "briskly" to "align his agency with the White House position enunciated by Mr. Fleischer," stating that he "never knew it was discussed at the White House. . . . I didn't know there was a meeting."114 The thrust of the news story in the New York Times was not that the White House was impermissibly intruding on the regulatory turf of Scully's agency but that Scully was something of a loose cannon, catching the White House off guard by speaking out on issues "when White House officials were trying to keep a low profile."115 As a result of the flap, the article suggested that the White House "was still trying to work out the lines of control and communication with agencies of the federal government."116 The clear implication was that [*pg 873] presidential authority existed, as a matter of course, if the White House could only "work out the lines of control and communication."117

The nursing home story supports the skeptics' view about critiques of the centralized regulatory process. The article did not address or even mention the issue of centralized presidential regulatory oversight; the exclusive focus was on the substantive content of the proposed regulatory action, backing the administration off its purported flirtation with an initiative deemed deregulatory in character. The strategy worked. When the Bush administration announced that it would strengthen, not ease, regulatory oversight of nursing homes, that was seen as a victory by nursing home advocacy groups. Any concern about the propriety or impropriety of centralized presidential regulatory review, if it existed, was left unvoiced, and the issue was not raised or addressed.118 Scully, the CMMS administrator, who was in charge of the issue as agency chief and who had floated the deregulatory initiative, appeared to be out of the loop on an issue that involved the regulatory jurisdiction of his own agency. Yet, he had no defenders of his administrative turf against presidential encroachment or overreaching. The structural critics of centralized presidential regulatory review were silent.

As this example illustrates, different presidential agendas can be promoted through a centralized system of presidential authority or influence over agency rulemaking. Accepting that proposition, it is appropriate now to consider the structural issue on its merits.

B. The Unitarian Accord: The Bright Line Between Independent Agencies and Other Executive Agencies

Professor Kagan contends that the "common ground" for discussing the issue of centralized presidential control or policy-based [*pg 874] influence over discretionary conduct of nonindependent executive agencies is that the "current system of administration is not strongly unitary."119 But her analysis actually supports quite a different conclusion: in one form or another, we all seem to be unitarians now. Despite Kagan's disagreement with the notion that the unitarian position is constitutionally mandated,120 her own defense of centralized presidential authority over nonindependent executive agency decisionmaking is aggressively unitarian. Kagan's unitarianism regarding EOP authority over and policy-based influence on discretionary regulatory conduct of nonindependent agencies is significant because, from the inception of centralized presidential regulatory review in 1981 during the Reagan administration, such agencies have been the site of the dispute over unitarianism. The Reagan executive orders did not assert such EOP authority on or influence over the independent agencies.

The Clinton administration not only accepted, but also extended the unitarian premises of the Reagan and Bush administrations. For example, Professor Kagan acknowledges that the Clinton administration's executive order, in contrast with earlier versions, claimed, at least "implied[ly]," the "authority ultimately to displace the judgment of agency officials," an authority that "even [President] Reagan . . . had disclaimed."121 That is, the Clinton administration apparently asserted "presidential directive authority over discretionary decisions assigned by Congress to specified executive branch officials (other than the President)."122 And that authority was asserted aggressively without the additional authorizing legislation that the ABA and former Carter administration Counsel Lloyd Cutler had recommended.123 This is a strongly unitarian position in that it assumes authority in the EOP to command agency decisionmaking, essentially substituting presidential priorities for those of agencies and displacing the exercise of enforcement discretion nominally lodged with agency leadership.124

[*pg 875]

Thus, Professor Kagan's vision (ostensibly expressing the Clinton administration's view) is quite unitarian. While rejecting the constitutional basis for this unitarian stance, Professor Kagan (and other nonconstitutional unitarians)125 reach their position on grounds of policy and through principles of statutory interpretation. To reach this unitarian end, Professor Kagan advocates a presumptive "interpretive principle" -- that, absent an express statutory provision that precludes presidential involvement in agency decisionmaking, a "standard delegation" to a nonindependent executive agency should be "read . . . as including the President."126 This means that the unitarian approach guides administration of executive agencies other than independent agencies, which were excluded from the Reagan and Bush executive orders and whose decisionmaking Professor Kagan recognizes to have been "insulate[d] . . . from the President's influence."127 I agree with Professor Kagan that, irrespective of how one might have wanted the law to develop normatively, this is a sensible way to read the cases128 and an appropriate methodology for both asserting considerable presidential authority over agency decisionmaking and respecting the statutorily conferred decisionmaking autonomy of independent agencies.

Although the distinction between EOP oversight of independent and nonindependent agencies seems to have secured greater consensus since its recognition by the Clinton administration, that distinction was not well recognized or accepted at the time of my OIRA administrator confirmation hearings. In my confirmation hearings, that is precisely the distinction I drew, to the edification (and consternation) of some members of the Senate Committee on Governmental Affairs, which has confirmation authority over the OIRA administrator (and oversight authority, in general, over OMB). As my experience illustrates, at least some members of the Senate Committee on Governmental Affairs found the distinction "interestingly described" and po- [*pg 876] tentially "troubling"129 and, at the time, not at all generally recognized or accepted on the Committee.

In a colloquy with Senator Lieberman,130 I asserted that OIRA's goal should be "a sensible decisionmaking process."131 This requires recognition that, "ultimately the Executive Branch is administered by the President, and within the confines of the legislation and the policies enacted by the Congress, to which you have to be faithful,"132 there is "wiggle room"133 or discretion on "how you are going to approach something within the confines of the Congressional mandate."134 That "calls for judgment, and ultimately it is the judgment of the President that has to prevail in that area."135 In response, Senator Lieberman asked about the "limits of OIRA's authority"136 since Congress "specifically delegated authority to the various agencies to carry out those laws, and with it, the power to implement regulations."137 For Senator Lieberman, the "cutting edge question"138 was whether OIRA can "hold up an agency regulation or return it because it disagrees with it on policy grounds."139

My answer to Senator Lieberman was straightforward: "I think there is a bright-line distinction that I would draw"140 regarding congressional delegation of authority to an agency. In the context of an "independent regulatory body such as the . . . Federal Communications Commission,"141 the president (through OIRA) cannot hold up an agency action because of a disagreement "on policy grounds,"142 and the "Executive Orders do not purport to establish a regulatory review process that controls those independent regulatory agencies."143 Senator Lieberman stated his view that OIRA had in fact [*pg 877] conducted regulatory review based on policy disagreements with the agency in question.144 I responded that the senator's example (the EPA) "is not an independent regulatory agency . . . . The President asserts jurisdiction over that agency" because the EPA "is not like the FCC. . . .There are differences. . . . [A]gencies that fall within the ken of the President's jurisdiction" are subject to the "exercise of supervision" on the part of OIRA "to be sure that those agencies are following Executive [Office of the President] policies . . . that . . . they are consistent with the parameters set forth by Congress."145 That is, there was no claim that the EOP "would have authority to countermand legislative judgments."146 However, where administrative discretion exists within a statutory framework, the choice among valid alternatives is for the "President [to] determine[,]"147 and his choice "is the one that should be pursued."148

Seeing the distinction I was drawing, Senator Lieberman restated my position that independent agencies were not subject to OIRA regulatory review but that "agencies that are more generally under the umbrella of the President's executive authority really deserve . . . less deference when their regulations come to OIRA, because OIRA plays a role in expressing the will of the Administration and of the President."149 My reply was that "OIRA is . . . the eyes and ears of the President. . . . So I don't think that it is inappropriate for the President to assert authority in the areas in which he has jurisdiction."150 Senator Lieberman's parting shot from this colloquy was that I had "interestingly described"151 the process and that "we may look back and say that your answer was a troubling one . . . or that it was a warning -- but that OIRA plays this role to carry out the policy of the President so long as it does not frustrate the intent of Congress."152

Thus, there is indeed much common ground between the position on centralized presidential regulatory review between Professor Kagan (ostensibly expressing the views advocated by the Clinton administration) and me (expressing views representative of those advo- [*pg 878] cated by the Reagan and Bush administrations). Although Kagan does not characterize her position as unitarian, once one pierces to the substance there is remarkable accord. This post-Clinton administration consensus surely was not a consensus a decade ago during my OIRA administrator confirmation process, when these issues were controversial and the distinctions, for purposes of centralized presidential regulatory review, between independent agencies and other executive agencies were not fully understood or accepted.

C. The Technocratic Premises

If there is accord on what I characterize as the unitarian stance, the premises of the stance are still debated. In this Section, I conclude that the technocratic rationale and its claims for centralized presidential regulatory review are widely accepted.153 In the following Section, I consider the more hotly contested substantive policy rationale for centralized presidential regulatory review.

The technocratic claims for centralized presidential regulatory review are now recognized as uncontroversial, at least in principle. These claims focus on "good government" themes that are compatible with the technocratic vision of agency rulemaking through use of specialized knowledge and expertise. In this uncontroversial category, I would include the following rationales:

1. Managing Regulatory Externalities. Some system, it is widely acknowledged, is needed to manage situations in which the conduct of one agency, in the exercise of authority conferred on it, can substantially affect interests or policies that fall within the purview of a different agency.154 Some mechanism for coordination across agencies is needed to mediate these interagency issues. In this regard, "the President is in a good position to centralize and coordinate the regulatory process."155

[*pg 879]

2. Achieving Better Quality Analysis in the Rulemaking Process. As the Clinton administration recognized and addressed with its Executive Order 12,866, an interagency "forum" for identifying and dealing with generic, interagency cross-cutting regulatory issues can be highly beneficial. Thus, Executive Order 12,866 established a "forum to assist agencies in identifying and analyzing important regulatory issues" such as the "development of innovative regulatory techniques" and the "development of . . . streamlined regulatory approaches for small businesses and other entities."156 The decisionmaking process of one agency often would be improved if it were to make beneficial use of data compiled by or the expertise present in another agency.157 The first two administrators of OIRA saw as perhaps the "greatest benefit of OMB review" improved agency ability to "respond to the kinds of questions that OMB raises."158 This was achieved when agencies "either established or enhanced their in-house capabilities to analyze their regulatory decisions."159 Thus, the very existence of external review can improve an agency's decisionmaking process by keeping the agency on its analytical toes. There is a value in keeping agency bureaucratic decisions intellectually honest and analytically rigorous, and external, centralized presidential regulatory review can bring this about by bringing to bear a new set of perspectives and analytical tools.160

3. Imposing Regulatory Uniformity. As the Clinton administration's executive order also recognized, there may be a value to establishing a uniform government-wide regulatory approach, across agencies, to certain kinds of recurring issues.161 This could include [*pg 880] establishing a methodology for performing cost-benefit analysis, developing a norm across agencies for cost-effectiveness, or developing an administration-wide approach for assessing comparative risk.162 Centralized presidential regulatory review provides a useful and beneficial mechanism for this type of interagency normalization.163

D. The Substantive Policy Issues

Centralized presidential regulatory review is well established in practice, and many of the technocratic advantages of the process are uncontroversial, at least in principle. More contentious are the substantive policy issues, which focus on its desirability, its legality, and the methods of its implementation.

In considering the substantive policy issues associated with centralized presidential regulatory review, this Section initially explains that the debate has been so heated in large part because analytical combatants come to the fray from such divergent analytical perspectives. Next, I contend that, on substantive policy grounds, centralized presidential regulatory review is a warranted process of institutionalizing presidential programs and priorities in agency rulemaking. In that context, I examine the influence of centralized presidential regulatory review on improving decisionmaking accountability within the executive branch. The vehicle for that discussion is a focus on three distinct sets of decisionmaking interrelationships.164 Finally, I show that existing constraints on centralized presidential regulatory review mitigate the risk of excessive politicization that critics of such a process condemn.

1. The Importance of an Analytical Framework. In the debate over centralized presidential regulatory review, as elsewhere, how one thinks about an issue and the way an issue is framed shape the way one analyzes it. An example of this phenomenon from a different context sets the stage. In health policy, there is a critical threshold issue: should public policy analysts think about health care as a techni- [*pg 881] cally driven, scientifically based professional social service (the professional paradigm) or as a market commodity (the market paradigm)?165 Advocates of the professional paradigm, the traditional model, note the asymmetry of information between doctors and their patients166 -- doctors are professional experts, and consumers lack the information or expertise needed to make informed choices (a prerequisite for a smoothly functioning market). Professional paradigm proponents at one time indulged an assumption that, empirically, economic incentives do not affect medical decisionmaking.167 That assumption is now more normative than empirical: proponents can only claim that economic incentives should not have such an effect. Thus, "[m]arket factors such as cost-benefit trade-offs are not only seen as irrelevant, but as corruptive of medical judgments."168

The professional paradigm "places power paternalistically in physician hands," with physicians "perceiv[ing] themselves as controlling decisionmaking."169 In this model, medical decisionmaking is viewed as "technical and scientific," with "[q]uality-cost trade offs [being] attenuated."170 The consequence is that "economics and trade offs become marginalized in the policy debate," since medical care is seen as an "exclusively technical-scientific enterprise."171 Under the professional paradigm, economic considerations undermine and corrode what are and should be technical scientific judgments.

The introduction of managed care reflected a move along a continuum toward an economic model, in which cost-benefit analysis and [*pg 882] economic trade-offs were taken seriously. Managed care "introduced an economic dimension to medical decisions," with HMOs asking not only whether a particular diagnosis or treatment is safe and effective -- the issue under the professional model -- but also whether the "treatment, even if effective, is of sufficient benefit or priority to warrant expenditure from a common pool of insurance money."172

The confrontation between market and professional paradigms can generate considerable vitriol. Some courts have reacted harshly, even punitively, to the market-oriented model. This is what one would expect from a clash of paradigms -- when a decisionmaker with one model in mind confronts a set of decisions based upon a different paradigm.173 The movement to enact a "patients' bill of rights" reflects, to a large extent, a backlash against the market-oriented model represented by managed care and a revanchist attempt to restore the professional paradigm, together with its professional prerogatives for physicians and its spare-no-expense style of practice (typically paid for out of a common insurance pool in the private sector or from tax funds for public beneficiaries on Medicare or Medicaid).174 The intense acrimony associated with this effort is illustrative of what happens in the policy arena when competing paradigmatic visions come into conflict.

This brief excursion into the world of health policy is designed to elucidate some of the harshness that has attended the debate about centralized presidential regulatory review. That debate entails two paradigm confrontations: one policy-based, the other structural.

The policy-based clash of paradigms is very similar to the health policy issues described above. In areas such as worker safety or environmental protection, the introduction of cost-benefit analysis by economic rationalists rankles those who hold, for example, a stewardship view of the environment, a secular-religious view quite hostile to considering economic trade-offs (and a view embodied in some health [*pg 883] and environmental legislation).175 This debate transcends nitty-gritty technical questions of how to measure costs and benefits and pits partisans with different teleological views against each other -- one side seeking to impose rational economic thought on the process, the other rejecting that very enterprise as corrosive of core secular-religious values and ideals.176 Thus, introduction of centralized presidential review that, from the outset, has been associated with a cost-benefit mindset was inherently and inevitably destined to become entwined inextricably with this deeply-rooted confrontation of weltanschauungs. The sharpness and intensity with which these issues have been debated bear the typical hallmarks of this provenance.

The structural divide is no less chasmatic. Many of the most vocal critics of centralized presidential regulatory review have a vision of the regulatory rulemaking process that, wrongly and unrealistically in my view, romanticizes the independence and expertise of agency decisionmakers. The strenuous objections to interagency contacts between line, or rulemaking, agencies and OIRA177 and to contacts between OIRA staff and nongovernmental personnel178 seem born of a vision of the rulemaking process that takes as its model the judicial process.179 In the judicial process, ex parte contacts are highly inappropriate, as they are in the context of administrative agencies per- [*pg 884] forming adjudicative functions. But rulemaking is not a purely scientific function; it also encompasses a heavy dose of normative judgment. The sense of impropriety here among critics seems founded on their view that "the expansion of presidential administration impinges on the neutral application of substantive knowledge and skills to regulatory problems."180 From that perspective, the problem is that politics is being substituted for expertise in the rulemaking process and that "politics is anathema to efficient administration."181 As politics becomes a more visible and legitimized component of agency rulemaking, these critics confront a trend they view as inappropriate and even corrosive.

Thus, it is not surprising that centralized presidential regulatory review has been criticized in such strong terms. Although foundational teleological disagreements remain, the structural issues become much less polarizing if one acknowledges that agency rulemaking is an admixture of expertise and normative policy judgment.182

2. Substantive Policy Claims for Centralized Regulatory Review. I now argue that centralized presidential regulatory review is justified on policy grounds and that the issues ultimately are more political than legal in character (although, in separation of powers matters, the political can influence the legal, at least in the long term). To frame the normative/policy discussion, I address three sets of relationships: (1) intra-agency relationships, between the bureaucratic/technical staff and the politically accountable presidential appointees ostensibly supervising their work product; (2) the relationships between EOP (OMB-OIRA) and the line agencies; and (3) the relationships that the executive branch has with non-executive branch constituents, both governmental (Congress) and nongovernmental.

a. Empowering Political Appointees Within Agencies. Contrary to the romantic view, agency rulemaking is not under the ultimate, formal control of technical experts. Political appointees supervise and are, at least nominally, in charge of agency decisionmaking. Thus, representatives of the president and his administration have always [*pg 885] been an important, even fundamental component of the agency decisionmaking process.

An administration whose policies and goals may be out of sync with the culture or values of an agency (and how that agency exercises its discretion in rulemaking) is well advised to improve the oversight role of its political appointees within the agency. It is fair for this point to travel under the label of improved political accountability.

Political appointees are notorious short-termers, a point made to me in my confirmation hearings by Senator Herb Kohl (who asked for a longer-term commitment from me).183 Career public servants who staff the agencies provide the historical memory, the expertise, and, not insignificantly, the human capital. Stories abound of staffers who have left their disapproved work product in a desk drawer until the political appointee who disapproved it moved on. Further, the political appointees may not understand or be committed to the administration's policies or priorities; consequently, it is the agency staff that shapes work product and presents options to an agency's political leadership. Delaying the presentation of staff work until a deadline looms is a time-honored bureaucratic strategy by which agency staff can thwart thinking that goes outside the box.184

Thus, empowering and thereby enhancing the political accountability of agency political appointees is as important a goal of centralized presidential regulatory review as improving the technical performance of agency decisionmaking.185 Since agency-level political appointees make normative policy judgments, and serve at the pleasure of the administration, it would appear appropriate that the administration take steps to educate them about the administration's priorities and to encourage accountability to the administration. Oversight from outside the agency can bolster the authority of political appointees, empowering them within the agency. Appointees may have limited political capital in the EOP to expend to support agency objectives. Staffers who want political appointees to use their political capital to promote staffers' agenda have an incentive to cultivate their [*pg 886] agency's political leadership, to husband these political resources, and to combat bureaucratic manipulation by other staff hostile to administration objectives.

Thus, agency rulemaking has always contemplated a blend of both technocratic and political concerns. It is hard to see a principled reason why a presidential administration should not identify and explain its policies to its political appointees and help political appointees actively participate in agency rulemaking at all stages. One of OIRA's important roles, in that regard, is to alert agency political appointees when career staffers' positions are at odds with administration goals and objectives. Without much of an independent staff of his or her own, that agency appointee may not have the knowledge or the resources to become an active player in the bureaucratic infighting within his or her own agency.186 OIRA can pinpoint issues, provide a form of expertise that may be lacking within the agency,187 help formulate administration policy on a common law, case-by-case basis, and seek to impose that vision on the agencies' rulemaking process through the agencies' political appointees.188

b. Providing a Vehicle for Implementing the President's Policies and Programs in Discretionary Agency Decisionmaking. The foregoing discussion presages the consideration of the relationships between EOP (OMB-OIRA) and the line agencies in the rulemaking context.

The rulemaking process typically involves a wide set of issues that are only partly technocratic in character. How should an agency allocate its time in areas in which enforcement initiatives are not mandatory but discretionary?189 How does an agency account for such administration values as respecting federalism in performing its rulemaking tasks or preferring incentive-based rather than command-and-control regulatory strategies?190 These are all normative considerations instrumental to agency rulemaking. Making policy that ac- [*pg 887] cords with these policy preferences necessarily requires technical expertise, data acquisition and analysis, and specialized knowledge of a substantive area. The technocratic input required, though, is not the whole story. Its relevance and usefulness ultimately depend on an agency's overarching approach to a problem. The normative and substantive policy preferences that create the overall framework for developing specific implementing rules and regulations are fundamentally political.191

Given that agency rulemaking does contemplate a role for political input -- at the very least on the part of political appointees within the agency -- it becomes hard to see why a presidential administration should not direct or at least influence its agents' exercise of discretion within the agency bureaucracy. Good private-sector management may call for decentralized decisionmaking, and many public-sector "good government" reforms would warrant more autonomy for agency leadership. But where agency rulemaking implicates fundamental components of an administration's political agenda, that kind of hands-off management style seems questionable.

Absent special circumstances,192 political appointees in executive agencies (other than independent agencies)193 are subject to the political control of the president,194 although it may at times be very expensive politically for a president to exercise that control.195 Through directives to agencies, President Clinton formally exerted authority [*pg 888] over agency agenda-setting and priority-setting activities.196 But, as Professor Kagan notes, the issue of formal legal authority is less important than the development of a mechanism through which the president's values and priorities can be identified and communicated to political appointees in agencies. In a multifaceted organization such as a government, "a President may face considerable constraints in imposing his will on administrative actors," even when asserting authority as President Clinton did.197

As a result, "authority," a comfortable model for lawyers, may be the wrong way to think about centralized presidential regulatory review of agency decisionmaking.198 A regulatory oversight structure allows a president, through the EOP, to assert centralized administration priorities to guide administration appointees in the agencies. But this triggers a negotiation model rather than a command-and-control corporate model. It requires an agency appointee to resist presidential directives, which that appointee also does at a price. Centralized regulatory review provides monitoring by the EOP, a forum for agency appointees, and a vehicle for a centralized resolution (typically, as under the Clinton administration, by the vice president). Once the process plays out, the agency appointee can either comply, continue to undermine the administration covertly (which will wear thin after a while in the EOP), resign, or force the president to fire him or her.

Especially for an administration whose political agenda runs counter to longstanding political or cultural traditions within an agency, some form of centralized regulatory oversight is essential. It is also highly desirable for an administration, like President Clinton's, that wants to move an agency in a direction that the agency's staff may not resist.199 The formal assertion of authority, which characterized President Clinton's directives, may over time set a tone and shift [*pg 889] the balance of power between the EOP and agency staff.200 This is how the "law" of separation of powers seems to work, as Justices Frankfurter and Jackson pointedly and sophisticatedly observed years ago in Youngstown Sheet & Tube Co. v. Sawyer.201 In this sense, the institutionalization of centralized presidential regulatory review for over twenty years seems likely to embed it within our understanding of how the modern presidency works (and, hopefully, soften the harsh polarized commentary of the past and focus it more on substantive policy issues). In this way, centralized presidential regulatory review also has shifted the balance of power within the executive branch. Regardless of whether the president exerts "authority" rather than persuasive "influence" over his political appointees within an agency is surely less important in explaining this shift in the balance of power202 than the president's ability to command respect and loyalty from his appointees and effectively to cajole agency appointees to adhere to EOP policies as they emerge in the centralized presidential regulatory process.203

c. Exerting Presidential Authority Over or Influence on the "Iron Triangle." The third set of relationships to consider is that among the executive branch, Congress, and nongovernmental actors. This set of [*pg 890] relationships focuses around what has traditionally been described as the "iron triangle" -- a congressional committee or subcommittee and its staff, an agency staff, and a set of organized interests or interest groups.204 Although the degree of congressional influence on agency decisionmaking is a matter of some debate,205 my experience in the OIRA nomination and confirmation process leads me to conclude that the influence can be quite substantial, at least with respect to some agencies and some committees.

The "iron triangle" theory suggests that stakeholders in an agency, on congressional committees (and their staffs), and among nongovernmental interests negotiate on regulatory issues facing an agency.206 Free informal discussion between agency staff and nonagency personnel is routine and contemplated in the rulemaking process, particularly in the early stages of deliberation on a certain set of issues.207 In that kind of environment, objections to OIRA oversight are ironic (but upon reflection quite understandable) because they suggest that the EOP has less claim on influencing agency decisionmaking than non-executive branch governmental players and even private interested parties.

The objections seem to stem from a perception on the part of the iron triangle stakeholders that they represent the universe of interested parties.208 This is a form of agency capture, with the self-appointed stakeholders in essence deeming themselves the exclusive set of persons with whom negotiations need be carried out. Once these stakeholders within the triangle strike a deal, they expect it to [*pg 891] be respected.209 Parties engage in a give-and-take process and expect the result of the bargain to become law. Outside players, such as OIRA, are seen as intruders on the triangle's turf, particularly when they exercise what is perceived as after-the-fact influence.210 In such circumstances, hostility to after-the-fact OIRA review is predictable and understandable.211

An analogy exists to a university's appointment or tenure-review process. Once a department or school evaluates a candidate's teaching and research, solicits outside letters of recommendation and evaluation, and has a robust debate about the merits of a candidate's file, the decision taken by that unit secures a sort of institutional legitimacy. External review by university-wide entities seems always to engender resentment, particularly when that external reviewer reaches a conclusion at odds with that taken in the department or the school. This is a challenge not only to the original unit's expertise and judgment, but also to its internal process of dialogue and debate that resulted in an affirmative vote. OIRA review has the hallmarks (including animosity) of a university-wide tenure-review committee turning back a tenure file to a department or a school with an admonition to do a new, more deliberative review of the file. The question is whether the EOP should have less of a role than the members of the iron triangle involved in a particular regulatory matter, especially congressional committee or subcommittee members and their staffs. The answer should be "no."

Congress and its committees have a number of constitutionally appropriate ways to influence agencies. They can conduct oversight [*pg 892] hearings and publicize their findings. They can slow walk or, in some situations, block the confirmation of political appointees. Committees often, in practice, control agency resources and jurisdiction as part of the authorization or reauthorization process or as part of the annual appropriations process. Additionally, committees initiate legislation and can impose specific mandates on agencies, controlling or limiting the scope of agency discretion.

Congressional influence is constitutionally limited, though. Under Chadha, Congress may act to make or implement the law only in its corporate capacity, following the requirements of bicameralism and presentment.212 Neither congressional committees nor their staffs have constitutional authority to affect the legal rights of individuals outside the legislative process.213 Further, Congress cannot control the implementation of the law, which is an executive branch function.214 Because of these limitations, Congress cannot claim to preclude executive influence on agency decisionmaking, at least outside the formal legislative process. Of course, as part of a committee's oversight function, a committee may influence an agency's decisionmaking, but that is a far cry from precluding EOP oversight of agency decisionmaking.

With respect to contacts between executive branch officials and nongovernmental individuals or groups, the administrative process allows for ex parte contacts during rulemaking and does not appear to require disclosure.215 Since nongovernmental groups can and do contact agency staffs, there should be no special concern when nongovernmental groups contact the EOP. Because a reviewing court will examine an agency's record, it is probably prudent to document and place in the public record "communications by interested parties (such as industry) to the OMB and the White House staff that provide significant factual information."216 But policy-oriented conversations with EOP should be regarded as the legitimate expression of the constitutionally protected right of private individuals or groups to petition their government. The president should know the impact of a [*pg 893] regulation or the position of a constituent group on a particular set of issues. If a particular perspective cannot be squared with the process of reasoned judgment imposed on agency decisionmaking, or if it does not comport with the requirements of the underlying legislation, it should not be followed, and an agency would be within its authority in declining to adopt such a position. But banning such contacts would be "troubling."217

3. Constraints on Centralized Presidential Regulatory Review. Acknowledging the importance of centralized presidential regulatory review does not mean accepting that the president or the executive branch generally should stray too far from the technocractic vision of agencies. Although centralized presidential regulatory oversight of agency discretion is appropriate, important constraints exist on a president's ability to move too far down the technocratic-political continuum toward politicization and away from the technocratic-scientific norm.218

First, the executive branch as an undifferentiated whole is constrained to operate within the parameters set forth in the specific legislation being implemented through the agency rulemaking process. This requirement is enforced by courts in challenges to agency rulemaking and enforcement action.219 The question is how and through whom the executive branch exercises the discretion left open by the legislation. The locus of executive branch authority to exercise legislatively conferred implementation discretion is the key issue.220

[*pg 894]

Second, administrative law places significant limits on the EOP's ability to politicize agency rulemaking. To change a rule, an agency must show more than that the president campaigned on a platform different from the one embodied in the rule at issue and that the EOP commanded a change.221 Under the hard look doctrine, courts protect the technocratic component of an agency's deliberative process, requiring a process of reasoned judgment in which an agency explains what it does and why. An agency's judgments should "address all significant issues, take into account all relevant data, consider all feasible alternatives, develop an extensive evidentiary record, and provide a detailed explanation of its conclusions."222 A court will evaluate an agency's rulemaking process based on the record before it and whether that record "adequately support[s]" the agency's decision.223 If an agency's "proffered reasons . . . do not adequately explain" its decision, the agency will face judicial remand "for more reasoned decision-making."224

These constraints place boundaries on and restrain the exercise of presidential control over or policy-based influence on an agency's exercise of discretion. A critic concerned about the nation's moving away from a technocratic vision of agency conduct based on expertise can take comfort in the fact that these constraints limit any president's ability to move too far away from that technocratic vision.



Since 1981, centralized presidential regulatory influence over the regulatory process of executive agencies has been practiced in four [*pg 895] ways, two by the Reagan and Bush administrations and two by the Clinton administration.

The first technique was the after-the-fact review of an agency's work product, usually at a late stage in its development, by the Reagan and Bush administrations pursuant to Executive Order 12,291.225 The second was the forward-looking regulatory planning process created by Reagan's Executive Order 12,498 and largely retained by Clinton, which allowed OIRA to shape agencies' policies and agendas at an earlier stage of an agency's planning process.226

The third technique was President Clinton's practice of proactively issuing directives to agencies, couched as legal mandates.227 These directives imposed presidential priorities and agendas on agencies and were an important tool for President Clinton to impel agencies to action in support of his administration's goals and objectives.228 The fourth kind of influence was Clinton's appropriation of agency regulations as part of his administration's portfolio of achievements, thereby giving them political visibility and, through a feedback loop, indirectly influencing future agency behavior, shaping it to his administration's last.229 This Part concludes that prospective directives can be a useful tool for achieving a president's agenda -- whether consistent with agency priorities or not -- and that use of a forward-looking regulatory planning process can enhance the effectiveness of the centralized presidential regulatory review process.

A. The Use of Prospective Directives: Implementing a Proactive Agenda

The Clinton administration's sophisticated use of advance directives to drive agency decisionmaking took presidential oversight to a new level. Although the directives were couched in mandatory language, their success in most circumstances turned on agency leadership acquiescence. Interagency disputes typically do not end up in court; courts do not issue mandatory injunctions requiring agencies to comply with such presidential directives. So, despite the mandatory language,230 which can shape perceptions if systematically used over [*pg 896] time and if otherwise unchallenged,231 the core question is whether a president can use the directive technique when the president's agenda is in tension with the political preferences or culture of an agency's staff. In other words, was the directive technique successful for President Clinton because he started a policy ball rolling downhill? Could an administration attempting to rein in what it perceives as an overly aggressive agency staff achieve the same types of success? The answer is at this point unknowable.

Nonetheless, the use of advance directives has much to commend it as compared to post hoc review. One way of viewing the after-the-fact OIRA reviews under Reagan and Bush is that they were developing, in a common law, case-by-case manner, a set of administration policies. In effect there was no administration position until a decision by OIRA was made. This allowed political appointees in the agency to advocate personal or agency positions without having to confront the question of fidelity to the president and his policies and priorities. A political appointee in an agency could feel comfortable, before the fact, in advancing institutional agency positions without being forewarned that a particular policy was out of sync with administration thinking. Once the EOP reached a decision, agency appointees were already locked into a particular stance, probably feeling duty-bound to represent forcefully the agency's viewpoint.

The proactive, advance directive has the advantage of formulating an administration position at the front end. Responsible political appointees in the agency will have helped develop the directive, so the EOP will have coopted them at an early stage. Once the administration takes a position, opposition becomes disloyalty. My sense is that that type of clearcut directive could help mobilize agency political appointees even when they face a hostile professional staff. The staff may be able to make changes at the margins, but the technique ensures the active involvement of the president's political agents (his appointees) within the agency. In particularly controversial matters, an administration confronting the values or culture of an agency probably will have to make available centralized staff resources (for example, through OIRA) to help implement the directive. Once the president's position is on the table, though, there is less opportunity [*pg 897] for indirect opposition through bureaucratic game-playing within an agency's staff.

B. The Regulatory Planning Process

If, as should be the case, agency staffs can accept EOP staff input as a legitimate part of an agency's decisionmaking process, that input should come as early in the process as possible. After-the-fact review inherently maximizes interagency conflict. Linking the planning process established in Executive Order 12,498 with the substantive standards adopted in Executive Order 12,291, and perhaps with a more comprehensive set of substantive directives, might best harmonize the centralized system of presidential regulatory review with the technocratic norms present within an agency. Broadening the horizon of agency staff is an important objective, and introducing staff to each other before decisions get too far along is an important way of allowing the staffs to see each other as professionals and to respect each other despite their differences in perspective. Staffs from agencies and OIRA should be encouraged to view each other as colleagues and as partners in the process. But, as Donald Elliott has counseled, the process also must include a set of incentives for cooperation so that an obdurate agency staff is divested of authority.232 Although Congress, at least nominally, assigns authority to specific agency heads, it does not specify which career personnel will be deputized to do the staff work on a particular matter. A recalcitrant agency staff should face a form of bureaucratic receivership, in which it loses effective staffing authority on specific matters to another set of career staff (most likely operating through the EOP). Formal authority at the appointee level would not change, but the underlying staff work would be vested elsewhere during the period of recalcitrance.



This legal and institutional policy analysis of centralized presidential regulatory review has made the case that the development of centralized presidential review of agency regulatory activity is an appropriate and desirable fixture of the modern American presidency. [*pg 898] It can play a useful role when a president seeks to pursue an activist regulatory agenda, shaping agency rulemaking agendas and priorities. It plays an essential and legitimate role when a president seeks to pursue a regulatory policy that reins in what the White House perceives as excessively intrusive regulatory initiatives that emerge from agency staffs.

Remarkable consensus has emerged, after the experience of the Clinton administration, about the desirability and propriety of centralized presidential regulatory review. In at least one form, the unitarian position has emerged as an appropriate executive branch norm; this norm allows presidential influence on executive agency rulemaking, while recognizing that existing doctrine precludes such influence on independent agencies.

It is now widely recognized that centralized presidential review can improve the functionalist or technocratic decisionmaking process of agencies by managing regulatory externalities, achieving better quality analysis in the rulemaking process, and providing a mechanism for establishing appropriate regulatory uniformity across agencies on commonly faced issues (such as assessing risk).

Controversy on the normative or policy rationale for centralized presidential regulatory review has been, historically, particularly divisive. This is largely attributable to different visions of the rulemaking process and to foundational substantive policy disagreements. The structural divide stems from a core paradigmatic disagreement about the nature of the agency rulemaking process -- whether it should take as its model the judicial process or whether it should realistically acknowledge that normative and political judgments necessarily and appropriately have a critical role to play in agency decisionmaking. If one takes the former view, then presidential influence is perceived as illegitimate, and overt centralized presidential regulatory review is rubbing salt in an open wound, expressly acknowledging and legitimizing what, in the view of critics, is corrosive conduct. Under the latter view, which recognizes the validity of mixed technical and political judgments in agency decisionmaking, presidential involvement is desirable as a means of asserting the president's policies and priorities and thereby assuring political accountability.

Similarly, a substantive policy clash of paradigms heightens tensions because centralized presidential regulatory review has been associated with the introduction of cost-benefit analysis. In a number of areas, such as worker safety or environmental protection, overt consideration of costs is considered heretical in some quarters. Thus, an- [*pg 899] other paradigm confrontation has existed, raising the temperature of the debates. During the Clinton administration, the heatedness associated with these issues abated, just as the consensus on the propriety of presidential influence on agency decisionmaking emerged. It will be interesting to observe whether the consensus on the structural issues will remain once the substantive policy debates recur, as they are likely to do in the second Bush administration. Of course, that recurrence, if it occurs, will feed the claims of the cynics, who have contended all along that the structural issues were merely a camouflage for core disagreements on the underlying policy issues.

This Essay has contended that centralized presidential regulatory review is warranted on policy grounds. Important normative and policy values are served by centralized presidential regulatory review, and those values are consistent with the traditional understanding of how the agency rulemaking process works, both in theory and in practice. Such review educates and enhances the accountability of agency political appointees, who are the president's representatives in the agencies, and who traditionally have had formal responsibility for and authority over agency decisionmaking.

Centralized presidential regulatory review both is and should be legally authorized. In practical terms, the core issues are political rather than legal in nature, much as that dissatisfies lawyers and legal analysts. At the same time, traditional legal constraints on centralized presidential regulatory review -- for example, the rules that no exercise of discretion by the executive branch can violate the terms of a statute and that an agency's decision must be supported by the rulemaking record -- limit excessive politicization of agency rulemaking.

In sum, centralized presidential regulatory review has become an important and constructive feature of the executive branch. A consensus has emerged on the core structural issues. It remains to be seen whether that consensus will persist once an administration advances a policy agenda that places it in conflict with agency staffs and nongovernmental advocacy groups dedicated to substantially different policy goals and regulatory approaches.



Copyright 2001 by James F. Blumstein.
Centennial Professor of Law, Vanderbilt Law School. This Essay is based on a paper presented at the Duke Law Journal's Administrative Law Conference held at Duke University School of Law on March 5, 2001. The Duke University School of Law's Program in Public Law provided additional financial support for the 2001 conference. I would like to thank my Vanderbilt Law School colleagues Lisa Schultz Bressman and Michael P. Vandenbergh for their thoughtful and constructive comments on a draft of this Essay. Neither is, of course, responsible for the ultimate product or for any views expressed in the Essay. I also would like to acknowledge my intellectual debt to C. Boyden Gray, White House counsel during the administration of President George H. W. Bush, and to the Honorable S. Jay Plager, who served as administrator of the Office of Information and Regulatory Affairs in the Office of Management and Budget at the end of the Reagan administration and the beginning of the George H. W. Bush administration. Together, albeit at different times, they provided me with critical and sophisticated insights about the nuances of regulatory review by the Executive Office of the President. Neither is responsible for the views expressed herein, however.
1. Compare, e.g., Alan B. Morrison, OMB Interference with Agency Rulemaking: The Wrong Way to Write a Regulation, 99 HARV. L. REV. 1059, 1059-60 (1986) ("Congress should eliminate OMB's involvement in the rulemaking process . . . ."), with, e.g., Christopher C. DeMuth & Douglas H. Ginsburg, White House Review of Agency Rulemaking, 99 HARV. L. REV. 1075, 1088 (1986) (arguing for a centralized regulatory review process).
2. See, e.g., Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245, 2383-84 (2001) (arguing that an increased presidential role in regulation "both satisfies legal requirements and promotes the values of administrative accountability and effectiveness").
3. See, e.g., Cynthia R. Farina, Undoing the New Deal Through the New Presidentialism, 22 HARV. J.L. & PUB. POL'Y 227, 227 (1998) (arguing that "the emergence of the new presidentialism, as a dominant theme in administrative and structural constitutional law, is, at least potentially, a profoundly anti-regulatory phenomenon"); Robert V. Percival, Checks Without Balance: Executive Office Oversight of the Environmental Protection Agency, 54 LAW & CONTEMP. PROBS. 127, 204 (Autumn 1991) (advocating reform of regulatory review "from a check without balance into a positive force for improving regulatory policy"); Peter L. Strauss, Presidential Rulemaking, 72 CHI.-KENT L. REV. 965, 968 (1997) (asserting that President Clinton's visible role in rulemaking "insufficiently respects the tension inherent in the Constitution between Congress's power to create the instruments of government and allocate authority among them and the fact of a single chief executive at the head of the agencies thus created, with intended and inevitable political relationships with all").
4. I use the term "unitarian" to reflect the exercise by the Executive Office of the President of either authority over or substantive, policy-based influence on discretionary conduct (for example, in the rulemaking process) of executive branch officials other than those in independent agencies. In the typical case, legislation has designated a specific senior official other than the president as having authority to exercise the discretion that the Executive Office of the President (or the president himself) either controls or seeks to influence on policy-based grounds. For a discussion of the imposition of an interpretation of a statute by the George H. W. Bush administration on the EPA, based on the administration's policy preferences, see Michael Herz, Imposing Unified Executive Branch Statutory Interpretation, 15 CARDOZO L. REV. 219, 224-25 & n.30, 249 (1993).
5. For a thoughtful call to combat the "unfortunate tendency . . . to deal with executive oversight on a purely theoretical basis, without considering the extent to which the model squares with inescapable realities of modern government," see id. at 219.
6. See, e.g., Steven G. Calabresi, Some Normative Arguments for the Unitary Executive, 48 ARK. L. REV. 23, 37 (1995) (discussing the Framers' desire to create a unitary executive in Article II); Steven G. Calabresi & Saikrishna B. Prakash, The President's Power to Execute the Laws, 104 YALE L.J. 541, 570-99 (1994) (analyzing the grant of power to the president in Article II). An eloquent and by-now-classic judicial statement of this view is that of Justice Scalia's dissent in Morrison v. Olson, 487 U.S. 654, 697 (1988) (Scalia, J., dissenting).
7. See Kagan, supra note 2, at 2320, 2326-31 (asserting that "unitarians have failed to establish their claim for plenary control as a matter of constitutional mandate"); Cass R. Sunstein, The Myth of the Unitary Executive, 7 ADMIN. L.J. AM. U. 299, 300-01 (1993) (arguing that the notion of the Framers' having a unitarian vision, intended to "put the President on top of a pyramid, with the administration below him," is an "ahistorical myth"); cf. Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 COLUM. L. REV. 1, 85-106 (1994) (inferring a strong presumption of unitariness using a structuralist method).
8. Exec. Order No. 12,291, 3 C.F.R. 127 (1982).
9. David Corn, OIRA Lives!, NATION, Oct. 30, 1989, at 484.
10. Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. CHI. L. REV. 1, 6 (1995).
11. Exec. Order No. 12,866 4(c), 3 C.F.R. 638, 642 (1994), reprinted in 5 U.S.C. 601 (2000).
12. Id. 2(b), 3 C.F.R. at 640.
13. Id. 2(b), 4, 4(c)(1)(A), 4(c)(5), 5, 5(a), 6(b), 3 C.F.R. at 642-45.
14. Pildes & Sunstein, supra note 10, at 15.
15. Kagan, supra note 2, at 2246-47.
16. Id. at 2248.
17. Id.
18. See Robert Pear, Bush Swiftly Denounces Nursing Home Plan, N.Y. TIMES, Sept. 8, 2001, at A24 (noting that the White House rejected a regulatory initiative emanating from the Department of Health and Human Services' Centers for Medicare and Medicaid Services that would have "refocus[ed] enforcement efforts to reduce the frequency of inspections at nursing homes that had good records of compliance with federal health and safety standards").
19. Memorandum from John D. Graham, Administrator, Office of Information and Regulatory Affairs, Office of Management and Budget, to the President's Management Council, regarding Presidential Review of Agency Rulemaking by OIRA (Sept. 20, 2001), available at (on file with the Duke Law Journal).
20. The delay in scheduling confirmation hearings and, overall, in the confirmation process, resulted in time running out in the legislative session. The nomination was voted out of the Senate's Governmental Affairs Committee and was supported in committee by Senator John Glenn, the committee's chairman. Faced with a potential floor amendment to the reauthorization of the Paperwork Reduction Act, which Senator Glenn was sponsoring and which had originally established OIRA, Senator Glenn withdrew the Act from floor consideration and blocked my nomination from coming to a floor vote.
21. The position previously had been occupied by S. Jay Plager, who resigned. He later was confirmed as a judge on the Court of Appeals for the Federal Circuit. The Bush administration had been criticized by supporters of a strong centralized presidential regulatory review presence for not moving the nomination more quickly. See Paul A. Gigot, Regulation Cop No Longer Walks His Beat, WALL ST. J., Apr. 13, 1990, at A10 (urging President Bush to fill the OIRA position).
22. For a criticism of the delay in the confirmation hearings, alleging that Senator Glenn was holding my nomination hostage to achieve his legislative goals regarding reauthorization of the Paperwork Reduction Act, see Editorial, Return of the Iron Age, WALL ST. J., Oct. 16, 1990, at A26. For a generally accurate discussion of the demise of the confirmation, see Percival, supra note 3, at 154-55 & n.167. Professor Percival's account is mistaken in its assertion that "the Senate Governmental Affairs Committee . . . refused to confirm the Bush nomination of a new OIRA administrator." Id. at 155. The committee approved the nomination (with Senator Glenn's support), see 136 CONG. REC. 36,320 (1990), but Senator Glenn blocked floor consideration when he lost unanimous consent to place the Paperwork Reduction Act before the Senate without risk of amendment, something he was unwilling to do. He then went to the floor of the Senate and announced that he would not allow my nomination to come before the Senate for a vote. Id. at 36,321.
23. See Role of the Office of Management and Budget in Regulation: Hearings Before the Subcomm. on Oversight and Investigations of the House Comm. on Energy and Commerce, 97th Cong. 54, 87-95 (1981) (statement of C. Boyden Gray, Counsel to Vice President Bush and the Presidential Task Force on Regulatory Relief) (reprinting a transcript of the Hall of Flags Regulatory Reform Briefing Before the United States Chamber of Commerce). See generally C. Boyden Gray, Regulation and Federalism, 1 YALE J. ON REG. 93 (1983) (exploring the Reagan administration's handling of regulatory programs).
24. That budget deal, again fatefully, would include a tax rate increase and be, in effect, a repudiation of President Bush's famous "read my lips, no new taxes" promise. At the same time, that budget deal would put in place a system of spending restraints that helped the Clinton administration achieve a balanced budget. Presumably, the budget deal that Richard Darman negotiated was aimed at achieving that objective during a putative second term for President George H. W. Bush.
25. See Peter M. Shane, Political Accountability in a System of Checks and Balances: The Case of Presidential Review of Rulemaking, 48 ARK. L. REV. 161, 167 (1995) ("[C]ontrary to President Bush's deregulatory rhetoric, the Bush Administration entered 1990 and 1991 with a record of significant increases in the number of new regulations being issued.").
26. Exec. Order No. 12,291, 3 C.F.R. 127 (1982).
27. The centralized presidential regulatory review process did not apply to independent regulatory agencies. Id. 1(d), 3 C.F.R. at 128.
28. Id. 2, 3 C.F.R. at 128.
29. Id. 2(b), 3(d), 6(a)(6), 3 C.F.R. at 128, 129, 131.
30. Id. 3(a), (d), 3 C.F.R. at 128, 129.
31. See, e.g., Pildes & Sunstein, supra note 10, at 11-16 (detailing presidents' "struggle[s] to assert more centralized control over the regulatory state").
32. COMM'N ON LAW & THE ECON., AM. BAR ASS'N, FEDERAL REGULATION: ROADS TO REFORM (1979). The ABA report built on a legislative proposal made by Lloyd Cutler a few years before he became counsel to President Carter. See Lloyd N. Cutler & David R. Johnson, Regulation and the Political Process, 84 YALE L.J. 1395, 1414-17 (1975) (describing Cutler's proposal).
33. Paperwork Reduction Act of 1980, Pub. L. No. 96-511, 94 Stat. 2812 (current version at 44 U.S.C. 3501-3520 (1994)). The Act became effective after the election of 1980, which President Carter lost to Ronald Reagan.
34. 44 U.S.C. 3503 (1994).
35. See supra note 23 and accompanying text.
36. Farina, supra note 3, at 235 (describing tensions between Congress and the president).
37. Although the Bush administration had been criticized in Congress and elsewhere for not speedily nominating an OIRA administrator, after the administration nominated me for the position members of Congress held my confirmation hostage to gain leverage in the fight over reauthorization of the Paperwork Reduction Act. See supra note 22.
38. See Transcript of Proceedings, Nomination Hearing Before the Senate Comm. on Governmental Affairs 72, 74-75 (Oct. 23, 1990) (unpublished transcript, on file with the Duke Law Journal) [hereinafter Nomination Hearing] (statement of Sen. Carl Levin, Member, Senate Comm. on Governmental Affairs) (discussing the negotiations on access to information about the status of OIRA regulatory review).
39. Id. at 63-76.
40. 839 F.2d 768 (D.C. Cir. 1988).
41. Id. at 769.
42. Id. at 770.
43. Id.
44. Id. at 773 (quoting 5 U.S.C. 552(b)(5) (1982)).
45. Id. at 770.
46. Id. at 776.
47. See, e.g., Nomination Hearing, supra note 38, at 64-65 (statement of James F. Blumstein, Nominee, Adm'r, Office of Info. and Regulatory Affairs) (arguing that the appropriate issue is private parties' access to governmental information, not access to information at a particular stage of the governmental deliberative process).
48. Wolfe, 839 F.2d at 775.
49. Id.
50. Id. at 776.
51. Id.
52. Id.
53. The nonlegislative compromise would have set voluntary time limits on OIRA review and would have required broader disclosure (after the conclusion of the decisionmaking process) by OIRA of its recommendations to an agency during the process of regulatory review. Percival, supra note 3, at 154.
54. Nomination Hearing, supra note 38, at 70-76 (statements of James F. Blumstein and Sen. Carl Levin).
55. See, e.g., Pildes & Sunstein, supra note 10, at 11-12 & n.32 (describing Roosevelt's agency reforms).
56. Id. at 11.
57. Budget and Accounting Act of 1921, Pub. L. No. 67-13, 42 Stat. 20.
58. See Act of June 30, 1932, 401-408, Pub. L. No. 72-212, 47 Stat. 382, 413-15 (authorizing the president to reorganize the administrative branch subject to a legislative veto). The creation of the Executive Office of the President stemmed from legislative authorization. See PERI E. ARNOLD, MAKING THE MANAGERIAL PRESIDENCY: COMPREHENSIVE REORGANIZA- TION PLANNING, 1905-1980, at 26-54 (1986) (describing Franklin D. Roosevelt's creation of the Executive Office of the President pursuant to authority granted in the 1939 Reorganization Act). The legislative veto was declared unconstitutional in INS v. Chadha, 462 U.S. 919, 959 (1983).
59. See Pildes & Sunstein, supra note 10, at 13 (noting that the Nixon administration administratively developed a system of "Quality of Life" reviews and that "[i]n the 'Quality of Life' review process, agencies were required to submit significant rules to OMB in advance of publication in the Federal Register").
60. Id. at 13-14. For a more thorough review of this historical evolution, see id. at 11-16.
61. Exec. Order No. 12,291 pmbl., 3 C.F.R. 127, 127 (1982).
62. Id. 2, 3 C.F.R. at 128.
63. Id. 2(a), 3 C.F.R. at 128.
64. Id. 2(b), 3 C.F.R. at 128.
65. Id. 2(c), 3 C.F.R. at 128.
66. Id. 2(d), 3 C.F.R. at 128.
67. Id. 3(d)(1)-(4), 3 C.F.R. at 129.
68. Id. 3(e)(1), 3 C.F.R. at 129.
69. Id. 3(f)(1), 3 C.F.R. at 129-30.
70. Id. 3(f)(2), 3 C.F.R. at 130.
71. Id. 6(a)(5), 3 C.F.R. at 131.
72. Id. 6(a)(8), 3 C.F.R. at 131. I was informed during my nomination process that this authority was used rarely, if ever.
73. Exec. Order No. 12,498 pmbl., 3 C.F.R. 323, 323 (1986).
74. Id. 1(a), 3 C.F.R. at 323.
75. Id. (requiring each agency not only to report its plans and objectives but also to provide "information concerning all significant regulatory actions underway or planned").
76. Id. 1(b), 3 C.F.R. at 323.
77. Id. 3(a)(i), 3 C.F.R. at 324.
78. Id. 3(a)(ii), 3 C.F.R. at 324.
79. Id.
80. Id. 1(c), 3 C.F.R. at 323.
81. Id. 1(d), 3 C.F.R. at 323.
82. Id. 1(b), 3 C.F.R. at 323.
83. Id. 1(d), 3 C.F.R. at 323.
84. Exec. Order No. 12,866, 3 C.F.R. 638 (1994), reprinted in 5 U.S.C. 601 (2000).
85. See Kagan, supra note 2, at 2286 (stating that the Clinton administration's Executive Order 12,866 "retained the most important features of President Reagan's oversight system" and "largely followed the Reagan model"). Executive Order 12,866 also "established an annual regulatory planning process similar to that initiated by [the Reagan administration's] Executive Order 12,498." Id.
86. See id. at 2285-86 (noting that Executive Order 12,866 "made clear that cost-benefit analysis, to the extent permitted by the relevant statute, would continue to serve as the basic criterion in assessing regulatory decisions"); see also Pildes & Sunstein, supra note 10, at 6 (emphasizing the cost-benefit calculus used).
87. Kagan, supra note 2, at 2288.
88. Id. Professor Kagan believes that, "[e]ven more important, the Clinton executive order, unlike the Reagan orders, suggested that the President had authority to direct executive department (though not independent agency) heads in the exercise of their delegated rulemaking power." Id. A formal dispute resolution process established the president or vice president as "arbiter" of disputes "arising out of OMB review." Id. The premise was that the "simple delegation of rulemaking authority to a specified agency head (the kind of delegation which underlies almost all regulations) would not prevent the President from making a final decision." Id. at 2289. This was precisely the process of dispute resolution contemplated when I was nominated to serve as administrator of OIRA. When Senator Glenn blocked my confirmation from coming to the Senate floor for a vote, id. at 2281, the Bush administration turned the vice president's office (specifically, the Competitiveness Council staffed out of his office) into an enforcement entity in addition to an adjudicatory entity (as originally contemplated) and never nominated another person to serve as OIRA administrator once I withdrew from the process. For a critical (and, factually, slightly erroneous) recounting of the Competitiveness Council's origin and influential role, see Percival, supra note 3, at 155-78; Shane, supra note 25, at 165-73.
89. Kagan, supra note 2, at 2281.
90. Id. at 2281-82. In addition to the retention of the Reagan-Bush centralized presidential regulatory review, the Clinton administration influenced agency policy at both the front end and the back end of the agency regulatory process.

At the front end, the Clinton administration used "formal directives . . . instructing one or more agencies to propose a rule or perform some other administrative action within a set period of time." Id. at 2285. These formal directives "assert[ed] legal authority to direct regulatory decisions," thereby purporting to "command" and not just influence agency decisionmaking by "persuasion." Id. at 2298-99.

At the same time, in what is now understood as a particularly Clintonesque style, the executive order appeared to disclaim what its proponents had claimed for it. The order defused opposition in part by "moderating controversial aspects of the Reagan oversight system." Id. at 2286. Thus, its "rhetoric . . . downplayed the substantive importance of OMB's role in reviewing agency decisions," by expressly listing as an objective "to reaffirm the primacy of Federal agencies in the regulatory decision-making process." Id. (quoting Exec. Order No. 12,866 pmbl., 3 C.F.R. 638, 638 (1994), reprinted in 5 U.S.C. 601 (2000)). Relying on this language, other provisions in the executive order, and "evidence . . . outside the four corners of the document," sympathetic commentators approvingly asserted that Executive Order 12,866 "establishe[d] the agencies as the principal decision makers, and in this way it insisted, more than its predecessors, on agency autonomy." Pildes & Sunstein, supra note 10, at 17. The regulatory oversight and planning process of Executive Order 12,866 "appear[ed] to involve a government-wide system of priority setting, rather than one concentrated in OMB." Id.; cf. Shane, supra note 25, at 164 (touting the Clinton administration executive order's "diffusion of authority" as promising a "significant gain in political accountability"). So, the controversy attending the Reagan-Bush system of regulatory review was allayed in part by soothing (but apparently false) "rhetoric" that purported to retain the "primacy of Federal agencies in the regulatory decision-making process" while, in actuality and "self-consciously," effectuating a "change in the form of presidential involvement in administrative decisionmaking," by "cross[ing] from one side [exerting presidential influence on agency decisionmaking] . . . to the other [exerting presidential command authority over agency decisonmaking]." Kagan, supra note 2, at 2286, 2298-99.

Regardless of whether the assertion of legal command authority was justified, the "repeated assertion of directive authority" by the president undoubtedly had an effect on "the understanding of agency and White House officials alike of their respective roles and powers." Id. at 2299. This effect accords with the recognition by Justice Frankfurter and Justice Jackson, in their influential concurrences in the Steel Seizure Case, that, in the context of separation of powers issues, power accretes to the party that, over time, uses it in a systematic way unchallenged by other separation of powers players. Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 593 (1952) (Frankfurter, J., concurring); id. at 634 (Jackson, J., concurring).

The president's issuance of formally mandatory "directives" to the agencies operated at the front end to "command" agency decisionmaking both substantively in terms of regulatory outcomes and in terms of agency agenda-setting. At the back end, President Clinton publicly appropriated the work product of agencies, "public[ly] assert[ing] ownership of agency action." Kagan, supra note 2, at 2299. President Clinton "himself unveiled . . . quantities of administrative work product," thereby "exert[ing] a substantive pull on administrative decisionmaking." Id. at 2299, 2301. For an account of how this form of back-end public appropriation fed back to influence agency decisionmaking, see id. at 2301-03.

91. Kagan, supra note 2, at 2287.
92. Id. (acknowledging that the lack of conflict between OMB-OIRA and agencies during the Clinton administration may have been in part the result of a "convergence of views about regulatory policy between rulemaking agencies and the OMB of a Democratic President").
93. See Shane, supra note 25, at 174 (observing that the Clinton administration's regulatory review process was "more consultative, more accessible, and more deferential to policy making by individual agencies").
94. Exec. Order No. 12,866 6(b)(2), 3 C.F.R. 638, 646 (1994), reprinted in 5 U.S.C. 601 (2000).
95. Kagan, supra note 2, at 2286. In my confirmation hearings, I endorsed the principle of timetables for OIRA review: "I support [timetables] . . . . [G]overnment by delay is not something that I advocate and not something I will be urging as a policy or pursuing as a policy. I think the hard issues have to be raised." Nomination Hearing, supra note 38, at 44-45 (statement of James F. Blumstein).
96. Exec. Order No. 12,866 4(d), 3 C.F.R. at 643; see Pildes & Sunstein, supra note 10, at 17 (noting that the Regulatory Working Group's "purpose . . . is to promote early interaction and cooperation by coordinating agency and OIRA behavior, and also to ensure exchange of information among agencies").
97. See Kagan, supra note 2, at 2286-87; Pildes & Sunstein, supra note 10, at 22-23.
98. Exec. Order No. 12,866 6(b)(3), 3 C.F.R. at 647.
99. Id. 6(b)(4), 3 C.F.R. at 647.
100. See Shane, supra note 25, at 176-77 (citing the executive order's regulatory philosophy contained in Section 1(a)). Actually, on the "hard variables" issue, Executive Order 12,291, issued by President Reagan in 1981, expressly called for the Regulatory Impact Analysis to describe and take into account both potential benefits and costs of a proposed regulatory action that included "effects that cannot be quantified in monetary terms." Exec. Order No. 12,291 3(d)(1), 3 C.F.R. 127, 129 (1982).
101. Shane, supra note 25, at 174.
102. Farina, supra note 3, at 227.
103. Morrison, supra note 1, at 1065.
104. Id. at 1074.
105. See generally Pildes & Sunstein, supra note 10 (supporting a limited concept of regulatory review); Peter L. Strauss & Cass R. Sunstein, The Role of the President and OMB in Informal Rulemaking, 38 ADMIN. L. REV. 181 (1986) (same).
106. Kagan, supra note 2, at 2247-48.
107. See id. at 2281 n.144 ("Among legal academics, Peter Strauss alone, to my knowledge, has noted Clinton's increased involvement in regulatory action.").
108. Id. at 2248.
109. Id.
110. Pear, supra note 18, at A24.
111. Id.
112. Id.
113. Id.
114. Id.
115. Id.
116. Id.
117. Id.
118. Robert Pear, the author of the New York Times article, was aware of the issue of centralized presidential regulatory review. See, e.g., Robert Pear, In Move to Spur Economy, Bush Is Urged to Order 90-Day Ban on New Federal Rules, N.Y. TIMES, Jan. 21, 1992, at A18 (noting that Senator John Glenn and Representative John Conyers "have charged that the White House improperly interfered in the rule-making of Federal Agencies" in the context of reporting on substantive criticism of a proposed moratorium on the issuance of new regulatory rules). If one's principal concern about centralized presidential regulatory control over agency rulemaking were the openness of the process, one could conclude that the Bush administration's direct and overt intervention negated structural criticism. But query how transparent the process is, from the agency's perspective, when the responsible agency official is unaware of the centralized policy decision that the White House made.
119. Kagan, supra note 2, at 2247.
120. See supra note 7 and accompanying text.
121. Kagan, supra note 2, at 2289-90.
122. Id. at 2290.
123. Id. at 2293 & nn.191-92.
124. See supra note 4. This role for centralized presidential regulatory review was controversial and the source of biting criticism. See, e.g., Herz, supra note 4, at 225-26 (noting that critics complained about EOP displacement of agency discretion in statutory interpretation); id. at 249 (noting that the administration's priorities prevailed over the agency's priorities in a case study under discussion "as the model of the unitary executive says it should [have]").
125. See, e.g., the case for greater presidential control of regulatory activity as presented in Pildes & Sunstein, supra note 10, at 43-46, and Strauss & Sunstein, supra note 105, at 189-90.
126. Kagan, supra note 2, at 2327.
127. Id.
128. See, e.g., Morrison v. Olson, 487 U.S. 654, 691-93 & n.31 (1988) (rejecting the hard constitutional version of the unitary executive position by recognizing that Congress can insulate certain officials wielding executive authority, such as members of independent agencies, from at-will termination by the president).
129. Nomination Hearing, supra note 38, at 49 (statement of Sen. Joseph Lieberman, Member, Senate Comm. on Governmental Affairs).
130. Id. at 45-49.
131. Id. at 45 (statement of James F. Blumstein).
132. Id. at 45-46.
133. Id. at 46.
134. Id.
135. Id.
136. Id. (statement of Sen. Joseph Lieberman).
137. Id.
138. Id.
139. Id.
140. Id. (statement of James F. Blumstein).
141. Id.
142. Id.
143. Id. at 46-47.
144. Id. at 47 (statement of Sen. Joseph Lieberman).
145. Id. at 47-48 (statement of James F. Blumstein).
146. Id. at 48.
147. Id.
148. Id.
149. Id. (statement of Sen. Joseph Lieberman).
150. Id. (statement of James F. Blumstein).
151. Id. at 49 (statement of Sen. Joseph Lieberman).
152. Id.
153. For a discussion of these issues and the contention that agencies should develop standards to assure adherence to the technocratic model of agency decisionmaking, see generally Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass'ns, 87 CORNELL L. REV. (forthcoming 2002) (manuscript on file with the Duke Law Journal).
154. See, e.g., Morrison, supra note 1, at 1064 ("I do not dispute that OMB can perform some useful functions in the rulemaking process. Its role in coordinating related proceedings between agencies . . . [is] entirely proper."); Percival, supra note 3, at 180 ("Even critics of regulatory review acknowledge that improved coordination and increased accountability are desirable goals.").
155. Strauss & Sunstein, supra note 105, at 189.
156. Exec. Order No. 12,866 4(d), 3 C.F.R. 638, 643 (1994), reprinted in 5 U.S.C. 601 (2000).
157. See Morrison, supra note 1, at 1064 (suggesting that it would be useful for OMB to ensure "that relevant scientific information, cost data, and alternative approaches are shared between agencies").
158. DeMuth & Ginsburg, supra note 1, at 1085.
159. Id.
160. See E. Donald Elliott, TQM-ing OMB: Or Why Regulatory Review Under Executive Order 12,291 Works Poorly and What President Clinton Should Do About It, 57 LAW & CONTEMP. PROBS. 167, 169 (Spring 1994) (noting that "economic analysis by the OMB generally improves, as well as delays, the EPA rules," and that "many (perhaps as much as eighty percent) of the major issues raised by the OMB had not previously received substantial consideration in internal EPA deliberations").
161. Professor Kagan views this uniform, cross-agency approach as part of the regulatory effectiveness of centralized presidential review. See Kagan, supra note 2, at 2340 (noting that, "[a]s agencies proliferated," their various jurisdictions began to overlap and that "[c]entral presidential oversight" could eliminate at least some of the redundancies that were created as a result).
162. Exec. Order No. 12,866 4(d), 3 C.F.R. 638, 643 (1994), reprinted in 5 U.S.C. 601 (2000) (suggesting that interagency comparative risk assessment methods be developed and then evaluated for their "efficacy" and "utility").
163. See Herz, supra note 4, at 255-56 (noting that "[e]xecutive oversight ensures consistency and coherence among agencies").
164. See infra Part III.D.2.
165. Language can shape perceptions of an issue. For example, advocates of a professional model in health care tend to speak in terms of a "system" of health care delivery. "Use of the term 'system' suggests a social services delivery model." James F. Blumstein, The Application of Antitrust Doctrine to the Healthcare Industry: The Interweaving of Empirical and Normative Issues, 31 IND. L. REV. 91, 94 (1998). Proponents of a market-oriented model tend to refer to it as an "'industry' to be policed through antitrust enforcement against anticompetitive conduct as are other economic sectors." Id. at 94-95.
166. See Kenneth Arrow, Uncertainty and the Welfare Economics of Medical Care, 53 AM. ECON. REV. 941, 946 (1963) (observing that "[w]hen there is uncertainty," knowledge tends to become "concentrated in those that can profit most from it").
167. See James F. Blumstein, Health Care Reform and Competing Visions of Medical Care: Antitrust and State Provider Cooperation Legislation, 79 CORNELL L. REV. 1459, 1466 (1994) (observing that under the professional model, medical care decisionmaking is an "exclusively technical-scientific enterprise").
168. James F. Blumstein, Medicine Isn't an Economics-Free Zone, WALL ST. J., June 22, 2001, at A14.
169. Blumstein, supra note 167, at 1465.
170. Id. at 1466.
171. Id.
172. Blumstein, supra note 168, at A14.
173. See, e.g., Muse v. Charter Hosp. of Winston-Salem, Inc., 452 S.E.2d 589, 593-94 (N.C. Ct. App. 1995) (sustaining an award of punitive damages and finding the medical decision in question, which was influenced by the defendant hospital's economic considerations, to be wanton and willful because it interfered with a physician's purely medical judgment), aff'd, 464 S.E.2d 44 (N.C. 1995). Compare Pegram v. Herdrich, 530 U.S. 211, 236-37 (2000) (reversing the court of appeals and embracing a market-oriented paradigm in a health-plan liability case under ERISA), with Pegram v. Herdrich, 154 F.3d 362, 374-80 (7th Cir. 1998) (warning against the evils of allowing economic considerations to play a role in medical decisionmaking).
174. See Blumstein, supra note 168, at A14 (observing that "the debate over the patients' bill of rights is actually a battle between competing visions of medical care").
175. See, e.g., Am. Lung Ass'n v. EPA, 134 F.3d 388, 389 (D.C. Cir. 1998) (noting that the Clean Air Act requires the EPA to decide what "margin of safety will protect the public health" from a "pollutant's adverse effects" and also to "promulgate national standards that limit emissions sufficiently to establish that margin of safety" without considering cost or "technological feasibility"); see also Am. Petroleum Inst. v. Costle, 665 F.2d 1176, 1185 (D.C. Cir. 1981) (holding that "attainability and technological feasibility are not relevant considerations in the promulgation of national ambient air quality standards"); Lead Indus. Ass'n, Inc. v. EPA, 647 F.2d 1130, 1148-50 (D.C. Cir. 1980) (holding that the EPA was not required to consider technological or economic feasibility in setting air quality standards under the Clean Air Act).
176. See THOMAS O. MCGARITY, REINVENTING RATIONALITY: THE ROLE OF REGULATORY ANALYSIS IN THE FEDERAL BUREAUCRACY 14 (1991) (explaining the "clash of rulemaking cultures" between the OMB and rulemaking agencies); Elliott, supra note 160, at 169 (noting that the major economic issues raised by OMB during the regulatory review process often had not been considered by the EPA); id. at 179 (noting that the EPA and the OMB "are institutions that see themselves as committed to different values, different missions, and different conceptions of what constitutes good policy").
177. Cf. Sierra Club v. Costle, 657 F.2d 298, 403 (D.C. Cir. 1981) (noting that "[m]any commentators agree that . . . communications which produce significant new information [that impacts rulemaking] should be noted on a public record").
178. See Percival, supra note 3, at 151 (noting that "OMB initially established policies . . . [that] prohibit[ed] low-level staff from meeting with industry").
179. Id. at 127 (describing "the tension between the rule of law and the politics of regulation" that is reflected in centralized presidential regulatory review).
180. Kagan, supra note 2, at 2352.
181. Id. at 2353; see also, e.g., Bruce Ackerman, The New Separation of Powers, 113 HARV. L. REV. 633, 689 (2000) (noting that unchecked "political intervention will . . . have predictably toxic effects on the rule of law").
182. The substantive policy disagreements, of course, remain.
183. Nomination Hearing, supra note 38, at 55-59 (statement of Sen. Herb Kohl, Member, Senate Comm. on Governmental Affairs). Senator Kohl asked if I would demonstrate long-term commitment if confirmed as the OIRA administrator. Id. at 55-56. Senator Kohl then observed that without such a commitment my likelihood of succeeding in the position would be "significantly lessened." Id. at 58-59.
184. See, e.g., Elliott, supra note 160, at 172-74 (noting that the EPA stated that a particular option could not be considered because it was too late to gather the necessary data).
185. See supra Part III.C.
186. See, e.g., Elliott, supra note 160, at 171-74 (noting that the political appointee may make decisions without "benefit of any input from the OMB").
187. See id. at 169 (stating that "economic analysis by the OMB generally improves . . . the EPA rules").
188. See Strauss & Sunstein, supra note 105, at 187-88 (suggesting that a virtue of centralized presidential regulatory review is "ensuring that regulatory policy is set by agency heads rather than staffs" and "enhancing the agency head's effective control over her staff").
189. For an interesting discussion of this agenda-setting role of EOP, see Kagan, supra note 2, at 2281-84.
190. See Elliott, supra note 160, at 174 (noting that an agency may not have "vetted" a broader administration policy within its internal decisionmaking process).
191. The agency rulemaking process will impose procedural discipline on the exercise of these policy preferences so that they are adequately informed and subjected to criticism. In turn, an agency must defend its policies through a process of rational decisionmaking and reasoned judgment. See New York v. Reilly, 969 F.2d 1147, 1150 (D.C. Cir. 1992) (stating that "an agency's conclusions [will be upheld] if they are supported by 'such relevant evidence as a reasonable mind might accept as adequate to support a conclusion'").
192. See, e.g., Morrison v. Olson, 487 U.S. 654, 660-69 (1988) (discussing the propriety of presidential control of an independent counsel).
193. See, e.g., Wiener v. United States, 357 U.S. 349, 356 (1958) (holding that the president could not remove an appointee from an independent "adjudicatory body like the War Crimes Commission" just because the president wanted a different person to have the job); Humphrey's Ex'r v. United States, 295 U.S. 602, 631-32 (1935) (holding that because the position of commissioner on the Federal Trade Commission was not a "purely executive office[]," the president was not free to replace a commissioner merely at his discretion).
194. See, e.g., Bowsher v. Synar, 478 U.S. 714, 723-27 (1986) (holding that the separation of powers doctrine does not allow Congress to control an official performing executive functions by removal from office by any means other than impeachment).
195. See, e.g., United States v. Nixon, 418 U.S. 683, 696 (1974) (noting that although it was "theoretically possible for [the Nixon administration] to revoke the Special Prosecutor's authority," the administration did not do so).
196. Kagan, supra note 2, at 2290.
197. Id. at 2298.
198. See DeMuth & Ginsburg, supra note 1, at 1083:
Agency heads exercise their statutory authority at the president's pleasure. . . . The tension between an agency head's statutory responsibilities and his accountability to the president . . . is a political question. . . . [T]he interesting general questions presented by White House review of agency rulemaking are not questions of law, but rather those of politics and policy.
199. For examples of this strategy used during the Clinton administration, see Kagan, supra note 2, at 2282-84 (stating that presidential initiatives aimed at reducing youth smoking and allowing funds from unemployment insurance to be used to pay for leave for new parents were characteristic of President Clinton's approach to presidential administration).
200. This has been referred to as changing the "psychology of government." Strauss, supra note 3, at 986; see also Kagan, supra note 2, at 2299 (noting that the president can "turn[] the spotlight on and creat[e] a constituency for the action ordered," thereby overcoming agency "inertia" by "increasing the costs of noncompliance to agency officials").
201. 343 U.S. 579, 610-11 (1952) (Frankfurter, J., concurring) (noting that "a systematic, unbroken, executive practice . . . engaged in by Presidents . . . may be treated as a gloss on 'executive Power'"); id. at 635 (Jackson, J., concurring) (stating that "[w]hile the Constitution diffuses power . . . , it also contemplates that practice will integrate the dispersed powers into a workable agreement").
202. See DeMuth & Ginsburg, supra note 1, at 1082 (noting that the "strictly legal questions raised by the review programs are not very difficult, and none of the legal attacks has been or is likely to be successful"); Kagan, supra note 2, at 2330 (referring, self-deprecatingly, to her "entire discussion" of the interpretation of legislative intent as having "something of a fictive aspect").
203. An analogous point comes from the period following the Supreme Court's decision in INS v. Chadha, 462 U.S. 919 (1983). Even though the Supreme Court held in Chadha that legislative vetoes were unconstitutional, id. at 959, Congress continued to enact legislation that contained them. See, e.g., KATHLEEN M. SULLIVAN & GERALD GUNTHER, CONSTITUTIONAL LAW 360 (14th ed. 2001) ("Despite Chadha, Congress continued to enact legislative veto provisions in a number of statutes. Congress apparently assumed that fear of budgetary retaliation would assure that the executive branch would honor those provisions."); Louis Fisher, Constitutional Interpretation by Members of Congress, 63 N.C. L. REV. 707, 740 & nn. 236-37 (1985) (noting this point and referring to examples compiled in Fisher's research). In essence, Congress viewed the legislative veto as a political tool even though it had no formal legal effect.
204. See, e.g., J. LEIPER FREEMAN, THE POLITICAL PROCESS 31 (rev. ed. 1965) (stating that "the scene of the activities of the major political parties as they attempt to organize and control the government" is the "political setting" defined by "the interactions . . . [among] congressional committees, executive bureaus, and interest groups").
205. For a summary and discussion of this debate, see Kagan, supra note 2, at 2255-60.
206. See id. at 2260 & n.41 (noting the view that "'iron triangles' . . . bind an agency tightly to a single congressional committee and organized interest" and further noting that some scholars "express[] skepticism about the strength and stability of these relationships").
207. See Morrison, supra note 1, at 1068 ("In the practice of administrative law as it has developed over the years, there have been virtually no restrictions on communications with agency officials at the early stages of their consideration of a particular problem."); Pildes & Sunstein, supra note 10, at 20-21 (explaining that the Administrative Procedure Act "contains no restrictions on ex parte communications during notice-and-comment rule making, and it does not require disclosure of such communications").
208. Cf. Elliott, supra note 160, at 177 (noting that "[w]hen the OMB raises fundamental issues about a rule in the final days before promulgation, after two years of agency work [and consultation], agency staffers [and other iron triangle stakeholders] understandably scream 'foul' and complain of 'late hits'").
209. See DOUGLASS CATER, POWER IN WASHINGTON 17-48 (1964) (characterizing iron triangles as striving "to be self-sustaining in control of power [within their] sphere[s]" and as being "resist[ant to] being overridden"); FREEMAN, supra note 204, at 22 (stating that political "sub-units, . . . encouraged by diffused power and functional specialization of political expertise, tend to enjoy a relatively wide range of political autonomy"); see also Morrison, supra note 1, at 1071 (suggesting that OMB be prohibited from "all . . . involvement in the substance of agency rulemakings, except through on-the-record comments that any interested person, either inside or outside government, could submit").
210. For a particularly critical perspective on centralized presidential regulatory review, see generally Erik D. Olson, The Quiet Shift of Power: Office of Management & Budget Supervision of Environmental Protection Agency Rulemaking under Executive Order 12,291, 4 VA. J. NAT. RESOURCES L. 1 (1984).
211. For many critics, the perceived lack of transparency in the process was an exacerbating factor, a criticism that the Clinton administration's executive order sought to remedy. See, e.g., Pildes & Sunstein, supra note 10, at 19-24 (summarizing several changes that the Clinton administration made in order to expose "internal executive branch communications [to greater] public scrutiny").
212. INS v. Chadha, 462 U.S. 919, 946-51 (1983).
213. Metro. Wash. Airports Auth. v. Citizens for the Abatement of Airport Noise, Inc., 501 U.S. 252, 255-58 (1991).
214. Bowsher v. Synar, 478 U.S. 714, 726 (1986).
215. See, e.g., Pildes & Sunstein, supra note 10, at 20-21 (explaining that the Administrative Procedure Act "contains no restrictions on ex parte communications during notice-and-comment rule making, and it does not require disclosure of such communications").
216. Elliott, supra note 160, at 168.
217. Pildes & Sunstein, supra note 10, at 20. Although Professors Pildes and Sunstein seem to oppose banning such contacts, they also note, quite correctly, that "a failure to disclose ex parte communications may breed harmful and unnecessary suspicion" and advocate that "undisclosed communications that had a real effect on the outcome . . . [should] be made part of the record for judicial review." Id. at 20-21.
218. Agencies also can assure greater adherence to the technocratic side of the continuum by developing standards that constrain their own exercise of discretion in the policymaking process. For development of this theme, see generally Bressman, supra note 153.
219. See, e.g., FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 125 (2000) (striking down an FDA rule that restricted the sale and distribution of cigarettes to children on the ground that the rule was outside the scope of the FDA's authority under the Food, Drug, and Cosmetic Act).
220. This does not gainsay that the issue raises matters of substance. For example, the Supreme Court has adjudicated intrabranch disputes even when the president is empowered to obviate such a dispute by the revocation of authority to an officer to whom he previously has delegated power. In United States v. Nixon, 418 U.S. 683 (1974), President Nixon resisted justiciability of his dispute with the Watergate special prosecutor, who sought to subpoena materials from the President. Id. at 692-93. The President moved to quash the subpoena on grounds that the "matter was an intra-branch dispute between a subordinate and superior officer of the Executive Branch and hence not subject to judicial resolution." Id. at 692. The Court rejected the President's argument, noting that the "Attorney General ha[d] delegated the [statutory] authority to represent the United States in these particular [Watergate] matters to a Special Prosecutor with unique authority and tenure." Id. at 694. The autonomous status of the special prosecutor and the limit on the president's authority over him were built into the regulation. Id. at 694 n.8. Although the attorney general was empowered to nullify or modify the regulation that established the special prosecutor, the Court held that the executive branch was bound by the regulation "[s]o long as [it] remain[ed] in force." Id. at 696.
221. See, e.g., Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 57 (1983) (stating that an agency changing its view of "'what is in the public interest must'" have a good reason for doing so).
222. Kagan, supra note 2, at 2270.
223. New York v. Reilly, 969 F.2d 1147, 1149 (D.C. Cir. 1992).
224. Id. at 1153.
225. See supra Part II.A.
226. See supra Part II.B.
227. For a discussion of these directives, see Kagan, supra note 2, at 2290-99.
228. Id.
229. Id. at 2299-2303.
230. For details of the Clinton directives on health care and gun control, see id. at 2303-09.
231. Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 593 (1952) (Frankfurter, J., concurring) (noting that the "systematic, unbroken" use of executive authority over time can alter the nature of presidential power).
232. See Elliott, supra note 160, at 175-81 (noting that management should work to create an environment in which the agency staff would perceive themselves as having a "shared stake in the long term goals of the enterprise").