Emory Law Journal

The Interdependent Relationship Between Internal and External Separation of Powers
Gillian E. Metzger *Professor of Law, Columbia Law School. This essay benefitted from insightful comments and discussion by participants at the Emory Law Journal’s 2009 Randolph W. Thrower Symposium, Executive Power: New Directions for the New Presidency?. Kevin Angle provided excellent research assistance.

It has been the best of times and the worst of times for internal separation of powers. Over the past few years, internal checks on executive power have been a central topic of legal academic debate—rarely have details of public administrative structure received so much attention. To some extent, this sudden popularity reflects growing interest in questions of institutional design. 1 See, e.g., Michael C. Dorf, Legal Indeterminacy and Institutional Design, 78 N.Y.U. L. Rev. 875, 886–88 (2003); Elizabeth Garrett & Adrian Vermeule, Institutional Design of a Thayerian Congress, 50 Duke L.J. 1277, 1279–86 (2001); Jerry L. Mashaw, Accountability and Institutional Design: Some Thoughts on the Grammar of Governance, in Public Accountability: Designs, Dilemmas and Experiences 115, 115–56 (Michael W. Dowdle ed., 2006). Unfortunately, however, another reason for this attention is the prominent erosion and impotence of such internal constraints under the recent administration of President George W. Bush.

Though differing in subject area and form, the instances in which the Bush Administration appeared to evade and perhaps violate internal constraints on administrative decisionmaking can largely be grouped under the heading of politicization of administration. 2David J. Barron, From Takeover to Merger: Reforming Administrative Law in an Age of Agency Politicization, 76 Geo. Wash. L. Rev. 1095, 1096 (2008); Donald P. Moynihan & Alasdair S. Roberts, The End of an Idea? The Bush Administration and the Exhaustion of the Politicized Presidency 4 (La Follette Sch., Working Paper No. 2008-024, 2008), available at http://www.lafollette.wisc.edu/publications/workingpapers. Some involved allegations that agency decisions, such as EPA’s denial of California’s application for a waiver to set automobile emission limits for greenhouse gases or FDA’s refusal to allow the Plan B emergency contraceptive to be sold over the counter, were being determined by White House ideology and politics instead of statutory criteria and professional assessment. 3 See Memorandum from the Comm. on Oversight & Gov’t Reform to Members of the Comm. on Oversight & Gov’t Reform 1–2 (May 19, 2008), available at http://oversight.house.gov/story.asp?ID=1956 (discussing EPA’s denial of a waiver requested by California to permit that state to regulate greenhouse gas emissions and noting that “[t]he record before the Committee suggests that the White House played a pivotal role in the decision to reject the California petition” and also that “[i]nternal EPA documents and transcribed interviews with EPA staff show that the agency career staff all supported granting the California petition”); Nina A. Mendelson, The California Greenhouse Gas Waiver Decision and Agency Interpretation: A Response to Professors Galle and Seidenfeld, 57 Duke L.J. 2157, 2169 (2008) (noting “apparent (though informal) White House involvement” in reviewing California’s waiver application); Gillian E. Metzger, Abortion, Equality, and Administrative Regulation, 56 Emory L.J. 865, 901–02 (2007) [hereinafter Metzger, Administrative Regulation] (“[A]fter-the-fact justifications reinforce the suspicion that the FDA’s decision [to deny over-the-counter status to Plan B emergency contraceptives] was driven more by moral opposition to teenage sex and politics than the public health concerns that constitute the agency’s statutory mandate.”). Others involved charges of misuse of personnel decisions for political purposes, such as claims that political affiliation and ideology were a basis for civil service hiring at the Department of Justice (DOJ). 4 See, e.g., DOJ, Office of the Inspector Gen. & Office of Prof’l Responsibility, An Investigation of Allegations of Politicized Hiring and Other Improper Personnel Actions in the Civil Rights Division 1 (2008) [hereinafter Politicized Hiring], available at http://www.usdoj.gov/oig/special/index.htm; DOJ, Office of the Inspector Gen. & Office of Prof’l Responsibility, An Investigation into the Removal of Nine U.S. Attorneys in 2006 (2008) [hereinafter U.S. Attorneys Removal], available at http://www.usdoj.gov/oig/special/s0809a/index.htm. Still others involved efforts to restrict information dissemination and insert White House appointees into agency rulemaking decisions—allegedly to serve the Bush Administration’s political agenda. 5 See, e.g., Mark Bowen, Censoring Science 15–16, 34, 36, 49–50, 56, 124, 227 (2008) (describing how NASA scientists were required to pre-clear media appearances); Holly Doremus, Scientific and Political Integrity in Environmental Policy, 86 Tex. L. Rev. 1601, 1603–17 (2008) (describing complaints lodged against the Bush Administration for editing scientific evaluations and censoring agency scientists); Michele Estrin Gilman, The President as Scientist-in-Chief, 45 Willamette L. Rev. 565, 566 (2009) (noting allegations by government scientists of political litmus tests, censorship, and political interference by the Bush Administration); Michael Specter, The Bush Administration’s War on the Laboratory, New Yorker, Mar.13, 2006, at 58, 62 (reporting that Administration officials “repeatedly altered government climate reports in order to minimize the relationship between such [greenhouse gas] emissions and global warming”); see also Exec. Order No. 13,422, 72 Fed. Reg. 2763, 2764 (Jan. 18, 2007) (adding a requirement that approval by agency regulatory policy officers ordinarily be required for rulemaking to commence and that such officers be presidential appointees chosen in consultation with the Office of Management and Budget (OMB)). Yet another category involved efforts to evade or silence dissenting internal voices, a well-documented phenomenon with respect to development of national security policy. 6 Jack Goldsmith, The Terror Presidency: Law and Judgment Inside the Bush Administration 166–68 (2007); id. at 167 (explaining that the Bush White House “made it a practice to limit readership of controversial legal opinions” and that “under directions from the White House, OLC [Office of Legal Counsel] did not show the opinion [the 2002 torture memo] to the State Department, which would have strenuously objected”—a practice Goldsmith “came to believe . . . was done to control outcomes in opinions and minimize resistance to them” (referencing Memorandum from Jay S. Bybee, Assistant Att’y Gen., Dep’t of Justice Office of Legal Counsel to Alberto R. Gonzales, Counsel to the President (Aug. 1, 2002))).

A possible lesson to draw from these incidents is that internal constraints ultimately are of limited effect in checking aggrandized presidential authority. That conclusion seems unduly pessimistic. Instances also exist in which internal resistance played an important role in constraining the Bush Administration’s efforts to push its policy beyond legal limits. 7 See Dan Eggen & Paul Kane, Gonzales Hospital Episode Detailed, Wash. Post, May 16, 2007, at A1 (describing an incident in which then-Attorney General John Ashcroft rebuffed White House aides attempting to obtain his approval for re-authorization of National Security Agency surveillance program); David Johnston & Scott Shane, Notes Detail Pressure on Ashcroft over Spying, N.Y. Times, Aug. 17, 2007, at A14 (describing efforts of White House aides to pressure then-Attorney General John Ashcroft to approve re-authorization of government surveillance program); see also Julian E. Barnes, Military Fought to Abide by War Rules, L.A. Times, June 30, 2006, at A1 (describing conflict between the Bush Administration and military lawyers over military commissions). Constraints that are ineffective in high-profile policy disputes may have significantly greater potency in less public and politically charged contexts—and in high-profile contexts, even internal checks with limited effect may be preferable to no checks at all. Moreover, presidential insistence on a policy position over internal resistance may not actually be an example of internal constraint failure. Instead, sometimes such insistence may exemplify the kind of constitutionally desirable direct presidential oversight of Executive Branch decisionmaking that fosters political accountability. At a minimum, no clear line separates forceful presidential assertion of regulatory priorities and presidential aggrandizement, as recent discussion of the Obama Administration’s expansion of White House policy staff demonstrates. 8President Obama has, for example, appointed numerous so-called “czars” to coordinate legislative and policy initiatives in the White House. See, e.g., Bruce Ackerman, A Role for Congress to Reclaim, Wash. Post, Mar. 11, 2009, at A15 (arguing that White House czars are likely to overshadow Cabinet officials); Tom Hamburger & Christi Parsons, White House Czars’ Power Stirs Criticism, Chi. Trib., Mar. 5, 2009, at C10 (describing criticism of expanded use of White House officials to coordinate policy matters).

I therefore see benefits from paying greater attention to internal administrative design and in particular to analyzing what types of administrative structures are likely to prove effective and appropriate in different contexts. 9A recent example of political attention to institutional design is the Obama Administration’s proposal to pull responsibility for consumer protection from current federal financial regulators and instead house this function, with expanded powers, in a new single-focus agency to ensure that consumer protection in financial contexts receives adequate attention. See Binyamin Appelbaum & David Cho, Obama Blueprint Deepens Federal Role in Markets, Wash. Post, June 17, 2009, at A1 (describing White House plan to create consumer protection agency). But I believe that attending to internal constraints alone is too narrow a focus because it excludes the crucial relationship between internal and external checks on the Executive Branch. Internal checks can be, and often are, reinforced by a variety of external forces—including not just Congress and the courts, but also state and foreign governments, international bodies, the media, and civil society organizations. Moreover, the reinforcement can also work in reverse, with internal constraints serving to enhance the ability of external forces, in particular Congress and the courts, to exert meaningful checks on the Executive Branch. Greater acknowledgment of this reciprocal relationship holds import both for fully understanding the separation of powers role played by internal constraints and for identifying effective reform strategies.

One aspect of this dynamic meriting additional attention is the link between internal Executive Branch constraints and external legal doctrine. Contemporary separation of powers doctrine makes little effort to reinforce internal constraints on the Executive Branch; instead it largely focuses on whether internal constraints intrude too far on presidential power, to the extent it considers such constraints at all. This stands in contrast to administrative law doctrine, which focuses primarily on behavior internal to the Executive Branch and often seeks to encourage Executive Branch adherence to constraints on agency action. This division of labor is not coincidental; the availability of administrative law restrictions on agencies is one reason why the courts have not sought to link internal and external constraints as a matter of separation of powers analysis. Judicial concerns about unduly intruding into congressional and presidential choices in structuring administration, and about the courts’ limited competency on questions of institutional design, are likely in play as well. Yet greater exploration of how separation of powers doctrine could be used to reinforce internal Executive Branch constraints appears justified, especially given the important separation of powers function that internal constraints can serve.

In this Essay, I first describe internal separation of powers mechanisms and the constitutional role they can play. I next take up the question of whether these constraints are effective checks on Executive Branch overreaching and emphasize the mutually reinforcing relationship between such internal constraints and external checks on the Executive Branch. Finally, I discuss the general failure of current separation of powers doctrine to directly connect internal and external constraints and analyze whether including such a linkage would be appropriate.

I. Internal Separation of Powers Mechanisms and Their Constitutional Implications

The meaning of “internal separation of powers” is not immediately self-evident. The Constitution says rather little on separation of powers, but the provisions that do address the issue focus overwhelmingly on external relations between the branches—whether it be the branches’ division and assignment of distinct powers (as in the Vesting Clauses) or their subsequent intermixing (as in provisions for a presidential veto and senatorial advice and consent). 10 See, e.g., U.S. Const. art. I, § 1; id. art. I, § 7, cl. 2–3; id. art. II, § 1, cl. 1; id. art. II, § 2, cl. 2; id. art. III, § 1. Although a few constitutional requirements directed at operations within each branch do exist, 11 See, e.g., id. art. II, § 2, cl. 1 (providing that the President may require opinions in writing from heads of departments); id. art. I, § 7, cl. 1 (providing that revenue bills must originate in the House); id. art. I, § 3, cl. 2 (providing for three classes of senators). the constitutional pattern for internal branch arrangements is either silence or express grants of discretion. 12For example, Article I gives the House and Senate discretion in determining their rules of procedure, id. art. I, § 5, cl. 2, and provides Congress with discretionary authority to shape the government through the “necessary and proper” clause, id. art. I, § 8, cl. 18. Indeed, to a constitutional formalist intent on maintaining sharp divisions among the branches and policing against efforts by each branch to exceed its proper sphere, 13 See INS v. Chadha, 462 U.S. 919, 951 (1983) (“The Constitution sought to divide the delegated powers of the new Federal Government into three defined categories, Legislative, Executive and Judicial . . . . The hydraulic pressure inherent within each of the separate Branches to exceed the outer limits of its power . . . must be resisted.”). the concept of internal separation of powers may seem a contradiction in terms. As a result, some explication and description of what internal separation of powers measures are, as well as an assessment of their constitutional status, is warranted.

A. Internal Separation of Powers Defined

The very idea of internal separation of powers is premised on a functionalist approach to constitutional interpretation that emphasizes general separation of powers principles rather than their specific manifestations in the constitutional text. 14For an account of the contrasting functionalist and formalist approaches to separation of powers, see generally Peter L. Strauss, Formal and Functional Approaches to Separation-of-Powers Questions—A Foolish Inconsistency?, 72 Cornell L. Rev. 488 (1987) [hereinafter Strauss, Formal and Functional]. These principles, well familiar from the Supreme Court’s separation of powers jurisprudence, include the division of the federal government’s powers “into three defined categories, Legislative, Executive, and Judicial,” 15 Chadha, 462 U.S. at 951; see also Bowsher v. Synar, 478 U.S. 714, 721–22 (1986). as well as the intermixing of the branches through “a carefully crafted system of checked and balanced power . . . .” 16Mistretta v. United States, 488 U.S. 361, 381 (1989); see also Buckley v. Valeo, 424 U.S. 1, 121 (1976); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 635 (1952) (Jackson, J., concurring). Though often invoking these principles in a somewhat conclusory and inconsistent manner, 17 See Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U. Pa. L. Rev. 1513, 1517 (1991) (“[T]he Supreme Court’s treatment of the constitutional separation of powers is an incoherent muddle.”); M. Elizabeth Magill, The Real Separation in Separation of Powers Law, 86 Va. L. Rev. 1127, 1132, 1174–83 (2000) [hereinafter Magill, Real Separation] (explaining that “[t]he set of principles constituting ‘separation of powers’ is far more easily invoked than specified” and noting inherent contradiction in current doctrine); see also Peter L. Strauss, The Place of Agencies in the Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573, 617 (1984) [hereinafter Strauss, Place of Agencies] (contrasting the separation of powers and checks and balances approaches). the Court has identified the ultimate goal of the separation of powers system as protecting liberty and prohibiting tyranny by preventing “the accumulation of excessive authority in a single Branch.” 18 Mistretta, 488 U.S. at 381; see also Youngstown, 343 U.S. at 635 (Jackson, J., concurring) (“[T]he Constitution diffuses power the better to secure liberty . . . .”); cf. The Federalist No. 47, at 298 (James Madison) (Clinton Rossiter ed., 1961) (“The accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self appointed, or elective, may justly be pronounced the very definition of tyranny.”). At the same time, in addition to deterring “arbitrary or tyrannical rule . . . [by] dispersing the federal power among three branches . . . [and] allocating specific powers and responsibilities to a branch fitted to the task, the Framers created a National Government that is both effective and accountable.” 19Loving v. United States, 517 U.S. 748, 757 (1996). For descriptions of accountability as a core separation of powers value, see Steven G. Calabresi, Some Normative Arguments for the Unitary Executive, 48 Ark. L. Rev. 23, 42–45 (1995) (describing accountability as one basis for unitary executive theory) and Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1, 93–94 (1994). The Court’s efforts to secure the sometimes contradictory goals of diffused or checked power on the one hand, and accountability on the other, have focused on preventing “the encroachment or aggrandizement of one branch at the expense of another.” 20 Buckley, 424 U.S. at 122; accord Mistretta, 488 U.S. at 382 (“It is this concern of encroachment and aggrandizement that has animated our separation-of-powers jurisprudence . . . .”).

The defining characteristic of internal separation of powers measures is that they seek to achieve these goals by operating within the confines of a single branch. In contrast, external separation of powers measures operate through interactions among the different branches of government or with other forces external to a particular branch’s operations. Although such internal measures are present in all the branches, 21Congressional analogues are easiest to identify; they include not just the bicameral structure of Congress but also the division of each house into separate and sometimes competing committees, rules limiting the power of leadership to force votes and end debate, independent research arms, the presence of majority and minority committee staff, and so on. Judicial checks also exist, however, such as division of the federal courts into geographic-based circuits, the use of three-judge appellate panels, the possibility of en banc review, and the recognition of jurisdictional limitations. My focus here, following the literature, is on internal checks within the Executive Branch. the focus of internal separation of powers scholarship is overwhelmingly on the Executive Branch, reflecting the view that the greatest threat of aggrandized power today lies in the broad delegations of power to the Executive Branch that characterize the modern administrative and national security state. 22 See, e.g., Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314, 2316 (2006) [hereinafter Katyal, Internal Separation] (“The result [of the War on Terror] is an executive that subsumes much of the tripartite structure of government.”); see also Martin S. Flaherty, The Most Dangerous Branch, 105 Yale L.J. 1725, 1816–20 (1996) (describing accumulation of authority and responsibility in the Executive Branch); Abner S. Greene, Checks and Balances in an Era of Presidential Lawmaking, 61 U. Chi. L. Rev. 123, 125 (1994) (“In the post-New Deal world, however, the framers’ factual assumptions have been displaced. Now, it is the President whose power has expanded and who therefore needs to be checked.”). Moreover, as that view suggests, internal separation of powers is most often equated with measures that check or constrain the Executive Branch, particularly presidential power.

A wide range of administrative structures and other mechanisms could be viewed as serving such an internal Executive Branch checking function. Some appear primarily animated by concerns about individual fairness and have a due process element—in particular, the division of functions within agencies and the separation of adjudication from legislative, investigatory, and enforcement activities. 23 See 5 U.S.C. §§ 554, 556 (2006) (detailing the procedures for adjudications and hearings within an agency); see also Strauss, Place of Agencies, supra note 17, at 622–25 (discussing individual fairness in considering separation of functions); Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869, 890, 896 (2009) (noting due process benefits from separating enforcement and adjudicative powers in the law enforcement context). Many others have a more systemic focus and seek to ensure regularity and the rule of law by depoliticizing governmental administration. One example of the latter prominent in separation of powers literature and case law is the independent agency, the head of which enjoys some independence from the President as a result of a term appointment and the requirement that removal be for cause. 24The extent of such independence is a matter of dispute. See Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. Rev. 459, 477–79, 485–95 (2008) (finding that party polarization increases presidential control over supposedly independent agencies but that it also delays the period before a new President can appoint a majority of the agencies’ commissioners). Other internal personnel measures offer independence even within executive agencies, the prime instance being the civil service and its prohibitions on politically-motivated employment decisions. 25 See 5 U.S.C. §§ 2301–2305 (2006) (setting forth a merit-based system for personnel decisions); Katyal, Internal Separation, supra note 22, at 2331–35 (highlighting the importance of an independent civil service and arguing for stronger protection from politicization). Tenure protections for administrative law judges (ALJs) are another example. See 5 U.S.C. § 7521(a) (2006) (requiring “good cause established and determined by the Merit Systems Protection Board on the record after opportunity for hearing” prior to removal, suspension, or reduction in pay of ALJs). Another important structural feature is the presence of independent agency watchdogs, such as inspectors general, who are protected by structural insulation within agencies and by independent reporting relationships with Congress. 26 See Paul C. Light, Monitoring Government: Inspectors General and the Search for Accountability 23–25, 56 (1993) [hereinafter Light, Monitoring Government]. Division of employees into distinct organizational units or agencies can also serve to limit the role of raw political calculations in setting policy, in part by breeding agency cultures that foster more professional decisionmaking based on expertise. 27This dynamic was evident in the FDA’s review of the application to make Plan B available over-the- counter and is also frequently discussed in regard to the creation of separate national science and health research institutes. See Thomas O. McGarity & Sidney A. Shapiro, Workers at Risk: The Failed Promise of the Occupational Safety and Health Administration 194 (1993) (explaining that “the Senate may have desired to protect NIOSH’s ‘scientific integrity’ by sheltering it from the day-to-day political and interest group pressures to which OSHA is constantly subjected, thereby allowing NIOSH to serve as a check on any propensity in OSHA to reach conclusions inconsistent with scientific knowledge,” but also noting attendant coordination problems); Metzger, Administrative Regulation, supra note 3, at 880 (noting “FDA’s decision to reject the recommendations of both its advisory committees and the directors and staff of the offices reviewing the application was a deviation from its usual practice regarding OTC [over-the-counter] applications” and a signal of internal opposition). Indeed, the structural mechanism of simply dividing staff with similar responsibilities into separate agencies can serve a checking function, as their separate administrative homes may foster different perspectives and lead to different sources of information. 28Some scholars have recently defended redundancy in national intelligence responsibilities on this ground. See Anne Joseph O’Connell, The Architecture of Smart Intelligence: Structuring and Overseeing Agencies in the Post 9-11 World, 94 Cal. L. Rev. 1655, 1689 (2006) (“[T]he most effective [national intelligence] structure probably would have redundant components as well as components that coordinate and centralize certain efforts.”); see also Katyal, Internal Separation, supra note 22, at 2324–28 (advocating bureaucratic overlap in national security). Internal constraints can also take a “soft” form, being rooted more in agency traditions and culture than “hard” structural features. A case in point is the Office of Legal Counsel (OLC) in the Department of Justice, which has at times operated as a check on the President as well as on other agencies, despite being headed by political appointees and lacking structural insulation. 29 See Goldsmith, supra note 6, at 33 (“OLC is, and views itself as, the frontline institution responsible for ensuring that the executive branch charged with executing the law is itself bound by law.”); Dawn E. Johnsen, Faithfully Executing the Laws: Internal Legal Constraints on Executive Power, 54 UCLA L. Rev. 1559, 1577–78 (2007) [hereinafter Johnsen, Internal Constraints] (explaining that “OLC’s legal interpretations typically are considered binding within the executive branch, unless overruled by the attorney general or the President (an exceedingly rare occurrence)” and noting OLC’s “tradition of accurate and principled legal advice”); Cornelia T.L. Pillard, The Unfulfilled Promise of the Constitution in Executive Hands, 103 Mich. L. Rev. 676, 703, 710–17 (2005) [hereinafter Pillard, Constitution] (discussing relative OLC independence and noting that OLC is “characterized by relative disengagement from [its] client entities, [is] staffed largely with career lawyers whose principal credentials are their legal skills, and ha[s] tended to foster within [its] own legal culture[] a distinction between politics and law”).

Separate from agency structure, personnel measures, and culture are those internal constraints that target agencies’ methods of operation and procedures. Most prominent among these might be the Administrative Procedure Act (APA), which imposes procedural requirements that agencies must follow in formal adjudication and in adopting or changing binding regulations—the latter feature receiving attention recently as the Obama Administration sought to undo last-minute Bush Administration rulemaking. 30 See Jack M. Beermann, Combating Midnight Regulation, 103 Nw. U. L. Rev. 352, 360 (2009) (discussing actions by the Obama Administration to reverse last-minute regulation by the Bush Administration). Such efforts also occurred eight years earlier, when the Bush Administration confronted a number of rules the Clinton Administration adopted just before leaving office. See Natural Res. Def. Council v. Abraham, 355 F.3d 179, 189–91 (2d Cir. 2004) (describing a Bush Administration attempt to reverse a Clinton-era rule published January 22, 2001). More important on a day-to-day basis are the agency guidance, policy manuals, and regulations that govern much of the operation of federal programs. 31 See Jerry L. Mashaw, Recovering American Administrative Law: Federalist Foundations, 1787-1801, 115 Yale L.J. 1256, 1261–62 (2006) (discussing “internal law of administration” as consisting of internal instructions issued by higher level officials to control the exercise of discretion by their subordinates). Although the APA regulates how agencies act, it is as much an external check as an internal one; not only do its procedural demands focus primarily on ensuring an opportunity for the public to participate in agency decisionmaking, the APA exempts many internal matters from its orbit. 32 See 5 U.S.C. § 553(b)–(c) (2006) (describing notice and comment requirements). But see § 553(a) (exempting rules relating to agency management, including personnel matters, from notice and comment requirements); § 553(b)(3)(A) (similarly exempting “rules of agency organization, procedure, or practice”). Publication and procedural requirements that attach to agency guidance lend it external dimensions as well, 33 See Peter L. Strauss, Publication Rules in the Rulemaking Spectrum: Assuring Proper Respect for an Essential Element, 53 Admin. L. Rev. 803, 804–07 (2001) [hereinafter Strauss, Publication Rules] (describing publication and procedural constraints that can apply to agency guidance). and other procedural checks, such as the requirement of advisory committee participation or review, similarly have both an internal and external character. 34These panels often form a core part of an agency’s regulatory approach, but they are composed primarily of outside experts. A good example is the use of scientific advisory panels in drug regulation. See Lars Noah, Scientific “Republicanism”: Expert Peer Review and the Quest for Regulatory Deliberation, 49 Emory L.J. 1033, 1054–57 (2000) (describing FDA use of technical advisory panels); see also 21 U.S.C. § 355(n) (2006) (setting out composition requirements for FDA advisory committees used in drug regulation and describing their functions). See generally 1 James T. O’Reilly, Food and Drug Administration § 13:94 (3d ed. 2007) (describing role of advisory committees in new drug applications). In 2007, Congress added new controls to help prevent conflicts of interest on FDA advisory panels. See Food and Drug Administration Amendments Act of 2007, Pub. L. No. 110-85, 121 Stat. 823 (to be codified at 21 U.S.C. § 379d-1).

By contrast, one central constraint on rulemaking—the requirement of OMB regulatory review 35Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Sept. 30, 1993), amended by Exec. Order No. 13,422, 72 Fed. Reg. 2763 (Jan. 18, 2007). —is admittedly internal but is not often thought of as a separation of powers mechanism because it fosters rather than checks presidential control over agencies. 36 See Peter L. Strauss, Overseer, or “The Decider”? The President in Administrative Law, 75 Geo. Wash. L. Rev. 696, 702 (2007) [hereinafter Strauss, Overseer] (“[Executive Orders 12,866 and 13,422] threaten the control of agency heads over their agencies’ agendas and effect a dramatic increase in presidential control over regulatory outcomes . . . .”). Although OMB enhances presidential control, it might also undermine an administration’s pro-regulatory agenda due to cost-cutting biases. See Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Regulatory State, 106 Colum. L. Rev. 1260, 1267 (2006) (suggesting that regardless of an administration’s political agenda, Executive Order 12,866 “contains within it several structural and institutional biases against regulation”). Moreover, as David Barron has argued, identifiable White House involvement in—and centralization of—policy setting may be less threatening to agency independence than some alternatives, such as increased politicization of agency appointments. See Barron, supra note 2, at 1120–21. See generally Terry M. Moe, The Politicized Presidency, in The New Direction in American Politics 235, 244–45 (John E. Chubb & Paul E. Peterson eds., 1985) (noting presidential incentives to centralize and politicize administration). Without doubt, presidential interventions and assertions of decisionmaking power can undermine expertise and independence in administration, particularly if pushed too far into agency structures and personnel. Thus, for example, one of the more worrying trends during the Bush Administration was the increase in the number of political appointees and their insertion deeper into agency structures. 37 See Barron, supra note 2, at 1128, 1142 (noting a “surge in the number of politically appointed positions created during the first term of President George W. Bush” and describing the effect of politicization on EPA during the Bush Administration); Strauss, Overseer, supra note 36, at 701–02 (describing the expanded role of Regulatory Policy Officers and the requirement of political appointment); David E. Lewis, The Politics of Presidential Appointments: Political Control and Bureaucratic Performance 19–21, 137 (2008) [hereinafter Lewis, Presidential Appointments] (showing that the number of political appointees increased during the Bush Administration, and noting that today, half of all civilian workforce positions are exempt from the merit system compared to only ten percent in 1951); see also Paul C. Light, Thickening Government: Federal Hierarchy and the Diffusion of Accountability 7 (1995) (finding a 430 percent increase in the number of senior executives and presidential appointees from 1960 to 1992). It is also not clear that broad presidential political control is the best method for ensuring politically responsive decisionmaking. Not only can such a presidential role undermine popular input on government policy through Congress, it can also restrict the political accountability of the Executive Branch by limiting transparency and minimizing the effect of participatory and open administrative processes. 38 See Heidi Kitrosser, The Accountable Executive, 93 Minn. L. Rev. 1741, 1743, 1774 (2009) [hereinafter Kitrosser, Accountable] (arguing that a unitary executive approach undermines accountability by increasing the President’s ability to control information and “make or implement policy behind closed doors”); see also Cynthia R. Farina, The Consent of the Governed: Against Simple Rules for a Complex World, 72 Chi.-Kent L. Rev. 987, 992–1007 (1997) (arguing that accountability justifications for strong presidential power are premised on false understandings of popular will); Flaherty, supra note 22, at 1821–26 (similarly arguing that political accountability justifications for broad presidential authority rest on an unduly simple understanding of accountability); Matthew C. Stephenson, Optimal Political Control of the Bureaucracy, 107 Mich. L. Rev. 53, 55 (2008) (“[A] moderate degree of bureaucratic insulation alleviates rather than exacerbates the countermajoritarian problems inherent in bureaucratic policymaking.”).

Yet at the same time, unilateral agency decisionmaking is also problematic from a separation of powers perspective, raising dangers of an unaccountable fourth branch and ineffective government. 39 See Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2263–64, 2331–46 (2001) (emphasizing the need for “direction and energy” as well as administrative constraints and arguing that presidential involvement leads to more accountable, responsive, and effective government); Lessig & Sunstein, supra note 19, at 98 (arguing that in the context of the modern administrative state, immunizing administrators from presidential control undermines separation of powers values by limiting accountability and increasing the risk of faction). Such unilateral decisionmaking is additionally at odds with constitutional provisions mandating some form of presidential oversight of Executive Branch officials and the constitutional decision to adopt a single rather than plural Executive. 40 See Strauss, Formal and Functional, supra note 14, at 495 (“Any workable theory must not only avoid placing excessive power in the President’s hands, but also maintain his claim to a central and unifying governmental role—that is, to a relationship with all agencies that permits the exercise of his characteristic functions.”). Put differently, the line between excessive politicization and appropriate presidential political input is often unclear. Presidential violation of governing statutes is plainly prohibited except in the rare instances when Congress intrudes on the President’s constitutional powers, 41Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 637 (1952) (Jackson, J., concurring) (“When the President takes measures incompatible with the expressed or implied will of Congress, his power is at its lowest ebb, for then he can rely only upon his own constitutional powers minus any constitutional powers of Congress over the matter.”). but statutes often leave broad room for presidential discretion. In those contexts at least, presidential oversight of assertions of authority may best reflect constitutional structure and separation of powers values. 42 See Kagan, supra note 39, at 2251, 2372–80 (arguing that “a statutory delegation to an executive agency official . . . usually should be read as allowing the President to assert directive authority” and proposing that judicial deference is most appropriate when evidence exists of actual presidential involvement); Strauss, Overseer, supra note 36, at 715 (finding a presidential role uncontroversial where “presidential authority readily fit the ‘oversight’ mold and/or may have been explicitly conferred by Congress”). As a result, categorically excluding mechanisms that enhance rather than check presidential oversight reflects an unjustifiably narrow conception of internal separation of powers.

It is also important not to lose sight of a centrally important fact: Presidents frequently support imposition of internal mechanisms that substantially constrain the Executive Branch and even sometimes adopt such measures voluntarily—on their own or at the initiative of an agency. 43For a recent analysis of the phenomenon of Executive Branch voluntary impositions, see Elizabeth Magill, Foreword: Agency Self-Regulation, 77 Geo. Wash. L. Rev. 859 (2009) [hereinafter Magill, Agency Self-Regulation]. Politics is a partial explanation for this, but another causal factor is that Presidents are judged on their ability to govern effectively. 44 See Lewis, Presidential Appointments, supra note 37, at 1, 55 (noting the abysmal federal response to Hurricane Katrina, specifically the role of political appointees, and arguing that “[s]ince voters and history judge presidents for the performance of the entire federal government during their tenure, this creates incentives for presidents to ensure that policy outcomes, both legislative and administrative, are under their control”); see also Moe, supra note 36, at 239 (explaining that Presidents are motivated by “political support and opposition, political strategy, and political tradeoffs,” and therefore value “‘responsive competence,’ not neutral competence”); Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. Econ. & Org. 81, 95 (1985) (asserting that Presidents, unlike legislators, are judged based on the effect of general government policies). Terry Moe has argued that such presidential performance accountability leads to core dynamics of Executive Branch centralization and politicization, as a President wants “an institutional system responsive to his needs as a political leader. He values organizational competence, to be sure, but what he seeks is ‘responsive competence,’ not neutral competence.” 45Moe, supra note 36, at 239, 244–45; see also Barron, supra note 2, at 1102 (identifying concepts of centralization and politicization); Kagan, supra note 39, at 2339 (describing presidential incentives). Yet a President’s political accountability may also lead him to support administrative structures that are more independent. As David Barron recently noted, sometimes “[a] system for making regulatory policy that is administrative in orientation may itself serve a given President’s agenda”—a situation Barron contends existed under President Franklin Roosevelt, who sought to “bulk[] up the regulatory state.” 46Barron, supra note 2, at 1111–12. Presidents may also find that responsiveness and competence conflict. In a recent study, David Lewis concluded that programs run by expert professional administrators perform better on the whole than those run by political appointees. 47 Lewis, Presidential Appointments, supra note 37, at 195–97 (using the Bush Administration’s Program Assessment Rating Tool (PART) scores to find lower performance ratings of political appointees and suggesting that “reducing the number of political appointees is one means of improving performance”). Presidents may well be willing to forego politicization or centralization and opt for a form of administration they can less easily control if they believe that doing so will yield more effective performance. Finally, Presidents may also conclude that internal constraints are in fact essential to ensure their ability to control administration by providing a mechanism that can limit on-the-ground discretion of agency officials. 48 See Magill, Agency Self-Regulation, supra note 43, at 884–86 (discussing self-regulatory measures as central to helping top-level decision makers control the authority of their agents); see also Jerry L. Mashaw, Reluctant Nationalists: Federal Administration and Administrative Law in the Republican Era, 1801–1829, 116 Yale L.J. 1636, 1685–86 (2007) (“Administrators . . . fear the centrifugal effects of discretion vested in subordinates. If for no reason other than self-protection . . . they inevitably construct supervisory routines and modes of instruction to bend peripheral discretion toward centralized control.”); Strauss, Publication Rules, supra note 33, at 814 (“From an agency perspective, uniformity of administration nationwide is desirable and the agency may doubt whether its pool of not-so-well-paid inspectors will be able to handle so much discretion.”).

B. The Constitutional Legitimacy of Internal Separation of Powers

The profusion of possible internal separation of powers mechanisms complicates assessments of their constitutional legitimacy. To be sure, many internal checks on presidential power are likely to be decried by unitary executive theorists who argue that under our constitutional scheme the President is granted control over all exercises of executive power. Under the unitary executive view, the President must be able to remove any officer or employee and set all administrative policy, even at the extreme of substituting his or her judgment for that of the agency head in whom a statute vested decisionmaking authority. 49 See Steven G. Calabresi & Saikrishna B. Prakash, The President’s Power to Execute the Laws, 104 Yale L.J. 541, 594–96 (1994) (arguing that the power of “removal, a power to act in [an inferior officer’s] stead, and a power to nullify [an inferior officer’s] acts when the President disapproves” are constitutionally required). For commentary critical of unitary executive claims, see Farina, supra note 38, at 987–89 (rejecting democratic legitimation argument of unitarians); Lessig & Sunstein, supra note 19, at 2–3 (“[T]he view that the framers constitutionalized anything like [the unitary] vision of the executive is just plain myth.”); and Strauss, Overseer, supra note 36, at 702–03 (finding the Constitution is “at best ambivalent on the question” of whether a President may direct agency determination of policy matters). According to unitary executive theorists, then, internal constraints such as independent agencies, the civil service, and assertions of independent agency policy-setting authority actually violate constitutional separation of powers principles. 50 Stephen G. Calabresi & Christopher S. Yoo, The Unitary Executive 420–22 (2008) (direction of subordinates); id. at 422–23 (civil service); id. at. 423–25 (independent agencies). That view is of course subject to substantial debate and, at any rate, has failed to find much support so far from the Supreme Court. 51 See, e.g., Morrison v. Olson, 487 U.S. 654, 685–96 (1988). Free Enterprise Fund v. Public Company Accounting Oversight Board, a case currently before the Court and argued on December 7, 2009, may provide the Court with an opportunity to address the extent to which Executive Branch officials must be subject to direct presidential removal and the constitutionality of for-cause removal restrictions. See 537 F.3d 667 (D.C. Cir. 2008), cert. granted, 129 S. Ct. 2378 (2009).

The more interesting point to note here is that even unitary executivists might not question the constitutionality of some of the measures described above. Voluntarily adopted measures are an obvious example; although the policy benefits and costs of such constraints could be disputed, it is hard to view self-imposed constraints on the Executive Branch as a significant threat to presidential constitutional authority. In addition, few deny that Congress has the power to require that standards be based on scientific criteria and evidence, to divide functions within agencies, and to assign similar responsibilities to multiple agencies, even if the effect of such measures is to bolster internal checks on presidential decisionmaking and control. 52 See, e.g., Saikrishna B. Prakash, Fragmented Features of the Constitution’s Unitary Executive, 45 Willamette L. Rev. 701, 705 (2009) (“The power to create offices is not merely the limited power to create generic offices, leaving the President to determine each office’s functions and duties. Rather, when Congress creates a Secretary of Treasury or a Secretary of the Interior, it may establish the powers and duties of these offices.”); see also Kendall v. United States ex. rel. Stokes, 37 U.S. (12 Pet.) 524 (1838). In like vein, those scholars who have raised concerns about expanding presidential authority generally accept the constitutionality of presidential efforts to oversee agency decisionmaking, provided such efforts do not extend so far as to involve presidential assumption of decisionmaking power that Congress has vested in agency heads. 53Strauss, Overseer, supra note 36, at 715–18; see also Harold H. Bruff, Balance of Forces: Separation of Powers Law in the Administrative State 455–74 (2006).

Whatever the scholars’ views, under the Court’s current doctrine the vast majority of internal separation of powers mechanisms within the Executive Branch are constitutional. 54Even in the context of the military, the Court has insisted on Executive Branch adherence to congressional strictures. See, e.g., Hamdan v. Rumsfeld, 548 U.S. 557, 613–25 (2006) (invalidating use of presidentially-authorized military commission because its procedures violate governing statutes). Most significantly, the Court has repeatedly upheld the constitutionality of restrictions on the President’s power to remove high-level Executive Branch officers. 55 See Morrison, 487 U.S. at 685–96; Wiener v. United States, 357 U.S. 349, 356 (1958); Humphrey’s Ex’r v. United States, 295 U.S. 602, 629–31 (1935). Even if the Court were to revisit that determination, 56If the Court determines that members of the Public Company Accounting Oversight Board are principal officers in the Free Enterprise case that is on the October 2009 term docket, it would not need to address the scope of the presidential removal power. See supra note 51. it is quite unlikely to call into question many other structural protections for intra-Executive Branch independence. In particular, the Court long ago upheld the constitutionality of the civil service and reiterated that view in a decision most favorable to presidential control. 57United States v. Perkins, 116 U.S. 483, 484 (1886); see also Myers v. United States, 272 U.S. 52, 160–63, 173–74 (1926) (acknowledging the constitutionality of removal restrictions on some inferior officers upheld in Perkins). The Court has also repeatedly enforced substantive statutory requirements over contrary presidential priorities, with Massachusetts v. EPA 58549 U.S. 497, 532–34 (2007). and Gonzales v. Oregon 59546 U.S. 243, 258, 263–64 (2006). being just two recent examples. 60 See, e.g., Jody Freeman & Adrian Vermeule, Massachusetts v. EPA: From Politics to Expertise, 2007 Sup. Ct. Rev. 51, 52 (identifying these decisions as instances in which a majority was “worrie[d] about the politicization of administrative expertise”); Kathryn A. Watts, Proposing a Place for Politics in Arbitrary and Capricious Review, 119 Yale L.J. 2, 7, 18–23 (2009) (discussing administrative law’s resistance to political justifications for agency action). Hence, as a practical matter, the constitutionality of statutory provisions mandating science and expert-based decisionmaking or segregating professional and expert employees organizationally is beyond debate.

II. The Interdependence of Internal and External Separation of Powers

A separate question about internal separation of powers mechanisms concerns their effectiveness, particularly as measures aimed at constraining Executive Branch aggrandizement. Are they actually able to constrain excessive presidential assertions of authority and other abuses? Or are they, in the end, little more than “parchment barriers” 61 The Federalist No. 48, at 305 (James Madison) (Clinton Rossiter ed., 1961). that are largely ineffective and, worse, may obscure the extent of accumulated presidential power?

The case in favor of internal mechanisms is in part comparative. Real limitations exist on the ability of traditional external constraints, specifically Congress and the courts, to check the power of the Executive Branch. The fundamental impediments for Congress are internal ones, in particular its need to proceed via the arduous process of bicameralism and presentment and the additional obstacles created by the operation of congressional committees and rules. 62 See William N. Eskridge, Jr., Vetogates, Chevron, Preemption, 83 Notre Dame L. Rev. 1441, 1444–46 (2008) (describing barriers to legislative action). The ordinary burdens of the legislative process are intensified in contexts involving efforts to check presidential authority given the frequent need to overcome a presidential veto. 63 Cf. Johnsen, Internal Constraints, supra note 29, at 1562 (arguing that Congress is an inadequate check on presidential power in part because of the ever-present possibility of presidential veto). Congress does wield important investigatory and oversight powers and has other tools that may give it leverage over the President, such as control over spending or the ability to add contentious measures to must-pass legislation. 64 See, e.g., William G. Howell & Jon C. Pevehouse, While Dangers Gather: Congressional Checks on Presidential War Powers 10–27 (2007) (describing congressional powers in the face of executive intransigence); Douglas Kriner, Can Enhanced Oversight Repair “the Broken Branch”?, 89 B.U. L. Rev. 765, 773–75 (2009) (describing the importance of congressional oversight). But the political reality of party allegiance dominating institutional interests, along with greater ideological cohesion among political parties in Congress, undermines these techniques and makes rigorous congressional constraints on presidential actions unlikely except in the context of divided government. 65Kagan, supra note 39, at 2311–12 (noting that “congressional parties have grown more ideologically coherent and partisan” but also arguing that divided power is the reality of modern government); Katyal, Internal Separation, supra note 22, at 2321 (“When the political branches are controlled by the same party, loyalty, discipline, and self-interest generally preclude interbranch checking.”). See generally Thomas E. Mann & Norman J. Ornstein, The Broken Branch: How Congress Is Failing America and How to Get It Back on Track (2006) (describing a shift in Congress from a highly decentralized, committee-based institution to a much more regimented one in which party increasingly trumps committee, and concluding that strong majority leadership in Congress leads not to a vigorous exertion of congressional authority but to a general passivity in the face of executive power, which results in the weakening of our system of checks and balances). Moreover, even if Congress is willing to actually engage in oversight, its ability to do so may be significantly hampered by the Executive Branch’s non-cooperation or intransigence, often in the form of assertion of executive privilege or failure to inform Congress of contentious activities. 66The Bush Administration repeatedly demonstrated such resistance to congressional oversight, from its refusal to turn over documents relating to Vice President Cheney’s energy task force early in 2001, see Mark J. Rozell & Mitchel A. Sollenberger, Executive Privilege and the Bush Administration, 24 J.L. & Pol. 1, 12–18 (2008), to its invocation of executive privilege in 2007 to shield high-level presidential aides from testifying about the removal of seven U.S. Attorneys, see id. at 37–38, and its 2008 attempt to block access to an EPA draft document finding that greenhouse gases endanger the environment, see Heidi Kitrosser, Accountability and Administrative Structure, 45 Willamette L. Rev. 607, 609 (2009) [hereinafter Kitrosser, Administrative Structure]. The Bush Administration has also been accused of tampering with publicly released data. See, e.g., Freeman & Vermeule, supra note 60, at 55 (“[T]here were suggestions of widespread tampering by the Bush administration with the global warming data reported by numerous federal agencies, including EPA.”). Whether Congress was adequately informed of the Bush Administration’s program for expanded domestic national security wiretapping is a question currently in dispute. See, e.g., Perry Bacon Jr. & Joby Warrick, CIA Chief Rebuts Pelosi’s Charges; Panetta Says Lawmakers Were Told About Use of Interrogation Methods, Wash. Post, May 16, 2009, at A1.

Courts, in turn, face jurisdictional barriers that limit their ability to review Executive Branch actions. 67Johnsen, Internal Constraints, supra note 29, at 1587 (“Courts employ a variety of jurisdictional and prudential limitations that either preclude review—such as standing and the political question doctrine—or that result in only partial review.”); Pillard, Constitution, supra note 29, at 688–91 (describing “acute practical and legal limitations on the courts’ ability and willingness to decide many constitutional issues that confront the executive branch”). Such barriers have recently surfaced in litigation challenging the government’s expansion of domestic wiretapping without complying with the Foreign Intelligence Surveillance Act; the Sixth Circuit held that the plaintiffs’ claims of injury from the program were too speculative to provide a basis for standing to challenge the program. 68ACLU v. Nat’l Sec. Agency, 493 F.3d 644, 656 (6th Cir. 2007), cert. denied, 128 S. Ct. 1334 (2008) (“Because there is no evidence that any plaintiff’s communications have ever been intercepted, and the state secrets privilege prevents discovery of such evidence . . . the anticipated harm is neither imminent nor concrete—it is hypothetical, conjectural, or speculative. Therefore, this harm cannot satisfy the ‘injury in fact’ requirement of standing.” (citation omitted)). Even when actions are justiciable, the courts’ effectiveness as a check can be significantly curtailed by their deference to reasonable Executive Branch policy determinations, particularly in the area of national security. 69Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv. L. Rev. 1095, 1140 (2009) (“[W]here judges perceive an emergency, . . . standards of rationality, statutory clarity, evidence, and reasonableness all become more capacious and forgiving.”); see also Pillard, Constitution, supra note 29, at 692 (“In cases involving foreign policy, national security, military, or immigration judgments, the courts systematically apply doctrines of overt deference that cause them to refrain from full enforcement of constitutional norms, leaving that task to the political branches.” (footnotes omitted)). Courts are also reluctant to intervene to correct general failures in administration or prompt Executive Branch action. 70 See Summers v. Earth Island Inst., 129 S. Ct. 1142, 1151 (2009) (“[D]eprivation of a procedural right without some concrete interest that is affected by the deprivation—a procedural right in vacuo—is insufficient to create Article III standing.”); Heckler v. Chaney, 470 U.S. 821, 831 (1985) (“[The Supreme Court] has recognized on several occasions over many years that an agency’s decision not to prosecute or enforce, whether through civil or criminal process, is a decision generally committed to an agency’s absolute discretion.”). Another major impediment is delay. Courts must wait for cases to come to them, and challenges to presidential action or policy are likely to be appealed, postponing final resolution of the underlying claims. 71 See Pillard, Constitution, supra note 29, at 689 (noting that “even where private parties can get courts to respond to their constitutional harms, they may face interstitial deprivations,” including delay and irreparable injury). This is not to say that deference and delay necessarily undermine judicial checks; the Supreme Court’s rejection of the Bush Administration’s refusal to regulate greenhouse gas emissions in Massachusetts v. EPA 72549 U.S. 497, 527–35 (2007). and recent decisions rebuffing broad presidential assertions of power regarding the Guantanamo Bay detainees 73 E.g., Boumediene v. Bush, 128 S. Ct. 2229 (2008); Hamdan v. Rumsfeld, 548 U.S. 557 (2006). are important testaments to the contrary. Yet even in these contexts, the limits of judicial constraints are evident. For example, although the EPA proposed regulating greenhouse gases under the Clean Air Act in response to the Massachusetts decision, the White House refused to act on the proposal and no formal action toward regulating greenhouse gases was taken in the remaining year and a half of the Bush Administration. 74Kitrosser, Administrative Structure, supra note 38, at 608–09 (describing Bush Administration attempts to stall EPA rulemaking on climate change following Supreme Court ruling). The ongoing, multi-year saga of habeas challenges involving the Guantanamo Bay detention center demonstrates even more vividly that it can be years before judicial review forces a change in Executive Branch behavior. 75 See Boumediene, 128 S. Ct. at 2275 (“In some of these cases six years have elapsed without the judicial oversight that habeas corpus or an adequate substitute demands.”).

Several bases exist for thinking that internal separation of powers mechanisms may have a comparative advantage. First, internal mechanisms operate ex ante, at the time when the Executive Branch is formulating and implementing policy, rather than ex post. As a result, they avoid the delay in application that can hamper both judicial and congressional oversight. 76Matthew D. McCubbins et al., Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 Va. L. Rev. 431, 433 (1989) (“[E]ffective political control of an agency requires ex ante constraints on the agency (that is, a means of restricting the agency’s decisionmaking before it actually makes policy choices), one source of which is manipulation of its structure and process.”); see also Laura A. Dickinson, Outsourcing War and Peace (forthcoming 2010) (emphasizing the internalization of rules and norms as being a more effective check than external controls and describing efforts by the military to encourage such internalization, such as integrating JAG officers into command structures); Kenneth A. Bamberger & Deirdre K. Mulligan, Privacy Decisionmaking in Administrative Agencies, 75 U. Chi. L. Rev. 75, 106 (2008) (noting that ex ante controls may overcome shortcomings of ex post monitoring). Second, internal mechanisms often operate continuously, rather than being limited to issues that generate congressional attention or arise in the form of a justiciable challenge. 77 See Matthew D. McCubbins, Abdication or Delegation? Congress, the Bureaucracy, and the Delegation Dilemma, Regulation, Summer 1999, at 30, 33 (“[Congressional] leaders do not have to spend a lot of time looking for trouble. Waiting for trouble to be brought to their attention assures [Congressional] leaders that the trouble is important to constituents.”); Pillard, Constitution, supra note 29, at 690–91 (describing deficiencies of ex post judicial review in ensuring constitutionality of executive actions). Third, internal mechanisms operate not just at the points at which policy proposals originate and are implemented but also at higher managerial levels, thus addressing policy and administration in both a granular and systemic fashion. In addition, policy recommendations generated through internal checks may face less resistance than those offered externally because the latter frequently arise after executive officials have already decided upon a policy course and are more likely to take an adversarial form. 78 See Gillian E. Metzger, Ordinary Administrative Law as Constitutional Common Law, 110 Colum. L. Rev. (forthcoming Mar. 2010) (manuscript at 34, on file with author) [hereinafter Metzger, Ordinary Administrative Law] (discussing greater effectiveness of internally-generated reforms); Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 U. Pitt. L. Rev. 589, 591 (2002) (suggesting that a “bias towards the maintenance of existing beliefs” makes agencies more receptive to new ideas before “lock-in” of policy preferences); Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 Colum. L. Rev. 458, 522 (2001) (“Judicially developed and imposed systems frequently trigger strong resentment and resistance, and they invite strategic behavior aimed at minimizing the impact of the law.”). Internal mechanisms may also gain credibility with Executive Branch officials to the extent they are perceived as contributing to more fully informed and expertise-based decisionmaking. 79 See supra text accompanying notes 27–28 (explaining how internal checks foster expertise and information generation); see also Mark Seidenfeld, Cognitive Loafing, Social Conformity, and Judicial Review of Agency Rulemaking, 87 Cornell L. Rev. 486, 515–16 (2002) [hereinafter Seidenfeld, Cognitive Loafing] (noting that agencies are more likely to respond to review where the reviewing audience is perceived as legitimate).

At the same time, it is important not to exaggerate the ability of internal separation of powers mechanisms to check presidential authority. Although as noted, Presidents have reasons to adhere to these mechanisms, they also have strong incentives to trump and evade internal checks in order to advance their political agendas and desired policy goals. Particularly in the face of a determined President, the constraining power of internal checks can be quite limited. This lesson, perhaps more than any other, was demonstrated by the Bush Administration. 80 See M. Elizabeth Magill, Can Process Cure Substance? A Response to Neal Katyal’s “Internal Separation of Powers”, 116 Yale L.J. Pocket Part 126, 130 (2006), http://thepocketpart.org/2006/11/2/magill.html [hereinafter Magill, Response] (emphasizing that President Bush pursued post-September 11 policies despite awareness of internal dissent); see also William G. Howell, Political Checks on a Politicized Presidency: A Response to Neal Katyal’s “Internal Separation of Powers”, 116 Yale L.J. Pocket Part 111, 114–15 (2006), http://thepocketpart.org/2006/10/26/howell.html (noting “the essentially political nature of bureaucratic structure” and describing methods the Executive might employ to circumvent internal opposition). Policy decisions were repeatedly made against the recommendations of career professional staff, often with evidence of direct White House intervention. 81 See Doremus, supra note 5, at 1603–09 (describing allegations that Bush Administration political appointees interfered in work of agency scientists); Gilman, supra note 5, at 566 (noting media reports about “scientists who claimed they were censored, forced to alter their conclusions, and prohibited [by the Bush Administration] from issuing reports and attending conferences”); Mendelson, supra note 3, at 2164 (“[T]hough Office of Management and Budget clearance was not required for this decision, EPA officials reportedly consulted with the White House on the decision [to deny California’s waiver for stricter greenhouse gas regulations] anyway.” (footnote omitted)); Metzger, Administrative Regulation, supra note 3, at 880–81 (describing the FDA’s rejection of “recommendations of both its advisory committees and the directors and staff of the offices reviewing the application” for over-the-counter status for Plan B). In other instances, most prominently the promulgation of OLC memos on torture and interrogation techniques, top presidential officials avoided consulting with career staff or involving other agencies with expertise on the issues at stake. 82 See Goldsmith, supra note 6, at 166–68 (describing the Bush Administration practice of “limit[ing] readership of controversial legal opinions [from OLC] to a very small group of lawyers” and specifically excluding the State Department from circulation of drafts of torture memos); Cornelia Pillard, Unitariness and Myopia: The Executive Branch, Legal Process, and Torture, 81 Ind. L.J. 1297, 1303–05 (2006) [hereinafter Pillard, Unitariness] (highlighting the importance of consultation within the Executive Branch and suggesting that key entities were not consulted during the drafting of torture memos); see also Jane Mayer, Annals of the Pentagon: The Memo, New Yorker, Feb. 27, 2006, at 32. Both entrenched practices and legal constraints guaranteeing political independence were violated. 83 See Politicized Hiring, supra note 4, at 64 (“[A Justice Department official] considered political and ideological affiliations in hiring career attorneys and in other personnel actions affecting career attorneys in the Civil Rights Division. In doing so, he violated federal law . . . and [Justice] Department policy . . . .”); U.S. Attorneys Removal, supra note 4, at 356 (“[T]he process used to remove the nine U.S. Attorneys in 2006 was fundamentally flawed.”); Johnsen, Internal Constraints, supra note 29, at 1578 (“Many [former OLC attorneys] were deeply outraged and saddened by what they saw as a dramatic and dangerous deviation from the office’s tradition of accurate and principled legal advice.”); see also U.S. Gov’t Accountability Office, Food and Drug Administration: Decision Process to Deny Initial Application for Over-the-Counter Marketing of the Emergency Contraceptive Drug Plan B Was Unusual 5 (2005) (noting how Plan B’s denial letter deviated from consistent FDA practice). As Elizabeth Magill has argued, even the strongest internal constraints are unlikely to be effective when the President is committed to a policy fundamentally at odds with the suggestions generated by independent actors within agencies. 84 See Magill, Response, supra note 80, at 130 (arguing that the belief that further internal process would impede the President was “mistaken” and that the Bush Administration faced internal dissent but “adopted the course that they did despite those objections”).

Yet efforts to strengthen internal checking mechanisms are not necessarily misdirected. The potential benefits of such mechanisms with respect to separation of powers make these efforts particularly worthwhile given the limitations of external checks. 85Johnsen, supra note 29, at 1562 (emphasizing the role of internal legal constraints in the face of “inherent inadequacies of the courts and Congress as external checks on the President”). Moreover, high-profile political disputes are too narrow a frame against which to assess the effectiveness of internal constraints. Even if internal constraints are unable to check a determined President in situations involving deep political disagreement, such constraints may still prove potent in more run-of-the-mill policy disputes or contexts in which political allegiances are more divided. Nor does this mean that internal constraints are ineffective when it counts. To the contrary, high-profile political disputes are arguably situations in which Presidents should be able to implement their policies of choice in order to ensure democratic accountability of the Executive Branch, assuming these policies accord with governing law. In such contexts, the success and effectiveness of internal constraints may be better understood not as forestalling presidential control of policy but rather as ensuring that contentious policy choices are made by the President and that the President’s role is publicly known. 86 Cf. Katyal, Internal Separation, supra note 22, at 2337–38 (recommending transfer of OLC’s adjudicatory function to an independent official, removable only for cause, but making that official’s decisions “subject to presidential overrule”). Katyal explains that “a presidential overruling of [an independent, internal executive adjudication] could trigger reporting to Congress,” id. at 2339, and could also foster accountability by checking unelected directors. Id. at 2338 (arguing that presidential override is appropriate because “lack of political accountability might dispose [directors] toward adventurism”). The onus for checking excessive presidential assertions of authority then falls to external forces, including Congress, the courts, and public opinion.

Most importantly, focusing on the effectiveness of internal constraints alone ignores the critical interdependent relationship between internal and external separation of powers. Internal mechanisms have more traction when reinforced externally, and external checks may have such a reinforcing effect even if their ability to constrain the President directly is more limited. Recent history demonstrates how external mechanisms can reinforce internal constraints. 87Daniel Carpenter’s fascinating study of the emergence of bureaucratic autonomy during the progressive era, in which he underscores the important role that external networks played, demonstrates that the dependence of internal constraints on external forces is not just a recent phenomenon. Daniel P. Carpenter, The Forging of Bureaucratic Autonomy 26–33 (2001). Congressional hearings on the politicization of DOJ hiring and the politically-motivated firings of U.S. Attorneys forced the resignations of Attorney General Alberto Gonzales and several of his staff members. 88Steven Lee Myers & Philip Shenon, Embattled Attorney General Resigns, N.Y. Times, Aug. 27, 2007, http://www.nytimes.com/2007/08/27/washington/27cnd-gonzales.html; see also David Johnston & Eric Lipton, Ex-aide Disputes Gonzales Stand over Dismissals, N.Y. Times, Mar. 30, 2007, at A1 (noting resignation of Attorney General Gonzales’s Chief of Staff over dismissal of U.S. Attorneys); David Stout & David Johnston, A Top Aide to Gonzales Resigns, Becoming Latest Fallout Casualty, N.Y. Times, Apr. 7, 2007, at A1 (reporting resignation of Monica Goodling who helped “coordinate” dismissals of U.S. Attorneys). Even if politicized hiring continued but was driven further underground, the hearings likely served to reinforce civil service protections by making clear the potential reputational costs of such behavior and emboldening career staff to come forward with examples of abuse of power. 89Indeed, former U.S. Attorney David Iglesias has argued that the scandal had a kind of paradoxical effect: “There was a sense at first that maybe it was going to make it less likely for U.S. attorneys to take more controversial cases, public corruption cases. It’s had the opposite effect. You’ve got U.S. attorneys that are really independent in a way they haven’t been in years.” Emma Schwartz, Looking Back on the Justice Department Scandal: A Conversation with Former U.S. Attorney David Iglesias, USNEWS.com, June 4, 2008, http://www.usnews.com/articles/news/politics/2008/06/04/looking-back-on-the-justice-department-scandal.html; cf. Evan J. Criddle, Fiduciary Foundations of Administrative Law, 54 UCLA L. Rev. 117, 162 (2006) (“[A]dministrators may care a good deal more about the reputational harms that flow from public censure than the legal consequences of an adverse judgment (for which they are not usually personally liable). An administrator’s perceived failure to act with reasonable prudence can have devastating reputational costs, as illustrated in the wake of Hurricane Katrina by the news media’s excoriation of Federal Emergency Management Agency (FEMA) director, Michael Brown.”). Another example of the external reinforcement of internal checks is the effect that two senators had in forcing the FDA to reconsider its denial of over-the-counter status for Plan B by blocking action on President Bush’s nominee for FDA Commissioner. 90Gardiner Harris, F.D.A. Approves Broader Access to Next-Day Pill, N.Y. Times, Aug. 25, 2006, at A1; see also Gardiner Harris, Bush Picks F.D.A. Chief, but Vote Is Unlikely Soon, N.Y. Times, Mar. 16, 2006, at A18. Peter Strauss has noted how the need to obtain senatorial consent for a successor can curb presidential power to fire agency heads, thereby reinforcing agency independence and expertise. Strauss, Overseer, supra note 36, at 735–36. External reinforcement from the courts, in the form of the Supreme Court’s habeas decisions, undermined the more extreme positions taken by high-level Bush Administration lawyers and—along with public outcry over OLC’s “torture memo”—appear to have led to some more moderate positions. 91 Bart Gellman, Angler: The Cheney Vice Presidency 354–57 (2008) (describing pullback from more extreme positions on treatment of detainees following the Supreme Court’s decision in Hamdan); Goldsmith, supra note 6, at 125 (“Whenever the Supreme Court threatened to review one of the administration’s terrorism policies, [Solicitor General] Paul Clement was able to eke out small concessions from the White House [such as] more formal procedural protections for detainees.”). Cornelia Pillard has argued more generally that the authority of OLC and the Solicitor General within the Executive Branch depends largely on their being able to “backstop [their] judgments in judicial doctrine.” Pillard, Constitution, supra note 29, at 685. Media exposure combined with pressure by professional organizations and other groups also abated efforts to “politicize science.” 92 See Gilman, supra note 5, at 605 (“Media reports [on politicization of science] spurred some government agencies to conduct internal investigations that generated new policies to protect agency scientists and promote transparency. Thus, the media clearly enhanced accountability . . . .”); see also Andrew C. Revkin, NASA Office Is Criticized on Climate Reports, N.Y. Times, June 3, 2008, at A16 (reporting that Michael Griffin, NASA agency head, quickly ordered review and policy changes when a pattern of distorting science was made public). Finally, the Court’s decision in Massachusetts v. EPA led the EPA political leadership to change its stance and accede to the agency staff’s view that greenhouse gases should be regulated under the Clean Air Act, although the White House continued to stonewall such a move. 93 See Kitrosser, Administrative Structure, supra note 66, at 608–09 (explaining that the “White House refused to see EPA’s plans and did their best to ensure that others could not see them,” even refusing to open e-mail from EPA containing a draft document recommending pollution controls); see also Felicity Barringer, White House Refused to Open E-Mail on Pollutants, N.Y. Times, June 25, 2008, at A15; Juliet Eilperin, White House Tried to Silence EPA Proposal on Car Emissions, Wash. Post, June 26, 2008, at A2. Instead, EPA sent out notice seeking further comment, delaying the rulemaking until the end of the Bush Administration. Juliet Eilperin & R. Jeffrey Smith, EPA Won’t Act on Emissions This Year, Wash. Post, July 11, 2008, at A1.

Equally important, the relationship between internal and external separation of powers is reciprocal: Internal and external checks reinforce and operate in conjunction with one another. Congress needs information to conduct meaningful oversight of the Executive Branch. 94 See Matthew D. McCubbins & Thomas Schwartz, Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms, in Congress, Structure and Policy 426, 427–30 (Matthew D. McCubbins & Terry Sullivan eds., 1987) (arguing that Congress is likely to depend on outside sources, such as citizens and interest groups, for information directing oversight activities); Matthew D. McCubbins et al., supra note 76, at 434 (noting that the cost of oversight provides an incentive “to set up a system in which someone else (that is, a third party outside of the principal-agent diad) monitors the agent and reports acts of noncompliance”); Patricia M. Wald & Jonathan R. Siegel, The D.C. Circuit and the Struggle for Control of Presidential Information, 90 Geo. L.J. 737, 739 (2002) (explaining that Congress requires information from the Executive for legislative and investigatory tasks because “the Executive is the repository of the country’s most important information for public policy formulation”). Internal agency experts and watchdogs are important sources of that information, whether in the guise of formal reports, studies, and testimony or informal conversations and leaks. 95Katyal, Internal Separation, supra note 22, at 2347 (“Over fifty inspectors general serve today, and they are structurally insulated from control by agency heads and required to report their findings biannually to Congress.”). See generally Light, Monitoring Government, supra note 26 (describing how the inspectors general assist cross-branch supervision by opening windows into the Executive Branch). As Seth Kreimer has argued, internal separation of powers constraints may also play a crucial role in enhancing transparency and accountability to external actors through the Freedom of Information Act (FOIA). Seth F. Kreimer, The Freedom of Information Act and the Ecology of Transparency, 10 U. Pa. J. Const. L. 1011, 1025–32, 1037–45 (2008). To be successful, requests under FOIA usually require prior knowledge, and requesters therefore largely depend on leaks from civil servants or other insiders to identify which claims for information to pursue. Id. Procedural constraints within agencies can serve a similar function, alerting Congress to agency activities. 96 See McCubbins & Schwartz, supra note 94, at 427–30 (describing congressional dependence on “fire alarms”—outside interest groups that highlight areas requiring oversight, allowing Congress to focus its resources on the needs of constituents). Internal mechanisms also reinforce congressional mandates by creating bodies of personnel within the Executive Branch who are committed to enforcing the governing statutory regime that sets out the parameters of their authority and regulatory responsibilities—and on whose expertise the functioning of these regulatory regimes often depends. 97Bruff, supra note 53, at 408 (“By training and inclination, bureaucrats seek legal authority for their actions. Accordingly, they constitute an often unappreciated bulwark to the rule of law in its everyday application to the citizen.”). Courts equally depend on information and evidence compiled by agency personnel to review agency actions, and they have invoked this dependence to justify the requirement that agencies disclose underlying information and offer detailed explanations of their decisions. 98 See, e.g., United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 249 (2d Cir. 1977) (“Adequate review of a determination requires an adequate record, if the review is to be meaningful.”). Moreover, despite courts regularly intoning that “it [is] not the function of the court to probe the mental processes of Secretar[ies] in reaching [their] conclusions,” 99Cobell v. Norton, 334 F.3d 1128, 1142 (D.C. Cir. 2003) (citing Morgan v. United States, 304 U.S. 1, 18 (1938)). judicial review of agency actions often appears to turn on judges’ perceptions of the role politics played in decisionmaking by agency officials. 100Freeman & Vermeule, supra note 60, at 52 (arguing that case law was driven by fear that executive expertise had been subordinated to politics). Evidence that decisions were made over the objections of career staff and agency professionals often triggers more rigorous review. 101Tummino v. Von Eschenbach, 427 F. Supp. 2d 212, 230–34 (E.D.N.Y. 2006) (noting concern that agency action was not approved by scientific advisory committee in finding sufficient bad faith to order discovery into reasons for agency action beyond those evident on the administrative record). Inconsistency in an agency’s opinion, particularly over a recent period, often indicates that politics rather than expertise affected the agency’s decision and can lead to greater scrutiny. See, e.g., Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41–42 (1983) (“Accordingly, an agency changing its course by rescinding a rule is obligated to supply a reasoned analysis for the change beyond that which may be required when an agency does not act in the first instance.”). But see FCC v. Fox Television Stations, Inc., 129 S. Ct. 1800, 1810 (2009) (“We find no basis in the Administrative Procedure Act or in our opinions for a requirement that all agency change be subjected to more searching review.”). A particularly striking suggestion of how internal checks can effect judicial review came in the recent Boumediene litigation. Just a few months after refusing to grant certiorari in order to allow the Combatant Status Review Tribunal process to proceed, the Court reversed course and granted review, apparently influenced by the concerns of military lawyers about how the tribunals were functioning. 102Joseph Landau, Muscular Procedure: Conditional Deference in the Executive Detention Cases, 84 Wash. L. Rev. 661, 694 (2009).

The claim that internal and external separation of powers mechanisms are interdependent is not novel. Recognition of this interdependence figures most prominently in commentary on how to enhance internal mechanisms, which often advocates strengthening congressional and public oversight of the Executive Branch through greater disclosure. 103 See Katyal, Internal Separation, supra note 22, at 2341–42 (explaining how congressional reporting requirements might strengthen internal checks); Pillard, Constitution, supra note 29, at 749–50 (arguing that transparency encourages internal executive constitutionalism, since the public put pressure on the President to comply with constitutional provisions); see also Johnsen, supra note 29, at 1596–97 (“Perhaps most essential to avoiding a culture in which OLC becomes merely an advocate of the administration’s policy preferences is transparency in the specific legal advice that informs executive action, as well as in the general governing processes and standards.”). In some cases, however, disclosure may actually weaken internal checks. See Pillard, Unitariness, supra note 82, at 1302–03 (arguing that routine publication of OLC legal opinions might discourage Executive Branch officials from seeking OLC’s legal advice in the first place). It also surfaces, though less frequently, in general separation of powers scholarship. 104 See, e.g., Bruce Ackerman, The New Separation of Powers, 113 Harv. L. Rev. 633 (2000) (analyzing the impact of the tripartite system of government on federal bureaucracy); M. Elizabeth Magill, Beyond Powers and Branches in Separation of Powers Law, 150 U. Pa. L. Rev. 603, 605–06 (2001) [hereinafter Magill, Beyond Powers] (arguing government is separated not into “three undifferentiated branches,” but rather into a “large and diverse set of government decisionmakers,” and considering the implications of this idea for separation of powers analysis); Strauss, Place of Agencies, supra note 17, at 622–25 (emphasizing the import of separation of function concerns for separation of powers analysis). Yet the connection between internal and external mechanisms is often implicit and is not the focus of sustained analysis. Indeed, this connection is sometimes obscured by discussions of how internal mechanisms can replace external checks on the Executive Branch or analyses assessing how well external checks function in isolation from internal Executive Branch constraints. 105 See Katyal, Internal Separation, supra note 22, at 2316 (“The first-best concept of ‘legislature v. executive’ checks and balances must be updated to contemplate second-best ‘executive v. executive’ divisions.”).

Bringing the interdependence of internal and external separation of powers mechanisms to the forefront facilitates a more realistic assessment of what internal Executive Branch constraints can accomplish. Although such mechanisms can act as important restraints on excessive Executive Branch power, they are not a panacea for the failure of other branches to adequately police the President. It would also be a mistake to conclude that the dependence of internal mechanisms on external checks makes the former irrelevant because this argument ignores the reciprocal and dynamic relationship between the two. Focusing on this internal–external interdependence also reveals important lessons for proposals to strengthen internal mechanisms. It suggests that the reforms most likely to succeed are those that explicitly link external and internal constraints, to the benefit of each. 106This linkage need not be specifically to Congress and the courts; other external forces—state and foreign governments, the media, and civil society organizations such as professional associations or advocacy groups—can also play a reinforcing role. As noted above, professional associations sought to publicize the Bush Administration’s interference with the work of government scientists, see supra text accompanying note 92, and advocacy organizations repeatedly have brought litigation to force disclosure of government policy.

III. Reinforcing Internal Constraints Through Separation of Powers Doctrine

In this final Part, I focus on one potential reform: reinforcing internal separation of powers mechanisms through constitutional separation of powers doctrine. This reform technique has to date received relatively little attention in internal separation of powers scholarship. 107Although varied in their specific subject matter focus, many reform proposals have emphasized the importance of greater disclosure as a mechanism for reinforcing internal constraints. See supra note 103 (describing various proposals to promote Executive Branch transparency). More importantly, as I argue below, it has received scant attention in separation of powers decisions, with reinforcement of internal constraints being relegated instead to the realm of administrative law. This doctrinal divide is analytically perplexing and unfortunate because it hides the important role that internal checks on the Executive Branch can—and do—play in our constitutional system.

A. The Absence of Internal Checks in Separation of Powers Analysis

Internal Executive Branch constraints feature regularly in separation of powers jurisprudence. The Court has repeatedly addressed the constitutionality of efforts to insulate Executive Branch officers from at-will presidential removal, and internal Executive Branch structure also determines an officer’s status for purposes of the Appointments Clause. 108Edmond v. United States, 520 U.S. 651, 658–66 (1997) (finding judges of the Coast Guard Court of Criminal Appeals to be “inferior officers,” and reviewing supervisory checks on their authority); Morrison v. Olson, 487 U.S. 654, 691 (1988); Humphrey’s Ex’r v. United States, 295 U.S. 602, 629 (1935). Notably, however, these decisions generally treat internal constraints as a given and focus their attention on determining if the specific constraints at issue represent constitutional violations because they intrude too far on presidential power. Less common, particularly in recent decisions, is judicial recognition of the potential constitutional benefits from internal controls, either in terms of guarding against aggrandized power from within or reinforcing the ability of the other branches to do so.

Morrison v. Olson, 109487 U.S. 654 (1988). which upheld the constitutionality of the independent counsel statute, is a case in point. The Court in Morrison analyzed whether limitations on the President’s ability to control and oversee an independent counsel, including the specification that the counsel could only be removed for good cause, violated constitutional separation of powers principles by “impermissibly interfer[ing] with the President’s exercise of his constitutionally appointed functions.” 110 Id. at 685. In concluding that no such interference was present, the Court acknowledged the constitutional relevance of internal constraints governing the activities of independent counsels, both for determining the counsels’ Appointments Clause status 111 Id. at 670–77. and for ensuring that such constraints did not impermissibly undermine the constitutional values of accountability and adherence to law. 112 Id. at 692–93, 695–96. Left unmentioned, however, was the possibility that the good-cause provision and other internal constraints on presidential oversight in fact advanced these constitutional values by ensuring that legal violations by high-level Executive Branch officials did not go unpunished.

To be sure, the independent counsel statute’s intrusions on presidential power were more immediately salient and represented the basis on which the constitutional challenge was framed. Time and experience have not been kind to the argument that the independent counsels served separation of powers goals; concerns about lack of accountability, prosecutorial excesses, and politicization led Congress to let the statute expire in 1999. 113Roberto Suro, As Special Counsel Law Expires, Power Will Shift to Reno, Wash. Post, June 30, 1999, at A6 (discussing the lapse of the Independent Counsel Act). It is nonetheless surprising that the Court did not discuss the potential constitutional benefits of such an internal constraint on the Executive Branch, even in the course of upholding the independent counsel’s constitutionality. Humphrey’s Executor v. United States, 114295 U.S. 602 (1935). the Court’s 1935 decision upholding for-cause limitations on the President’s ability to remove Federal Trade Commissioners, offers an instructive comparison. The Humphrey’s Executor Court stated that at-will presidential removal would exert a “coercive influence . . . [on] the independence of the commission” 115Id. at 629–30. and portrayed the for-cause limitation as serving a separation of powers function by guarding against presidential assumption of legislative and adjudicative powers. 116 Id.

Similarly absent in recent separation of powers jurisprudence are efforts to use separation of powers doctrine to encourage the Executive Branch’s adherence to—or adoption of—internal constraints. A striking example of this is found in Whitman v. American Trucking Associations, a 2001 decision rejecting the claim that the Clean Air Act’s delegation of authority to set emission standards for pollutants that were “requisite to protect public health” was an unconstitutional delegation of legislative power. 117531 U.S. 457, 472–76 (2001). The Whitman Court adamantly rejected the relevance of agency-imposed constraints to assessing a delegation challenge, stating “[t]he idea that an agency can cure an unconstitutionally standardless delegation of power by declining to exercise some of that power seems to us internally contradictory.” 118 Id. at 473. As Peter Strauss has noted, the Court also gave no weight to the fact that the process used by EPA in promulgating the air quality standards at issue involved independent checks on agency discretion, specifically participation by outside experts on the statutorily-mandated Clean Air Scientific Advisory Council. Peter L. Strauss, On Capturing the Possible Significance of Institutional Design and Ethos, 61 Admin. L. Rev. 259, 270–71 (2009) [hereinafter Strauss, On Capturing].Whitman’s failure to acknowledge the separation of powers benefits of internal constraints is particularly notable because the transfer of broad standard-setting authority to the Executive Branch was at issue. Such standard-setting is a core legislative activity, even if it is one the Executive Branch could also constitutionally perform. 119 See Whitman, 531 U.S. at 487–90 (Stevens, J., concurring) (arguing that the power at issue was legislative). Unlike Morrison, therefore, the constitutional danger most clearly presented by Whitman was the excessive accumulation of power in the Executive Branch, precisely the type of danger to which internal Executive Branch constraints would appear most constitutionally relevant. Indeed, a number of commentators had identified the importance of internal constraints that restrict agency regulatory discretion in addressing delegation fears and guarding against arbitrary or abusive agency action—as had some earlier delegation doctrine precedent. 120 See, e.g., Yakus v. United States, 321 U.S. 414, 425 (1944) (noting the power of the courts to “ascertain whether the will of Congress has been obeyed”); Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 55–57, 219–20 (1969) (suggesting that agencies should provide standards to limit their discretion); Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass’ns, 87 Cornell L. Rev. 452, 477, 479–82 (2001) (arguing that “the answer cannot be that agencies simply lack any authority whatsoever to adopt narrowing constructions in the delegation situation,” and that such authority should instead be derived from administrative law); Metzger, Ordinary Administrative Law, supra note 78 (manuscript at 7) (contrasting Whitman with prior case law looking to other checks, such as judicial review, to determine constitutionality). The net effect of Whitman, however, was to disable delegation doctrine as a means of encouraging adoption of such constraints and reinforcing adherence to their requirements.

An interesting contrast to Whitman’s insistence on the irrelevance of internal constraints is the Court’s 2008 decision in Boumediene v. Bush, which held that the Military Commissions Act (MCA)’s restrictions on the ability of Guantanamo Bay detainees to challenge their detention through habeas corpus violated the Suspension Clause. 121Boumediene v. Bush, 128 S. Ct. 2229, 2240 (2008). Although involving a habeas challenge, Boumediene underscored the importance of the writ of habeas corpus to the separation of powers, explaining that through the writ the Judiciary retains “a time-tested device . . . to maintain the ‘delicate balance of governance’ that is itself the surest safeguard of liberty.” 122 Id. at 2247 (citation omitted). Most significantly, in holding the MCA unconstitutional, Justice Kennedy’s majority opinion repeatedly referenced procedural deficiencies within the government’s internal administrative proceedings, at times suggesting that use of more robust internal procedural protections could have led to a different result. In particular, the Court emphasized that such alternative procedures can be an adequate substitute for habeas and that in determining adequacy, “[w]hat matters is the sum total of procedural protections afforded to the detainee at all stages, direct and collateral.” 123 Id. at 2269; see also id. at 2268 (“[T]he necessary scope of habeas review in part depends upon the rigor of any earlier proceedings . . . .”).

The effect of Boumediene was thus counter to that of Whitman in that the Court used its constitutional scrutiny to encourage the Executive Branch to adopt more robust internal constraints. Although it seems fair to say this was an intentional move on the Court’s part, at a minimum Boumediene demonstrates that a lack of attention to the separation of powers benefits of internal constraints is not universal across the Court’s decisions. 124 Hamdan v. Rumsfeld, 548 U.S. 557 (2006), is arguably another instance of judicial reinforcement of internal constraints, albeit more tacit. According to Neil Katyal, driving the result in Hamdan was the fact that internal Executive Branch experts on the Geneva Convention and the law of war opposed the President’s position. See Neal Kumar Katyal, Hamdan v. Rumsfeld: The Legal Academy Goes to Practice, 120 Harv. L. Rev. 65, 105, 109–12 (2006) (“The Justices consciously refused to award deference to the presidential determinations at issue because they lacked support from the bureaucracy, and in particular the Judge Advocates General and the State Department.”). Yet as Katyal acknowledges, Hamdan’s strong emphasis on the importance of congressional sanction at a minimum obscures the decision’s concern about the lack of internal expertise underlying the Administration’s position. Id. at 112–13; see also Hamdan, 548 U.S. at 593–95, 593 n.23, 601–03, 613; id. at 636 (Breyer, J., concurring) (explaining that Congress had denied the President the necessary authority “to create military commissions of the kind at issue here”). Instead, the more obvious incentive created by Hamdan was to encourage the Executive Branch to obtain congressional sanction for its policy of detaining and trying enemy combatants in military commissions, which the Bush Administration promptly proceeded to do and which resulted in enactment of the MCA. See Boumediene, 128 S. Ct. at 2242 (“Congress responded [to Hamdan] by passing the MCA . . . .”); Goldsmith, supra note 6, at 137–40 (explaining that following Hamdan, “only Congress could help the administration out of its predicament”). Yet Boumediene’s express linkage of separation of powers doctrine and internal constraints remains a rarity and reflects in part specific features of habeas jurisprudence, which have long required absence of adequate alternatives before a habeas claim can lie. 125 See, e.g., Boumediene, 128 S. Ct. at 2262 (“In light of this holding [that non-citizens detained at Guantanamo are entitled to constitutional habeas protections] the question becomes whether the statute stripping jurisdiction to issue the writ avoids the Suspension Clause mandate because Congress has provided adequate substitute procedures for habeas corpus.”); Swain v. Pressley, 430 U.S. 372, 381 (1977) (“[S]ubstitution of a collateral remedy which is neither inadequate nor ineffective to test the legality of a person’s detention does not constitute a suspension of the writ of habeas corpus.”).

Internal Executive Branch constraints and external judicial review are much more frequently connected in administrative law doctrine. Here the most salient recent example is United States v. Mead Corporation, the Court’s 2000 decision indicating that Chevron deference is predominantly granted to agency statutory interpretations promulgated using procedures that carry the force of law. 126United States v. Mead Corp., 533 U.S. 218, 226–27, 230 (2001). By linking deference to particular procedures, Mead gave agencies an incentive to undertake notice-and-comment rulemaking or formal adjudication, both processes that impose significant constraints on an agency’s policy-setting discretion. 127 See id. at 246 (Scalia, J., dissenting) (“Another practical effect of today’s opinion will be an artificially induced increase in informal rulemaking. Buy stock in the GPO.”); Matthew C. Stephenson, The Price of Public Action: Constitutional Doctrine and the Judicial Manipulation of Legislative Enactment Costs, 118 Yale L.J. 2, 31–32 (2008) (“[I]n the nonconstitutional administrative law context, whether an agency used formal decision-making procedures in promulgating an interpretation of a statute is a central factor courts consider in deciding whether the agency’s interpretation is entitled to deference.”). Administrative law decisions have tied judicial review to internal constraints in other ways, such as justifying extensive agency duties of explanation in rulemaking as necessary to allow meaningful judicial review or expressing concern that judicial review may lead to excessive proceduralization of agency decisionmaking. 128United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 249 (2d Cir. 1977) (“Adequate review of a determination requires an adequate record, if the review is to be meaningful.”); see also Vt. Yankee Nuclear Corp. v. Natural Res. Def. Council, Inc., 435 U.S. 519, 547 (1978) (“Monday morning quarterbacking [of the type engaged in by the lower court] not only encourages but almost compels the agency to conduct all rulemaking proceedings with the full panoply of procedural devices normally associated only with adjudicatory hearings.”). Interestingly, some proposed linkages between judicial review and other constraints have yet to find judicial favor. The Court does not give more deference to Executive Branch actions for which there is evidence of greater presidential oversight and sanction, see Kagan, supra note 39, at 2372 (advocating such deference), or of political involvement more generally, see Peter L. Strauss, Revisiting Overton Park: Political and Judicial Controls over Administrative Actions Affecting the Community, 39 UCLA L. Rev. 1251, 1329 (1992) (“The risks created by accepted judicial participation in the political process should lead judges to pay serious attention to the realities of political controls over administrative action before acting on the assumption that such controls will not prove effective.”). Nor is the involvement of state governments or expert advisory committees generally deemed an acceptable basis for expanded deference. But see Am. Trucking Ass’ns, Inc. v. EPA, 175 F.3d 1027, 1061 (D.C. Cir. 1999) (Tatel, J., dissenting) (arguing that because politically accountable state governments play a primary role in determining how to distribute burden or regulation, “courts have less reason to second-guess the specificity of the congressional delegation”); see also Strauss, On Capturing, supra note 118, at 271 (noting that in rejecting the nondelegation challenge in Whitman the Court gave no weight to the presence of an expert advisory committee as a check on agency decisionmaking). Sometimes the connection is tacit, but it is nonetheless an evident dynamic. The prime example is the higher level of scrutiny that the Court often applies to agency decisions that appear driven by political considerations rather than expertise. 129 See, e.g., Freeman & Vermeule, supra note 60, at 54 (“[J]ust as State Farm held deregulatory decisions reviewable, in order to allow a judicial hard look at a decision that allegedly injected politics into an expert judgment, so too MA v EPA held the denial of a petition requesting regulation to be reviewable, and for similar reasons.”). Katherine Watts has recently suggested that courts take a different approach, deferring to express acknowledgment by agencies of the role politics played in their decisionmaking—when such political calculations are allowable by statute—as a way to limit the extent to which politics undermines expertise in agency decisionmaking. See Watts, supra note 60, at 5 (“[W]hat counts as a ‘valid’ reason under arbitrary-and-capricious review should be expanded to include influences from the President, other executive officials and members of Congress, so long as these political influences are openly and transparently disclosed in the agency’s rulemaking record.”). Another central linkage between administrative law doctrine and internal agency constraints is the Accardi principle, or the rule that courts will force agencies to follow their existing regulations. 130 See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–68 (1954). For discussions of Accardi, see Magill, Agency Self-Regulation, supra note 43, at 873–81, and Thomas W. Merrill, The Accardi Principle, 74 Geo. Wash. L. Rev. 569 (2006).

B. Should Separation of Powers Analysis Be Used to Reinforce Internal Executive Branch Constraints?

Assessing whether the Court should use separation of powers analysis to reinforce internal Executive Branch constraints requires greater understanding of why the Court’s recent decisions largely fail to do so. No doubt, a major reason is the availability of ordinary administrative law to fill this role. 131 See Metzger, Ordinary Administrative Law, supra note 78, at 6–7 (noting contexts in which the Court has relied on ordinary administrative constraints to avoid constitutional violations); see also Bressman, supra note 120, at 479–81 (“[Whitman] shifts the source of authority for that requirement from constitutional law to administrative law . . . .”). Ordinary administrative law provides a mechanism by which the Court can often reinforce internal constraints without expressly linking them to constitutional law. But this descriptive explanation just clarifies the phenomenon at issue. The Court is plainly willing to enforce internal Executive Branch constraints—both those imposed by Congress and, in some contexts, those assumed voluntarily by Executive Branch actors. 132 See supra note 130 and accompanying text (discussing Accardi). Many voluntary constraints, most notably Executive Order 12,866, providing for centralized regulatory review by the Office of Management and Budget (OMB), are expressly made not subject to judicial review. See Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Oct. 4, 1993), amended by Exec. Order No. 13,422, 72 Fed. Reg. 2763 (Jan. 18, 2007). Its reluctance lies instead in acknowledging the constitutional role these constraints can play. 133I have argued elsewhere that the Court is similarly reluctant to openly acknowledge the constitutional basis of many core administrative law doctrines. See Metzger, Ordinary Administrative Law, supra note 120, at 4–5.

I find this reluctance puzzling. The strong judicial inclination to avoid unnecessary constitutional questions does not explain the Court’s reluctance, as it is hardly avoiding constitutional questions in cases like Morrison and Whitman. A better explanation might be that, by emphasizing the separation of powers benefits of internal Executive Branch constraints, the Court would suggest that such constraints are constitutionally required and would risk intruding unduly on congressional prerogatives to fashion the administrative structure of federal government. 134U.S. Const. art. I, § 8, cl. 18. Yet it is surely possible to take internal constraints into account as one factor in a separation of powers analysis without conveying that a particular set of constraints is mandatory. 135 See Boumediene v. Bush, 128 S. Ct. 2229, 2269 (2008) (explaining, in reviewing whether military commissions are an adequate substitute for habeas corpus, “[w]hat matters is the sum total of procedural protections afforded to the detainee at all stages, direct and collateral”); Walters v. Nat’l Ass’n of Radiation Survivors, 473 U.S. 305, 326 (1985) (“The flexibility of our approach in due process cases is intended in part to allow room for other forms of dispute resolution; with respect to the individual interests at stake here, legislatures are to be allowed considerable leeway to formulate such processes without being forced to conform to a rigid constitutional code of procedural necessities.”). The Court has done this in the Appointments Clause context, 136Edmond v. United States, 520 U.S. 651, 662–63 (1997) (stating that “[w]hether one is an ‘inferior’ officer depends on whether he has a superior” and that “‘inferior officers’ are officers whose work is directed and supervised at some level by [principal officers]”); Morrison v. Olson, 487 U.S. 654, 671–72 (1988) (considering a variety of factors in determining that independent counsels are inferior officers). and the general balancing and functionalist character of separation of powers analysis would seem easily able to absorb an additional consideration into the mix. 137 See, e.g., Boumediene, 128 S. Ct. at 2246–47, 2263–74 (finding the writ of habeas corpus essential to the separation of powers, but considering the possibility of an adequate alternative); Clinton v. Jones, 520 U.S. 681, 697–706 (1997) (considering the role of the President and the judiciary and the extent to which participating in court proceedings would burden the President’s official duties); Morrison, 487 U.S. at 685–96 (weighing a range of factors to determine whether the independent counsel law upsets the constitutional balance). Moreover, perhaps some internal Executive Branch constraints are constitutionally required to address the separation of powers concerns raised by the expansion of executive power in the modern administrative and national security state—or, at a minimum, the Court should more directly engage that possibility before rejecting it. 138In this regard, Whitman’s formalistic emphasis on congressional guidance as the sole constitutional consideration in nondelegation challenges and its lack of attention to more functional considerations is particularly unsatisfying, especially given the decision’s functionalist justification for why limitations on delegations are not rigorously enforced. Whitman v. Am. Trucking Ass’ns, 531 U.S. 457, 474–75 (2001) (“[W]e have ‘almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.’” (quoting Mistretta v. United States, 488 U.S. 361, 416 (1989) (Scalia, J., dissenting))).

A more significant concern is institutional competency. Courts may have difficulty assessing internal separation of powers mechanisms in a principled yet meaningful manner. Internal constraints may simultaneously advance some separation of powers values while undermining others. In particular, at the same time as they serve the constitutional goal of checking excessive Executive Branch power, such constraints arguably undermine political accountability and the Executive Branch’s overall unitary structure. 139In her article on presidential administration, for example, now-Solicitor General Elena Kagan famously argued that, to enhance accountability and improve effectiveness, courts should more readily defer to agency determinations when the President is more directly involved. Kagan, supra note 39, at 2372. Such deference, she maintained, would enhance important constitutional values, including accountability, id. at 2331–39, and effectiveness, id. at 2339–46, without undermining the separation of powers, id. at 2319–31 (accepting the argument that Congress may constitutionally limit presidential control of agency activity, but suggesting courts should generally interpret statues as providing for a presidential role). Indeed, to some scholars the role that internal constraints play in strengthening external checks, particularly external checks by Congress, makes them constitutionally suspect because they represent Congress aggrandizing itself at the President’s expense. 140 See FCC v. Fox Television Stations,129 S. Ct. 1800, 1815–16 (2009) (“The independent agencies are sheltered not from politics but from the President, and it has often been observed that their freedom from presidential oversight (and protection) has simply been replaced by increased subservience to congressional direction.”); Calabresi & Prakash, supra note 49, at 583 (“[I]ndirect political control [by Congress] will necessarily exist with any so-called ‘independent’ agency or officer because absent presidential control, congressional oversight and appropriations powers become the only concern for the officers of the allegedly ‘independent’ agencies.”). Separate from this concern about indeterminacy is the problem that the practical effect of such mechanisms is often contested and rests on aspects of government operations unfamiliar to many judges. Assessing the impact of specific mechanisms may therefore prove difficult even when agreement exists on the relative priority of the separation of powers values at stake.

These institutional competency concerns are quite real, but ultimately are not a persuasive reason to forego reinforcing internal constraints through separation of powers analysis. Neither concern is unique to assessing the potential constitutional benefits of internal Executive Branch constraints. The same concern with constitutional indeterminacy underlies criticism of the Court’s willingness to move beyond specific constitutional provisions and base its constitutional determinations on general constitutional values and principles. 141 See John F. Manning, Federalism and the Generality Problem in Constitutional Interpretation, 122 Harv. L. Rev. 2003, 2040–47 (2009) (finding that constitutional decisions derived from background norms are at odds with the understanding of the Constitution as a finely crafted compromise). But this has long been the case with respect to separation of powers analysis, in which general constitutional values and principles play a central role. 142 See Gillian E. Metzger, Response: The Constitutional Legitimacy of Freestanding Federalism, 122 Harv. L. Rev. 98, 103–06 (2009) (“[T]he Court has a longstanding practice of invoking freestanding federalism in resolving disputes about the scope of federal and state powers or immunities.”). Difficulties in assessing practical impact are equally present when courts address other separation of powers issues, such as the intrusiveness of removal restrictions on a President’s authority and ability to perform constitutional functions. Indeed, courts regularly make such assessments in a variety of constitutional contexts. 143 See, e.g., Wilkinson v. Austin, 545 U.S. 209, 224–30 (2005) (finding detention procedures provided detainees to be adequate process); Forsyth County v. Nationalist Movement, 505 U.S. 123, 128–37 (1992) (finding parade ordinance invalid under First and Fourteenth Amendments). Moreover, the implications of this competency concern are not easily cabined to constitutional analysis and also call into question efforts to reinforce internal checks and otherwise encourage agency self-regulating behavior through administrative law. 144Indeed, administrative law scholars have often criticized the Court on exactly this basis, with a prime example being complaints that the Court’s decisions have served to ossify rulemaking. See, e.g., Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process, 41 Duke L.J. 1385, 1387–96, 1400–03, 1419 (1992) (explaining that the vigorous judicial review “of complex rulemaking” has led to ossification). As a result, this competency concern cannot justify the current disparate treatment of internal constraints in constitutional as opposed to administrative law.

In addition, this concern rests on a misperception of the role courts would need to play if separation of powers doctrine were used to reinforce internal constraints. Courts could, as in Boumediene, identify a fairly specific set of internal constraints as necessary to avoid a constitutional violation. But they could also serve a reinforcing role simply by taking such constraints into account in the manner suggested above, as a factor that may support a finding of constitutionality or greater deference. 145This was the approach taken in Wyeth v. Levine last term, in which the Court indicated that it would determine the weight given to agency assessments of the burden that state laws impose on federal regulatory schemes based on the degree to which those assessments represent an “informed determination[].” Wyeth v. Levine, 129 S. Ct. 1187, 1201 (2009) (“The weight we accord the agency’s explanation of state law’s impact on the federal scheme depends on its thoroughness, consistency, and persuasiveness.”). They could even continue to rely primarily on administrative law doctrine, both as a means of reinforcing internal checks and more directly policing against undue agency politicization, while simply acknowledging more openly the constitutional separation of powers function that administrative law is then performing. 146For suggestions along these lines, see Metzger, Ordinary Administrative Law, supra note 78, at 7–12.

Thus, neither concerns with judicial overstepping nor judicial institutional competency justify ignoring the potential separation of powers role that internal Executive Branch constraints can serve. That still leaves the question of what is gained by including this recognition and seeking to reinforce such internal constraints through constitutional separation of powers analysis—rather than leaving this task, as at present, to the realm of ordinary administrative law. One practical issue is that a number of instances involving alleged presidential overreaching do not arise in a form that allows a direct administrative law challenge. 147 See, e.g., 5 U.S.C. § 553(a) (2006) (exempting matters involving military and foreign affairs from the scope of the Administrative Procedure Act (APA)); Franklin v. Massachusetts, 505 U.S. 788, 796 (1992) (holding that the President is not an agency under the APA). But viewed more systematically, administrative law often functions well as a reinforcement mechanism, particularly given the Court’s willingness to manipulate doctrine if necessary to ensure that perceived excesses of presidential politicization do not escape judicial administrative law review. 148 Massachusetts v. EPA is perhaps the clearest recent example of this phenomenon. See Freeman & Vermeule, supra note 60, at 108–09 (suggesting that the Court in Massachusetts may have attempted to “nudge” agencies away from politicization and strong presidential administration by employing more rigorous scrutiny).

Although effective in practice, the Court’s current approach suffers from analytic and normative deficiencies. Relying on ordinary administrative law obscures the legitimate constitutional role that internal Executive Branch constraints can play, potentially leading to an incomplete separation of powers analysis that perceives such constraints in unduly negative terms. It also fuels misconceptions about the status of ordinary administrative law doctrine and undermines transparency in administrative law contexts. Further, the Court’s reliance on ordinary administrative law too rarely acknowledges the constitutional concerns that actually motivate its decisions. 149 See Metzger, Ordinary Administrative Law, supra note 78, at 3 (“What is less often acknowledged . . . is the degree to which constitutional concerns permeate ordinary administrative law . . . .”). Particularly given the constant battle over the proper scope of judicial review in administrative law, fuller recognition of the constitutional role that internal Executive Branch constraints can play—and thus of the potential constitutional benefits of judicial reinforcement of such constraints—is warranted.

Conclusion

The public administration scandals of the Bush Administration and recent regulatory failures have rightly focused scholarly and public attention on questions of institutional design. Often disparaged mechanisms for ensuring Executive Branch accountability, such as the Freedom of Information Act or the civil service, are suddenly being viewed in a more positive light. 150 See Scott Shane, A.C.L.U. Lawyers Mine Documents for Truth About Detainees and Interrogations, N.Y. Times, Aug. 30, 2009, at A4 (describing surprise at the success of the ACLU’s Freedom of Information Act requests with respect to the government’s detention policies and activities); see also Evan Perez & Deborah Solomon, Treasury Retreats on TARP Watchdog, Wall St. J., Sept. 2, 2009, at A3 (describing Republican lawmakers’ push for the independence of a special inspector general for the Troubled Asset Relief Program). Although these design questions are largely approached in policy or functional terms, they also carry constitutional resonance. Highlighting this constitutional dimension offers the possibility not only of reinforcing internal Executive Branch constraints through separation of powers analysis, but also of fostering greater appreciation of how to achieve separation of powers goals in the contemporary world of administrative governance.

Footnotes

*Professor of Law, Columbia Law School. This essay benefitted from insightful comments and discussion by participants at the Emory Law Journal’s 2009 Randolph W. Thrower Symposium, Executive Power: New Directions for the New Presidency?. Kevin Angle provided excellent research assistance.

1 See, e.g., Michael C. Dorf, Legal Indeterminacy and Institutional Design, 78 N.Y.U. L. Rev. 875, 886–88 (2003); Elizabeth Garrett & Adrian Vermeule, Institutional Design of a Thayerian Congress, 50 Duke L.J. 1277, 1279–86 (2001); Jerry L. Mashaw, Accountability and Institutional Design: Some Thoughts on the Grammar of Governance, in Public Accountability: Designs, Dilemmas and Experiences 115, 115–56 (Michael W. Dowdle ed., 2006).

2David J. Barron, From Takeover to Merger: Reforming Administrative Law in an Age of Agency Politicization, 76 Geo. Wash. L. Rev. 1095, 1096 (2008); Donald P. Moynihan & Alasdair S. Roberts, The End of an Idea? The Bush Administration and the Exhaustion of the Politicized Presidency 4 (La Follette Sch., Working Paper No. 2008-024, 2008), available at http://www.lafollette.wisc.edu/publications/workingpapers.

3 See Memorandum from the Comm. on Oversight & Gov’t Reform to Members of the Comm. on Oversight & Gov’t Reform 1–2 (May 19, 2008), available at http://oversight.house.gov/story.asp?ID=1956 (discussing EPA’s denial of a waiver requested by California to permit that state to regulate greenhouse gas emissions and noting that “[t]he record before the Committee suggests that the White House played a pivotal role in the decision to reject the California petition” and also that “[i]nternal EPA documents and transcribed interviews with EPA staff show that the agency career staff all supported granting the California petition”); Nina A. Mendelson, The California Greenhouse Gas Waiver Decision and Agency Interpretation: A Response to Professors Galle and Seidenfeld, 57 Duke L.J. 2157, 2169 (2008) (noting “apparent (though informal) White House involvement” in reviewing California’s waiver application); Gillian E. Metzger, Abortion, Equality, and Administrative Regulation, 56 Emory L.J. 865, 901–02 (2007) [hereinafter Metzger, Administrative Regulation] (“[A]fter-the-fact justifications reinforce the suspicion that the FDA’s decision [to deny over-the-counter status to Plan B emergency contraceptives] was driven more by moral opposition to teenage sex and politics than the public health concerns that constitute the agency’s statutory mandate.”).

4 See, e.g., DOJ, Office of the Inspector Gen. & Office of Prof’l Responsibility, An Investigation of Allegations of Politicized Hiring and Other Improper Personnel Actions in the Civil Rights Division 1 (2008) [hereinafter Politicized Hiring], available at http://www.usdoj.gov/oig/special/index.htm; DOJ, Office of the Inspector Gen. & Office of Prof’l Responsibility, An Investigation into the Removal of Nine U.S. Attorneys in 2006 (2008) [hereinafter U.S. Attorneys Removal], available at http://www.usdoj.gov/oig/special/s0809a/index.htm.

5 See, e.g., Mark Bowen, Censoring Science 15–16, 34, 36, 49–50, 56, 124, 227 (2008) (describing how NASA scientists were required to pre-clear media appearances); Holly Doremus, Scientific and Political Integrity in Environmental Policy, 86 Tex. L. Rev. 1601, 1603–17 (2008) (describing complaints lodged against the Bush Administration for editing scientific evaluations and censoring agency scientists); Michele Estrin Gilman, The President as Scientist-in-Chief, 45 Willamette L. Rev. 565, 566 (2009) (noting allegations by government scientists of political litmus tests, censorship, and political interference by the Bush Administration); Michael Specter, The Bush Administration’s War on the Laboratory, New Yorker, Mar.13, 2006, at 58, 62 (reporting that Administration officials “repeatedly altered government climate reports in order to minimize the relationship between such [greenhouse gas] emissions and global warming”); see also Exec. Order No. 13,422, 72 Fed. Reg. 2763, 2764 (Jan. 18, 2007) (adding a requirement that approval by agency regulatory policy officers ordinarily be required for rulemaking to commence and that such officers be presidential appointees chosen in consultation with the Office of Management and Budget (OMB)).

6 Jack Goldsmith, The Terror Presidency: Law and Judgment Inside the Bush Administration 166–68 (2007); id. at 167 (explaining that the Bush White House “made it a practice to limit readership of controversial legal opinions” and that “under directions from the White House, OLC [Office of Legal Counsel] did not show the opinion [the 2002 torture memo] to the State Department, which would have strenuously objected”—a practice Goldsmith “came to believe . . . was done to control outcomes in opinions and minimize resistance to them” (referencing Memorandum from Jay S. Bybee, Assistant Att’y Gen., Dep’t of Justice Office of Legal Counsel to Alberto R. Gonzales, Counsel to the President (Aug. 1, 2002))).

7 See Dan Eggen & Paul Kane, Gonzales Hospital Episode Detailed, Wash. Post, May 16, 2007, at A1 (describing an incident in which then-Attorney General John Ashcroft rebuffed White House aides attempting to obtain his approval for re-authorization of National Security Agency surveillance program); David Johnston & Scott Shane, Notes Detail Pressure on Ashcroft over Spying, N.Y. Times, Aug. 17, 2007, at A14 (describing efforts of White House aides to pressure then-Attorney General John Ashcroft to approve re-authorization of government surveillance program); see also Julian E. Barnes, Military Fought to Abide by War Rules, L.A. Times, June 30, 2006, at A1 (describing conflict between the Bush Administration and military lawyers over military commissions).

8President Obama has, for example, appointed numerous so-called “czars” to coordinate legislative and policy initiatives in the White House. See, e.g., Bruce Ackerman, A Role for Congress to Reclaim, Wash. Post, Mar. 11, 2009, at A15 (arguing that White House czars are likely to overshadow Cabinet officials); Tom Hamburger & Christi Parsons, White House Czars’ Power Stirs Criticism, Chi. Trib., Mar. 5, 2009, at C10 (describing criticism of expanded use of White House officials to coordinate policy matters).

9A recent example of political attention to institutional design is the Obama Administration’s proposal to pull responsibility for consumer protection from current federal financial regulators and instead house this function, with expanded powers, in a new single-focus agency to ensure that consumer protection in financial contexts receives adequate attention. See Binyamin Appelbaum & David Cho, Obama Blueprint Deepens Federal Role in Markets, Wash. Post, June 17, 2009, at A1 (describing White House plan to create consumer protection agency).

10 See, e.g., U.S. Const. art. I, § 1; id. art. I, § 7, cl. 2–3; id. art. II, § 1, cl. 1; id. art. II, § 2, cl. 2; id. art. III, § 1.

11 See, e.g., id. art. II, § 2, cl. 1 (providing that the President may require opinions in writing from heads of departments); id. art. I, § 7, cl. 1 (providing that revenue bills must originate in the House); id. art. I, § 3, cl. 2 (providing for three classes of senators).

12For example, Article I gives the House and Senate discretion in determining their rules of procedure, id. art. I, § 5, cl. 2, and provides Congress with discretionary authority to shape the government through the “necessary and proper” clause, id. art. I, § 8, cl. 18.

13 See INS v. Chadha, 462 U.S. 919, 951 (1983) (“The Constitution sought to divide the delegated powers of the new Federal Government into three defined categories, Legislative, Executive and Judicial . . . . The hydraulic pressure inherent within each of the separate Branches to exceed the outer limits of its power . . . must be resisted.”).

14For an account of the contrasting functionalist and formalist approaches to separation of powers, see generally Peter L. Strauss, Formal and Functional Approaches to Separation-of-Powers Questions—A Foolish Inconsistency?, 72 Cornell L. Rev. 488 (1987) [hereinafter Strauss, Formal and Functional].

15 Chadha, 462 U.S. at 951; see also Bowsher v. Synar, 478 U.S. 714, 721–22 (1986).

16Mistretta v. United States, 488 U.S. 361, 381 (1989); see also Buckley v. Valeo, 424 U.S. 1, 121 (1976); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 635 (1952) (Jackson, J., concurring).

17 See Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U. Pa. L. Rev. 1513, 1517 (1991) (“[T]he Supreme Court’s treatment of the constitutional separation of powers is an incoherent muddle.”); M. Elizabeth Magill, The Real Separation in Separation of Powers Law, 86 Va. L. Rev. 1127, 1132, 1174–83 (2000) [hereinafter Magill, Real Separation] (explaining that “[t]he set of principles constituting ‘separation of powers’ is far more easily invoked than specified” and noting inherent contradiction in current doctrine); see also Peter L. Strauss, The Place of Agencies in the Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573, 617 (1984) [hereinafter Strauss, Place of Agencies] (contrasting the separation of powers and checks and balances approaches).

18 Mistretta, 488 U.S. at 381; see also Youngstown, 343 U.S. at 635 (Jackson, J., concurring) (“[T]he Constitution diffuses power the better to secure liberty . . . .”); cf. The Federalist No. 47, at 298 (James Madison) (Clinton Rossiter ed., 1961) (“The accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self appointed, or elective, may justly be pronounced the very definition of tyranny.”).

19Loving v. United States, 517 U.S. 748, 757 (1996). For descriptions of accountability as a core separation of powers value, see Steven G. Calabresi, Some Normative Arguments for the Unitary Executive, 48 Ark. L. Rev. 23, 42–45 (1995) (describing accountability as one basis for unitary executive theory) and Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1, 93–94 (1994).

20 Buckley, 424 U.S. at 122; accord Mistretta, 488 U.S. at 382 (“It is this concern of encroachment and aggrandizement that has animated our separation-of-powers jurisprudence . . . .”).

21Congressional analogues are easiest to identify; they include not just the bicameral structure of Congress but also the division of each house into separate and sometimes competing committees, rules limiting the power of leadership to force votes and end debate, independent research arms, the presence of majority and minority committee staff, and so on. Judicial checks also exist, however, such as division of the federal courts into geographic-based circuits, the use of three-judge appellate panels, the possibility of en banc review, and the recognition of jurisdictional limitations. My focus here, following the literature, is on internal checks within the Executive Branch.

22 See, e.g., Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314, 2316 (2006) [hereinafter Katyal, Internal Separation] (“The result [of the War on Terror] is an executive that subsumes much of the tripartite structure of government.”); see also Martin S. Flaherty, The Most Dangerous Branch, 105 Yale L.J. 1725, 1816–20 (1996) (describing accumulation of authority and responsibility in the Executive Branch); Abner S. Greene, Checks and Balances in an Era of Presidential Lawmaking, 61 U. Chi. L. Rev. 123, 125 (1994) (“In the post-New Deal world, however, the framers’ factual assumptions have been displaced. Now, it is the President whose power has expanded and who therefore needs to be checked.”).

23 See 5 U.S.C. §§ 554, 556 (2006) (detailing the procedures for adjudications and hearings within an agency); see also Strauss, Place of Agencies, supra note 17, at 622–25 (discussing individual fairness in considering separation of functions); Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869, 890, 896 (2009) (noting due process benefits from separating enforcement and adjudicative powers in the law enforcement context).

24The extent of such independence is a matter of dispute. See Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. Rev. 459, 477–79, 485–95 (2008) (finding that party polarization increases presidential control over supposedly independent agencies but that it also delays the period before a new President can appoint a majority of the agencies’ commissioners).

25 See 5 U.S.C. §§ 2301–2305 (2006) (setting forth a merit-based system for personnel decisions); Katyal, Internal Separation, supra note 22, at 2331–35 (highlighting the importance of an independent civil service and arguing for stronger protection from politicization). Tenure protections for administrative law judges (ALJs) are another example. See 5 U.S.C. § 7521(a) (2006) (requiring “good cause established and determined by the Merit Systems Protection Board on the record after opportunity for hearing” prior to removal, suspension, or reduction in pay of ALJs).

26 See Paul C. Light, Monitoring Government: Inspectors General and the Search for Accountability 23–25, 56 (1993) [hereinafter Light, Monitoring Government].

27This dynamic was evident in the FDA’s review of the application to make Plan B available over-the- counter and is also frequently discussed in regard to the creation of separate national science and health research institutes. See Thomas O. McGarity & Sidney A. Shapiro, Workers at Risk: The Failed Promise of the Occupational Safety and Health Administration 194 (1993) (explaining that “the Senate may have desired to protect NIOSH’s ‘scientific integrity’ by sheltering it from the day-to-day political and interest group pressures to which OSHA is constantly subjected, thereby allowing NIOSH to serve as a check on any propensity in OSHA to reach conclusions inconsistent with scientific knowledge,” but also noting attendant coordination problems); Metzger, Administrative Regulation, supra note 3, at 880 (noting “FDA’s decision to reject the recommendations of both its advisory committees and the directors and staff of the offices reviewing the application was a deviation from its usual practice regarding OTC [over-the-counter] applications” and a signal of internal opposition).

28Some scholars have recently defended redundancy in national intelligence responsibilities on this ground. See Anne Joseph O’Connell, The Architecture of Smart Intelligence: Structuring and Overseeing Agencies in the Post 9-11 World, 94 Cal. L. Rev. 1655, 1689 (2006) (“[T]he most effective [national intelligence] structure probably would have redundant components as well as components that coordinate and centralize certain efforts.”); see also Katyal, Internal Separation, supra note 22, at 2324–28 (advocating bureaucratic overlap in national security).

29 See Goldsmith, supra note 6, at 33 (“OLC is, and views itself as, the frontline institution responsible for ensuring that the executive branch charged with executing the law is itself bound by law.”); Dawn E. Johnsen, Faithfully Executing the Laws: Internal Legal Constraints on Executive Power, 54 UCLA L. Rev. 1559, 1577–78 (2007) [hereinafter Johnsen, Internal Constraints] (explaining that “OLC’s legal interpretations typically are considered binding within the executive branch, unless overruled by the attorney general or the President (an exceedingly rare occurrence)” and noting OLC’s “tradition of accurate and principled legal advice”); Cornelia T.L. Pillard, The Unfulfilled Promise of the Constitution in Executive Hands, 103 Mich. L. Rev. 676, 703, 710–17 (2005) [hereinafter Pillard, Constitution] (discussing relative OLC independence and noting that OLC is “characterized by relative disengagement from [its] client entities, [is] staffed largely with career lawyers whose principal credentials are their legal skills, and ha[s] tended to foster within [its] own legal culture[] a distinction between politics and law”).

30 See Jack M. Beermann, Combating Midnight Regulation, 103 Nw. U. L. Rev. 352, 360 (2009) (discussing actions by the Obama Administration to reverse last-minute regulation by the Bush Administration). Such efforts also occurred eight years earlier, when the Bush Administration confronted a number of rules the Clinton Administration adopted just before leaving office. See Natural Res. Def. Council v. Abraham, 355 F.3d 179, 189–91 (2d Cir. 2004) (describing a Bush Administration attempt to reverse a Clinton-era rule published January 22, 2001).

31 See Jerry L. Mashaw, Recovering American Administrative Law: Federalist Foundations, 1787-1801, 115 Yale L.J. 1256, 1261–62 (2006) (discussing “internal law of administration” as consisting of internal instructions issued by higher level officials to control the exercise of discretion by their subordinates).

32 See 5 U.S.C. § 553(b)–(c) (2006) (describing notice and comment requirements). But see § 553(a) (exempting rules relating to agency management, including personnel matters, from notice and comment requirements); § 553(b)(3)(A) (similarly exempting “rules of agency organization, procedure, or practice”).

33 See Peter L. Strauss, Publication Rules in the Rulemaking Spectrum: Assuring Proper Respect for an Essential Element, 53 Admin. L. Rev. 803, 804–07 (2001) [hereinafter Strauss, Publication Rules] (describing publication and procedural constraints that can apply to agency guidance).

34These panels often form a core part of an agency’s regulatory approach, but they are composed primarily of outside experts. A good example is the use of scientific advisory panels in drug regulation. See Lars Noah, Scientific “Republicanism”: Expert Peer Review and the Quest for Regulatory Deliberation, 49 Emory L.J. 1033, 1054–57 (2000) (describing FDA use of technical advisory panels); see also 21 U.S.C. § 355(n) (2006) (setting out composition requirements for FDA advisory committees used in drug regulation and describing their functions). See generally 1 James T. O’Reilly, Food and Drug Administration § 13:94 (3d ed. 2007) (describing role of advisory committees in new drug applications). In 2007, Congress added new controls to help prevent conflicts of interest on FDA advisory panels. See Food and Drug Administration Amendments Act of 2007, Pub. L. No. 110-85, 121 Stat. 823 (to be codified at 21 U.S.C. § 379d-1).

35Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Sept. 30, 1993), amended by Exec. Order No. 13,422, 72 Fed. Reg. 2763 (Jan. 18, 2007).

36 See Peter L. Strauss, Overseer, or “The Decider”? The President in Administrative Law, 75 Geo. Wash. L. Rev. 696, 702 (2007) [hereinafter Strauss, Overseer] (“[Executive Orders 12,866 and 13,422] threaten the control of agency heads over their agencies’ agendas and effect a dramatic increase in presidential control over regulatory outcomes . . . .”). Although OMB enhances presidential control, it might also undermine an administration’s pro-regulatory agenda due to cost-cutting biases. See Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Regulatory State, 106 Colum. L. Rev. 1260, 1267 (2006) (suggesting that regardless of an administration’s political agenda, Executive Order 12,866 “contains within it several structural and institutional biases against regulation”). Moreover, as David Barron has argued, identifiable White House involvement in—and centralization of—policy setting may be less threatening to agency independence than some alternatives, such as increased politicization of agency appointments. See Barron, supra note 2, at 1120–21. See generally Terry M. Moe, The Politicized Presidency, in The New Direction in American Politics 235, 244–45 (John E. Chubb & Paul E. Peterson eds., 1985) (noting presidential incentives to centralize and politicize administration).

37 See Barron, supra note 2, at 1128, 1142 (noting a “surge in the number of politically appointed positions created during the first term of President George W. Bush” and describing the effect of politicization on EPA during the Bush Administration); Strauss, Overseer, supra note 36, at 701–02 (describing the expanded role of Regulatory Policy Officers and the requirement of political appointment); David E. Lewis, The Politics of Presidential Appointments: Political Control and Bureaucratic Performance 19–21, 137 (2008) [hereinafter Lewis, Presidential Appointments] (showing that the number of political appointees increased during the Bush Administration, and noting that today, half of all civilian workforce positions are exempt from the merit system compared to only ten percent in 1951); see also Paul C. Light, Thickening Government: Federal Hierarchy and the Diffusion of Accountability 7 (1995) (finding a 430 percent increase in the number of senior executives and presidential appointees from 1960 to 1992).

38 See Heidi Kitrosser, The Accountable Executive, 93 Minn. L. Rev. 1741, 1743, 1774 (2009) [hereinafter Kitrosser, Accountable] (arguing that a unitary executive approach undermines accountability by increasing the President’s ability to control information and “make or implement policy behind closed doors”); see also Cynthia R. Farina, The Consent of the Governed: Against Simple Rules for a Complex World, 72 Chi.-Kent L. Rev. 987, 992–1007 (1997) (arguing that accountability justifications for strong presidential power are premised on false understandings of popular will); Flaherty, supra note 22, at 1821–26 (similarly arguing that political accountability justifications for broad presidential authority rest on an unduly simple understanding of accountability); Matthew C. Stephenson, Optimal Political Control of the Bureaucracy, 107 Mich. L. Rev. 53, 55 (2008) (“[A] moderate degree of bureaucratic insulation alleviates rather than exacerbates the countermajoritarian problems inherent in bureaucratic policymaking.”).

39 See Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2263–64, 2331–46 (2001) (emphasizing the need for “direction and energy” as well as administrative constraints and arguing that presidential involvement leads to more accountable, responsive, and effective government); Lessig & Sunstein, supra note 19, at 98 (arguing that in the context of the modern administrative state, immunizing administrators from presidential control undermines separation of powers values by limiting accountability and increasing the risk of faction).

40 See Strauss, Formal and Functional, supra note 14, at 495 (“Any workable theory must not only avoid placing excessive power in the President’s hands, but also maintain his claim to a central and unifying governmental role—that is, to a relationship with all agencies that permits the exercise of his characteristic functions.”).

41Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 637 (1952) (Jackson, J., concurring) (“When the President takes measures incompatible with the expressed or implied will of Congress, his power is at its lowest ebb, for then he can rely only upon his own constitutional powers minus any constitutional powers of Congress over the matter.”).

42 See Kagan, supra note 39, at 2251, 2372–80 (arguing that “a statutory delegation to an executive agency official . . . usually should be read as allowing the President to assert directive authority” and proposing that judicial deference is most appropriate when evidence exists of actual presidential involvement); Strauss, Overseer, supra note 36, at 715 (finding a presidential role uncontroversial where “presidential authority readily fit the ‘oversight’ mold and/or may have been explicitly conferred by Congress”).

43For a recent analysis of the phenomenon of Executive Branch voluntary impositions, see Elizabeth Magill, Foreword: Agency Self-Regulation, 77 Geo. Wash. L. Rev. 859 (2009) [hereinafter Magill, Agency Self-Regulation].

44 See Lewis, Presidential Appointments, supra note 37, at 1, 55 (noting the abysmal federal response to Hurricane Katrina, specifically the role of political appointees, and arguing that “[s]ince voters and history judge presidents for the performance of the entire federal government during their tenure, this creates incentives for presidents to ensure that policy outcomes, both legislative and administrative, are under their control”); see also Moe, supra note 36, at 239 (explaining that Presidents are motivated by “political support and opposition, political strategy, and political tradeoffs,” and therefore value “‘responsive competence,’ not neutral competence”); Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. Econ. & Org. 81, 95 (1985) (asserting that Presidents, unlike legislators, are judged based on the effect of general government policies).

45Moe, supra note 36, at 239, 244–45; see also Barron, supra note 2, at 1102 (identifying concepts of centralization and politicization); Kagan, supra note 39, at 2339 (describing presidential incentives).

46Barron, supra note 2, at 1111–12.

47 Lewis, Presidential Appointments, supra note 37, at 195–97 (using the Bush Administration’s Program Assessment Rating Tool (PART) scores to find lower performance ratings of political appointees and suggesting that “reducing the number of political appointees is one means of improving performance”).

48 See Magill, Agency Self-Regulation, supra note 43, at 884–86 (discussing self-regulatory measures as central to helping top-level decision makers control the authority of their agents); see also Jerry L. Mashaw, Reluctant Nationalists: Federal Administration and Administrative Law in the Republican Era, 1801–1829, 116 Yale L.J. 1636, 1685–86 (2007) (“Administrators . . . fear the centrifugal effects of discretion vested in subordinates. If for no reason other than self-protection . . . they inevitably construct supervisory routines and modes of instruction to bend peripheral discretion toward centralized control.”); Strauss, Publication Rules, supra note 33, at 814 (“From an agency perspective, uniformity of administration nationwide is desirable and the agency may doubt whether its pool of not-so-well-paid inspectors will be able to handle so much discretion.”).

49 See Steven G. Calabresi & Saikrishna B. Prakash, The President’s Power to Execute the Laws, 104 Yale L.J. 541, 594–96 (1994) (arguing that the power of “removal, a power to act in [an inferior officer’s] stead, and a power to nullify [an inferior officer’s] acts when the President disapproves” are constitutionally required). For commentary critical of unitary executive claims, see Farina, supra note 38, at 987–89 (rejecting democratic legitimation argument of unitarians); Lessig & Sunstein, supra note 19, at 2–3 (“[T]he view that the framers constitutionalized anything like [the unitary] vision of the executive is just plain myth.”); and Strauss, Overseer, supra note 36, at 702–03 (finding the Constitution is “at best ambivalent on the question” of whether a President may direct agency determination of policy matters).

50 Stephen G. Calabresi & Christopher S. Yoo, The Unitary Executive 420–22 (2008) (direction of subordinates); id. at 422–23 (civil service); id. at. 423–25 (independent agencies).

51 See, e.g., Morrison v. Olson, 487 U.S. 654, 685–96 (1988). Free Enterprise Fund v. Public Company Accounting Oversight Board, a case currently before the Court and argued on December 7, 2009, may provide the Court with an opportunity to address the extent to which Executive Branch officials must be subject to direct presidential removal and the constitutionality of for-cause removal restrictions. See 537 F.3d 667 (D.C. Cir. 2008), cert. granted, 129 S. Ct. 2378 (2009).

52 See, e.g., Saikrishna B. Prakash, Fragmented Features of the Constitution’s Unitary Executive, 45 Willamette L. Rev. 701, 705 (2009) (“The power to create offices is not merely the limited power to create generic offices, leaving the President to determine each office’s functions and duties. Rather, when Congress creates a Secretary of Treasury or a Secretary of the Interior, it may establish the powers and duties of these offices.”); see also Kendall v. United States ex. rel. Stokes, 37 U.S. (12 Pet.) 524 (1838).

53Strauss, Overseer, supra note 36, at 715–18; see also Harold H. Bruff, Balance of Forces: Separation of Powers Law in the Administrative State 455–74 (2006).

54Even in the context of the military, the Court has insisted on Executive Branch adherence to congressional strictures. See, e.g., Hamdan v. Rumsfeld, 548 U.S. 557, 613–25 (2006) (invalidating use of presidentially-authorized military commission because its procedures violate governing statutes).

55 See Morrison, 487 U.S. at 685–96; Wiener v. United States, 357 U.S. 349, 356 (1958); Humphrey’s Ex’r v. United States, 295 U.S. 602, 629–31 (1935).

56If the Court determines that members of the Public Company Accounting Oversight Board are principal officers in the Free Enterprise case that is on the October 2009 term docket, it would not need to address the scope of the presidential removal power. See supra note 51.

57United States v. Perkins, 116 U.S. 483, 484 (1886); see also Myers v. United States, 272 U.S. 52, 160–63, 173–74 (1926) (acknowledging the constitutionality of removal restrictions on some inferior officers upheld in Perkins).

58549 U.S. 497, 532–34 (2007).

59546 U.S. 243, 258, 263–64 (2006).

60 See, e.g., Jody Freeman & Adrian Vermeule, Massachusetts v. EPA: From Politics to Expertise, 2007 Sup. Ct. Rev. 51, 52 (identifying these decisions as instances in which a majority was “worrie[d] about the politicization of administrative expertise”); Kathryn A. Watts, Proposing a Place for Politics in Arbitrary and Capricious Review, 119 Yale L.J. 2, 7, 18–23 (2009) (discussing administrative law’s resistance to political justifications for agency action).

61 The Federalist No. 48, at 305 (James Madison) (Clinton Rossiter ed., 1961).

62 See William N. Eskridge, Jr., Vetogates, Chevron, Preemption, 83 Notre Dame L. Rev. 1441, 1444–46 (2008) (describing barriers to legislative action).

63 Cf. Johnsen, Internal Constraints, supra note 29, at 1562 (arguing that Congress is an inadequate check on presidential power in part because of the ever-present possibility of presidential veto).

64 See, e.g., William G. Howell & Jon C. Pevehouse, While Dangers Gather: Congressional Checks on Presidential War Powers 10–27 (2007) (describing congressional powers in the face of executive intransigence); Douglas Kriner, Can Enhanced Oversight Repair “the Broken Branch”?, 89 B.U. L. Rev. 765, 773–75 (2009) (describing the importance of congressional oversight).

65Kagan, supra note 39, at 2311–12 (noting that “congressional parties have grown more ideologically coherent and partisan” but also arguing that divided power is the reality of modern government); Katyal, Internal Separation, supra note 22, at 2321 (“When the political branches are controlled by the same party, loyalty, discipline, and self-interest generally preclude interbranch checking.”). See generally Thomas E. Mann & Norman J. Ornstein, The Broken Branch: How Congress Is Failing America and How to Get It Back on Track (2006) (describing a shift in Congress from a highly decentralized, committee-based institution to a much more regimented one in which party increasingly trumps committee, and concluding that strong majority leadership in Congress leads not to a vigorous exertion of congressional authority but to a general passivity in the face of executive power, which results in the weakening of our system of checks and balances).

66The Bush Administration repeatedly demonstrated such resistance to congressional oversight, from its refusal to turn over documents relating to Vice President Cheney’s energy task force early in 2001, see Mark J. Rozell & Mitchel A. Sollenberger, Executive Privilege and the Bush Administration, 24 J.L. & Pol. 1, 12–18 (2008), to its invocation of executive privilege in 2007 to shield high-level presidential aides from testifying about the removal of seven U.S. Attorneys, see id. at 37–38, and its 2008 attempt to block access to an EPA draft document finding that greenhouse gases endanger the environment, see Heidi Kitrosser, Accountability and Administrative Structure, 45 Willamette L. Rev. 607, 609 (2009) [hereinafter Kitrosser, Administrative Structure]. The Bush Administration has also been accused of tampering with publicly released data. See, e.g., Freeman & Vermeule, supra note 60, at 55 (“[T]here were suggestions of widespread tampering by the Bush administration with the global warming data reported by numerous federal agencies, including EPA.”). Whether Congress was adequately informed of the Bush Administration’s program for expanded domestic national security wiretapping is a question currently in dispute. See, e.g., Perry Bacon Jr. & Joby Warrick, CIA Chief Rebuts Pelosi’s Charges; Panetta Says Lawmakers Were Told About Use of Interrogation Methods, Wash. Post, May 16, 2009, at A1.

67Johnsen, Internal Constraints, supra note 29, at 1587 (“Courts employ a variety of jurisdictional and prudential limitations that either preclude review—such as standing and the political question doctrine—or that result in only partial review.”); Pillard, Constitution, supra note 29, at 688–91 (describing “acute practical and legal limitations on the courts’ ability and willingness to decide many constitutional issues that confront the executive branch”).

68ACLU v. Nat’l Sec. Agency, 493 F.3d 644, 656 (6th Cir. 2007), cert. denied, 128 S. Ct. 1334 (2008) (“Because there is no evidence that any plaintiff’s communications have ever been intercepted, and the state secrets privilege prevents discovery of such evidence . . . the anticipated harm is neither imminent nor concrete—it is hypothetical, conjectural, or speculative. Therefore, this harm cannot satisfy the ‘injury in fact’ requirement of standing.” (citation omitted)).

69Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv. L. Rev. 1095, 1140 (2009) (“[W]here judges perceive an emergency, . . . standards of rationality, statutory clarity, evidence, and reasonableness all become more capacious and forgiving.”); see also Pillard, Constitution, supra note 29, at 692 (“In cases involving foreign policy, national security, military, or immigration judgments, the courts systematically apply doctrines of overt deference that cause them to refrain from full enforcement of constitutional norms, leaving that task to the political branches.” (footnotes omitted)).

70 See Summers v. Earth Island Inst., 129 S. Ct. 1142, 1151 (2009) (“[D]eprivation of a procedural right without some concrete interest that is affected by the deprivation—a procedural right in vacuo—is insufficient to create Article III standing.”); Heckler v. Chaney, 470 U.S. 821, 831 (1985) (“[The Supreme Court] has recognized on several occasions over many years that an agency’s decision not to prosecute or enforce, whether through civil or criminal process, is a decision generally committed to an agency’s absolute discretion.”).

71 See Pillard, Constitution, supra note 29, at 689 (noting that “even where private parties can get courts to respond to their constitutional harms, they may face interstitial deprivations,” including delay and irreparable injury).

72549 U.S. 497, 527–35 (2007).

73 E.g., Boumediene v. Bush, 128 S. Ct. 2229 (2008); Hamdan v. Rumsfeld, 548 U.S. 557 (2006).

74Kitrosser, Administrative Structure, supra note 38, at 608–09 (describing Bush Administration attempts to stall EPA rulemaking on climate change following Supreme Court ruling).

75 See Boumediene, 128 S. Ct. at 2275 (“In some of these cases six years have elapsed without the judicial oversight that habeas corpus or an adequate substitute demands.”).

76Matthew D. McCubbins et al., Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 Va. L. Rev. 431, 433 (1989) (“[E]ffective political control of an agency requires ex ante constraints on the agency (that is, a means of restricting the agency’s decisionmaking before it actually makes policy choices), one source of which is manipulation of its structure and process.”); see also Laura A. Dickinson, Outsourcing War and Peace (forthcoming 2010) (emphasizing the internalization of rules and norms as being a more effective check than external controls and describing efforts by the military to encourage such internalization, such as integrating JAG officers into command structures); Kenneth A. Bamberger & Deirdre K. Mulligan, Privacy Decisionmaking in Administrative Agencies, 75 U. Chi. L. Rev. 75, 106 (2008) (noting that ex ante controls may overcome shortcomings of ex post monitoring).

77 See Matthew D. McCubbins, Abdication or Delegation? Congress, the Bureaucracy, and the Delegation Dilemma, Regulation, Summer 1999, at 30, 33 (“[Congressional] leaders do not have to spend a lot of time looking for trouble. Waiting for trouble to be brought to their attention assures [Congressional] leaders that the trouble is important to constituents.”); Pillard, Constitution, supra note 29, at 690–91 (describing deficiencies of ex post judicial review in ensuring constitutionality of executive actions).

78 See Gillian E. Metzger, Ordinary Administrative Law as Constitutional Common Law, 110 Colum. L. Rev. (forthcoming Mar. 2010) (manuscript at 34, on file with author) [hereinafter Metzger, Ordinary Administrative Law] (discussing greater effectiveness of internally-generated reforms); Stephanie Stern, Cognitive Consistency: Theory Maintenance and Administrative Rulemaking, 63 U. Pitt. L. Rev. 589, 591 (2002) (suggesting that a “bias towards the maintenance of existing beliefs” makes agencies more receptive to new ideas before “lock-in” of policy preferences); Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 Colum. L. Rev. 458, 522 (2001) (“Judicially developed and imposed systems frequently trigger strong resentment and resistance, and they invite strategic behavior aimed at minimizing the impact of the law.”).

79 See supra text accompanying notes 27–28 (explaining how internal checks foster expertise and information generation); see also Mark Seidenfeld, Cognitive Loafing, Social Conformity, and Judicial Review of Agency Rulemaking, 87 Cornell L. Rev. 486, 515–16 (2002) [hereinafter Seidenfeld, Cognitive Loafing] (noting that agencies are more likely to respond to review where the reviewing audience is perceived as legitimate).

80 See M. Elizabeth Magill, Can Process Cure Substance? A Response to Neal Katyal’s “Internal Separation of Powers”, 116 Yale L.J. Pocket Part 126, 130 (2006), http://thepocketpart.org/2006/11/2/magill.html [hereinafter Magill, Response] (emphasizing that President Bush pursued post-September 11 policies despite awareness of internal dissent); see also William G. Howell, Political Checks on a Politicized Presidency: A Response to Neal Katyal’s “Internal Separation of Powers”, 116 Yale L.J. Pocket Part 111, 114–15 (2006), http://thepocketpart.org/2006/10/26/howell.html (noting “the essentially political nature of bureaucratic structure” and describing methods the Executive might employ to circumvent internal opposition).

81 See Doremus, supra note 5, at 1603–09 (describing allegations that Bush Administration political appointees interfered in work of agency scientists); Gilman, supra note 5, at 566 (noting media reports about “scientists who claimed they were censored, forced to alter their conclusions, and prohibited [by the Bush Administration] from issuing reports and attending conferences”); Mendelson, supra note 3, at 2164 (“[T]hough Office of Management and Budget clearance was not required for this decision, EPA officials reportedly consulted with the White House on the decision [to deny California’s waiver for stricter greenhouse gas regulations] anyway.” (footnote omitted)); Metzger, Administrative Regulation, supra note 3, at 880–81 (describing the FDA’s rejection of “recommendations of both its advisory committees and the directors and staff of the offices reviewing the application” for over-the-counter status for Plan B).

82 See Goldsmith, supra note 6, at 166–68 (describing the Bush Administration practice of “limit[ing] readership of controversial legal opinions [from OLC] to a very small group of lawyers” and specifically excluding the State Department from circulation of drafts of torture memos); Cornelia Pillard, Unitariness and Myopia: The Executive Branch, Legal Process, and Torture, 81 Ind. L.J. 1297, 1303–05 (2006) [hereinafter Pillard, Unitariness] (highlighting the importance of consultation within the Executive Branch and suggesting that key entities were not consulted during the drafting of torture memos); see also Jane Mayer, Annals of the Pentagon: The Memo, New Yorker, Feb. 27, 2006, at 32.

83 See Politicized Hiring, supra note 4, at 64 (“[A Justice Department official] considered political and ideological affiliations in hiring career attorneys and in other personnel actions affecting career attorneys in the Civil Rights Division. In doing so, he violated federal law . . . and [Justice] Department policy . . . .”); U.S. Attorneys Removal, supra note 4, at 356 (“[T]he process used to remove the nine U.S. Attorneys in 2006 was fundamentally flawed.”); Johnsen, Internal Constraints, supra note 29, at 1578 (“Many [former OLC attorneys] were deeply outraged and saddened by what they saw as a dramatic and dangerous deviation from the office’s tradition of accurate and principled legal advice.”); see also U.S. Gov’t Accountability Office, Food and Drug Administration: Decision Process to Deny Initial Application for Over-the-Counter Marketing of the Emergency Contraceptive Drug Plan B Was Unusual 5 (2005) (noting how Plan B’s denial letter deviated from consistent FDA practice).

84 See Magill, Response, supra note 80, at 130 (arguing that the belief that further internal process would impede the President was “mistaken” and that the Bush Administration faced internal dissent but “adopted the course that they did despite those objections”).

85Johnsen, supra note 29, at 1562 (emphasizing the role of internal legal constraints in the face of “inherent inadequacies of the courts and Congress as external checks on the President”).

86 Cf. Katyal, Internal Separation, supra note 22, at 2337–38 (recommending transfer of OLC’s adjudicatory function to an independent official, removable only for cause, but making that official’s decisions “subject to presidential overrule”). Katyal explains that “a presidential overruling of [an independent, internal executive adjudication] could trigger reporting to Congress,” id. at 2339, and could also foster accountability by checking unelected directors. Id. at 2338 (arguing that presidential override is appropriate because “lack of political accountability might dispose [directors] toward adventurism”).

87Daniel Carpenter’s fascinating study of the emergence of bureaucratic autonomy during the progressive era, in which he underscores the important role that external networks played, demonstrates that the dependence of internal constraints on external forces is not just a recent phenomenon. Daniel P. Carpenter, The Forging of Bureaucratic Autonomy 26–33 (2001).

88Steven Lee Myers & Philip Shenon, Embattled Attorney General Resigns, N.Y. Times, Aug. 27, 2007, http://www.nytimes.com/2007/08/27/washington/27cnd-gonzales.html; see also David Johnston & Eric Lipton, Ex-aide Disputes Gonzales Stand over Dismissals, N.Y. Times, Mar. 30, 2007, at A1 (noting resignation of Attorney General Gonzales’s Chief of Staff over dismissal of U.S. Attorneys); David Stout & David Johnston, A Top Aide to Gonzales Resigns, Becoming Latest Fallout Casualty, N.Y. Times, Apr. 7, 2007, at A1 (reporting resignation of Monica Goodling who helped “coordinate” dismissals of U.S. Attorneys).

89Indeed, former U.S. Attorney David Iglesias has argued that the scandal had a kind of paradoxical effect: “There was a sense at first that maybe it was going to make it less likely for U.S. attorneys to take more controversial cases, public corruption cases. It’s had the opposite effect. You’ve got U.S. attorneys that are really independent in a way they haven’t been in years.” Emma Schwartz, Looking Back on the Justice Department Scandal: A Conversation with Former U.S. Attorney David Iglesias, USNEWS.com, June 4, 2008, http://www.usnews.com/articles/news/politics/2008/06/04/looking-back-on-the-justice-department-scandal.html; cf. Evan J. Criddle, Fiduciary Foundations of Administrative Law, 54 UCLA L. Rev. 117, 162 (2006) (“[A]dministrators may care a good deal more about the reputational harms that flow from public censure than the legal consequences of an adverse judgment (for which they are not usually personally liable). An administrator’s perceived failure to act with reasonable prudence can have devastating reputational costs, as illustrated in the wake of Hurricane Katrina by the news media’s excoriation of Federal Emergency Management Agency (FEMA) director, Michael Brown.”).

90Gardiner Harris, F.D.A. Approves Broader Access to Next-Day Pill, N.Y. Times, Aug. 25, 2006, at A1; see also Gardiner Harris, Bush Picks F.D.A. Chief, but Vote Is Unlikely Soon, N.Y. Times, Mar. 16, 2006, at A18. Peter Strauss has noted how the need to obtain senatorial consent for a successor can curb presidential power to fire agency heads, thereby reinforcing agency independence and expertise. Strauss, Overseer, supra note 36, at 735–36.

91 Bart Gellman, Angler: The Cheney Vice Presidency 354–57 (2008) (describing pullback from more extreme positions on treatment of detainees following the Supreme Court’s decision in Hamdan); Goldsmith, supra note 6, at 125 (“Whenever the Supreme Court threatened to review one of the administration’s terrorism policies, [Solicitor General] Paul Clement was able to eke out small concessions from the White House [such as] more formal procedural protections for detainees.”). Cornelia Pillard has argued more generally that the authority of OLC and the Solicitor General within the Executive Branch depends largely on their being able to “backstop [their] judgments in judicial doctrine.” Pillard, Constitution, supra note 29, at 685.

92 See Gilman, supra note 5, at 605 (“Media reports [on politicization of science] spurred some government agencies to conduct internal investigations that generated new policies to protect agency scientists and promote transparency. Thus, the media clearly enhanced accountability . . . .”); see also Andrew C. Revkin, NASA Office Is Criticized on Climate Reports, N.Y. Times, June 3, 2008, at A16 (reporting that Michael Griffin, NASA agency head, quickly ordered review and policy changes when a pattern of distorting science was made public).

93 See Kitrosser, Administrative Structure, supra note 66, at 608–09 (explaining that the “White House refused to see EPA’s plans and did their best to ensure that others could not see them,” even refusing to open e-mail from EPA containing a draft document recommending pollution controls); see also Felicity Barringer, White House Refused to Open E-Mail on Pollutants, N.Y. Times, June 25, 2008, at A15; Juliet Eilperin, White House Tried to Silence EPA Proposal on Car Emissions, Wash. Post, June 26, 2008, at A2. Instead, EPA sent out notice seeking further comment, delaying the rulemaking until the end of the Bush Administration. Juliet Eilperin & R. Jeffrey Smith, EPA Won’t Act on Emissions This Year, Wash. Post, July 11, 2008, at A1.

94 See Matthew D. McCubbins & Thomas Schwartz, Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms, in Congress, Structure and Policy 426, 427–30 (Matthew D. McCubbins & Terry Sullivan eds., 1987) (arguing that Congress is likely to depend on outside sources, such as citizens and interest groups, for information directing oversight activities); Matthew D. McCubbins et al., supra note 76, at 434 (noting that the cost of oversight provides an incentive “to set up a system in which someone else (that is, a third party outside of the principal-agent diad) monitors the agent and reports acts of noncompliance”); Patricia M. Wald & Jonathan R. Siegel, The D.C. Circuit and the Struggle for Control of Presidential Information, 90 Geo. L.J. 737, 739 (2002) (explaining that Congress requires information from the Executive for legislative and investigatory tasks because “the Executive is the repository of the country’s most important information for public policy formulation”).

95Katyal, Internal Separation, supra note 22, at 2347 (“Over fifty inspectors general serve today, and they are structurally insulated from control by agency heads and required to report their findings biannually to Congress.”). See generally Light, Monitoring Government, supra note 26 (describing how the inspectors general assist cross-branch supervision by opening windows into the Executive Branch). As Seth Kreimer has argued, internal separation of powers constraints may also play a crucial role in enhancing transparency and accountability to external actors through the Freedom of Information Act (FOIA). Seth F. Kreimer, The Freedom of Information Act and the Ecology of Transparency, 10 U. Pa. J. Const. L. 1011, 1025–32, 1037–45 (2008). To be successful, requests under FOIA usually require prior knowledge, and requesters therefore largely depend on leaks from civil servants or other insiders to identify which claims for information to pursue. Id.

96 See McCubbins & Schwartz, supra note 94, at 427–30 (describing congressional dependence on “fire alarms”—outside interest groups that highlight areas requiring oversight, allowing Congress to focus its resources on the needs of constituents).

97Bruff, supra note 53, at 408 (“By training and inclination, bureaucrats seek legal authority for their actions. Accordingly, they constitute an often unappreciated bulwark to the rule of law in its everyday application to the citizen.”).

98 See, e.g., United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 249 (2d Cir. 1977) (“Adequate review of a determination requires an adequate record, if the review is to be meaningful.”).

99Cobell v. Norton, 334 F.3d 1128, 1142 (D.C. Cir. 2003) (citing Morgan v. United States, 304 U.S. 1, 18 (1938)).

100Freeman & Vermeule, supra note 60, at 52 (arguing that case law was driven by fear that executive expertise had been subordinated to politics).

101Tummino v. Von Eschenbach, 427 F. Supp. 2d 212, 230–34 (E.D.N.Y. 2006) (noting concern that agency action was not approved by scientific advisory committee in finding sufficient bad faith to order discovery into reasons for agency action beyond those evident on the administrative record). Inconsistency in an agency’s opinion, particularly over a recent period, often indicates that politics rather than expertise affected the agency’s decision and can lead to greater scrutiny. See, e.g., Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41–42 (1983) (“Accordingly, an agency changing its course by rescinding a rule is obligated to supply a reasoned analysis for the change beyond that which may be required when an agency does not act in the first instance.”). But see FCC v. Fox Television Stations, Inc., 129 S. Ct. 1800, 1810 (2009) (“We find no basis in the Administrative Procedure Act or in our opinions for a requirement that all agency change be subjected to more searching review.”).

102Joseph Landau, Muscular Procedure: Conditional Deference in the Executive Detention Cases, 84 Wash. L. Rev. 661, 694 (2009).

103 See Katyal, Internal Separation, supra note 22, at 2341–42 (explaining how congressional reporting requirements might strengthen internal checks); Pillard, Constitution, supra note 29, at 749–50 (arguing that transparency encourages internal executive constitutionalism, since the public put pressure on the President to comply with constitutional provisions); see also Johnsen, supra note 29, at 1596–97 (“Perhaps most essential to avoiding a culture in which OLC becomes merely an advocate of the administration’s policy preferences is transparency in the specific legal advice that informs executive action, as well as in the general governing processes and standards.”). In some cases, however, disclosure may actually weaken internal checks. See Pillard, Unitariness, supra note 82, at 1302–03 (arguing that routine publication of OLC legal opinions might discourage Executive Branch officials from seeking OLC’s legal advice in the first place).

104 See, e.g., Bruce Ackerman, The New Separation of Powers, 113 Harv. L. Rev. 633 (2000) (analyzing the impact of the tripartite system of government on federal bureaucracy); M. Elizabeth Magill, Beyond Powers and Branches in Separation of Powers Law, 150 U. Pa. L. Rev. 603, 605–06 (2001) [hereinafter Magill, Beyond Powers] (arguing government is separated not into “three undifferentiated branches,” but rather into a “large and diverse set of government decisionmakers,” and considering the implications of this idea for separation of powers analysis); Strauss, Place of Agencies, supra note 17, at 622–25 (emphasizing the import of separation of function concerns for separation of powers analysis).

105 See Katyal, Internal Separation, supra note 22, at 2316 (“The first-best concept of ‘legislature v. executive’ checks and balances must be updated to contemplate second-best ‘executive v. executive’ divisions.”).

106This linkage need not be specifically to Congress and the courts; other external forces—state and foreign governments, the media, and civil society organizations such as professional associations or advocacy groups—can also play a reinforcing role. As noted above, professional associations sought to publicize the Bush Administration’s interference with the work of government scientists, see supra text accompanying note 92, and advocacy organizations repeatedly have brought litigation to force disclosure of government policy.

107Although varied in their specific subject matter focus, many reform proposals have emphasized the importance of greater disclosure as a mechanism for reinforcing internal constraints. See supra note 103 (describing various proposals to promote Executive Branch transparency).

108Edmond v. United States, 520 U.S. 651, 658–66 (1997) (finding judges of the Coast Guard Court of Criminal Appeals to be “inferior officers,” and reviewing supervisory checks on their authority); Morrison v. Olson, 487 U.S. 654, 691 (1988); Humphrey’s Ex’r v. United States, 295 U.S. 602, 629 (1935).

109487 U.S. 654 (1988).

110 Id. at 685.

111 Id. at 670–77.

112 Id. at 692–93, 695–96.

113Roberto Suro, As Special Counsel Law Expires, Power Will Shift to Reno, Wash. Post, June 30, 1999, at A6 (discussing the lapse of the Independent Counsel Act).

114295 U.S. 602 (1935).

115Id. at 629–30.

116 Id.

117531 U.S. 457, 472–76 (2001).

118 Id. at 473. As Peter Strauss has noted, the Court also gave no weight to the fact that the process used by EPA in promulgating the air quality standards at issue involved independent checks on agency discretion, specifically participation by outside experts on the statutorily-mandated Clean Air Scientific Advisory Council. Peter L. Strauss, On Capturing the Possible Significance of Institutional Design and Ethos, 61 Admin. L. Rev. 259, 270–71 (2009) [hereinafter Strauss, On Capturing].

119 See Whitman, 531 U.S. at 487–90 (Stevens, J., concurring) (arguing that the power at issue was legislative).

120 See, e.g., Yakus v. United States, 321 U.S. 414, 425 (1944) (noting the power of the courts to “ascertain whether the will of Congress has been obeyed”); Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 55–57, 219–20 (1969) (suggesting that agencies should provide standards to limit their discretion); Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass’ns, 87 Cornell L. Rev. 452, 477, 479–82 (2001) (arguing that “the answer cannot be that agencies simply lack any authority whatsoever to adopt narrowing constructions in the delegation situation,” and that such authority should instead be derived from administrative law); Metzger, Ordinary Administrative Law, supra note 78 (manuscript at 7) (contrasting Whitman with prior case law looking to other checks, such as judicial review, to determine constitutionality).

121Boumediene v. Bush, 128 S. Ct. 2229, 2240 (2008).

122 Id. at 2247 (citation omitted).

123 Id. at 2269; see also id. at 2268 (“[T]he necessary scope of habeas review in part depends upon the rigor of any earlier proceedings . . . .”).

124 Hamdan v. Rumsfeld, 548 U.S. 557 (2006), is arguably another instance of judicial reinforcement of internal constraints, albeit more tacit. According to Neil Katyal, driving the result in Hamdan was the fact that internal Executive Branch experts on the Geneva Convention and the law of war opposed the President’s position. See Neal Kumar Katyal, Hamdan v. Rumsfeld: The Legal Academy Goes to Practice, 120 Harv. L. Rev. 65, 105, 109–12 (2006) (“The Justices consciously refused to award deference to the presidential determinations at issue because they lacked support from the bureaucracy, and in particular the Judge Advocates General and the State Department.”). Yet as Katyal acknowledges, Hamdan’s strong emphasis on the importance of congressional sanction at a minimum obscures the decision’s concern about the lack of internal expertise underlying the Administration’s position. Id. at 112–13; see also Hamdan, 548 U.S. at 593–95, 593 n.23, 601–03, 613; id. at 636 (Breyer, J., concurring) (explaining that Congress had denied the President the necessary authority “to create military commissions of the kind at issue here”). Instead, the more obvious incentive created by Hamdan was to encourage the Executive Branch to obtain congressional sanction for its policy of detaining and trying enemy combatants in military commissions, which the Bush Administration promptly proceeded to do and which resulted in enactment of the MCA. See Boumediene, 128 S. Ct. at 2242 (“Congress responded [to Hamdan] by passing the MCA . . . .”); Goldsmith, supra note 6, at 137–40 (explaining that following Hamdan, “only Congress could help the administration out of its predicament”).

125 See, e.g., Boumediene, 128 S. Ct. at 2262 (“In light of this holding [that non-citizens detained at Guantanamo are entitled to constitutional habeas protections] the question becomes whether the statute stripping jurisdiction to issue the writ avoids the Suspension Clause mandate because Congress has provided adequate substitute procedures for habeas corpus.”); Swain v. Pressley, 430 U.S. 372, 381 (1977) (“[S]ubstitution of a collateral remedy which is neither inadequate nor ineffective to test the legality of a person’s detention does not constitute a suspension of the writ of habeas corpus.”).

126United States v. Mead Corp., 533 U.S. 218, 226–27, 230 (2001).

127 See id. at 246 (Scalia, J., dissenting) (“Another practical effect of today’s opinion will be an artificially induced increase in informal rulemaking. Buy stock in the GPO.”); Matthew C. Stephenson, The Price of Public Action: Constitutional Doctrine and the Judicial Manipulation of Legislative Enactment Costs, 118 Yale L.J. 2, 31–32 (2008) (“[I]n the nonconstitutional administrative law context, whether an agency used formal decision-making procedures in promulgating an interpretation of a statute is a central factor courts consider in deciding whether the agency’s interpretation is entitled to deference.”).

128United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 249 (2d Cir. 1977) (“Adequate review of a determination requires an adequate record, if the review is to be meaningful.”); see also Vt. Yankee Nuclear Corp. v. Natural Res. Def. Council, Inc., 435 U.S. 519, 547 (1978) (“Monday morning quarterbacking [of the type engaged in by the lower court] not only encourages but almost compels the agency to conduct all rulemaking proceedings with the full panoply of procedural devices normally associated only with adjudicatory hearings.”). Interestingly, some proposed linkages between judicial review and other constraints have yet to find judicial favor. The Court does not give more deference to Executive Branch actions for which there is evidence of greater presidential oversight and sanction, see Kagan, supra note 39, at 2372 (advocating such deference), or of political involvement more generally, see Peter L. Strauss, Revisiting Overton Park: Political and Judicial Controls over Administrative Actions Affecting the Community, 39 UCLA L. Rev. 1251, 1329 (1992) (“The risks created by accepted judicial participation in the political process should lead judges to pay serious attention to the realities of political controls over administrative action before acting on the assumption that such controls will not prove effective.”). Nor is the involvement of state governments or expert advisory committees generally deemed an acceptable basis for expanded deference. But see Am. Trucking Ass’ns, Inc. v. EPA, 175 F.3d 1027, 1061 (D.C. Cir. 1999) (Tatel, J., dissenting) (arguing that because politically accountable state governments play a primary role in determining how to distribute burden or regulation, “courts have less reason to second-guess the specificity of the congressional delegation”); see also Strauss, On Capturing, supra note 118, at 271 (noting that in rejecting the nondelegation challenge in Whitman the Court gave no weight to the presence of an expert advisory committee as a check on agency decisionmaking).

129 See, e.g., Freeman & Vermeule, supra note 60, at 54 (“[J]ust as State Farm held deregulatory decisions reviewable, in order to allow a judicial hard look at a decision that allegedly injected politics into an expert judgment, so too MA v EPA held the denial of a petition requesting regulation to be reviewable, and for similar reasons.”). Katherine Watts has recently suggested that courts take a different approach, deferring to express acknowledgment by agencies of the role politics played in their decisionmaking—when such political calculations are allowable by statute—as a way to limit the extent to which politics undermines expertise in agency decisionmaking. See Watts, supra note 60, at 5 (“[W]hat counts as a ‘valid’ reason under arbitrary-and-capricious review should be expanded to include influences from the President, other executive officials and members of Congress, so long as these political influences are openly and transparently disclosed in the agency’s rulemaking record.”).

130 See United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260, 266–68 (1954). For discussions of Accardi, see Magill, Agency Self-Regulation, supra note 43, at 873–81, and Thomas W. Merrill, The Accardi Principle, 74 Geo. Wash. L. Rev. 569 (2006).

131 See Metzger, Ordinary Administrative Law, supra note 78, at 6–7 (noting contexts in which the Court has relied on ordinary administrative constraints to avoid constitutional violations); see also Bressman, supra note 120, at 479–81 (“[Whitman] shifts the source of authority for that requirement from constitutional law to administrative law . . . .”).

132 See supra note 130 and accompanying text (discussing Accardi). Many voluntary constraints, most notably Executive Order 12,866, providing for centralized regulatory review by the Office of Management and Budget (OMB), are expressly made not subject to judicial review. See Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Oct. 4, 1993), amended by Exec. Order No. 13,422, 72 Fed. Reg. 2763 (Jan. 18, 2007).

133I have argued elsewhere that the Court is similarly reluctant to openly acknowledge the constitutional basis of many core administrative law doctrines. See Metzger, Ordinary Administrative Law, supra note 120, at 4–5.

134U.S. Const. art. I, § 8, cl. 18.

135 See Boumediene v. Bush, 128 S. Ct. 2229, 2269 (2008) (explaining, in reviewing whether military commissions are an adequate substitute for habeas corpus, “[w]hat matters is the sum total of procedural protections afforded to the detainee at all stages, direct and collateral”); Walters v. Nat’l Ass’n of Radiation Survivors, 473 U.S. 305, 326 (1985) (“The flexibility of our approach in due process cases is intended in part to allow room for other forms of dispute resolution; with respect to the individual interests at stake here, legislatures are to be allowed considerable leeway to formulate such processes without being forced to conform to a rigid constitutional code of procedural necessities.”).

136Edmond v. United States, 520 U.S. 651, 662–63 (1997) (stating that “[w]hether one is an ‘inferior’ officer depends on whether he has a superior” and that “‘inferior officers’ are officers whose work is directed and supervised at some level by [principal officers]”); Morrison v. Olson, 487 U.S. 654, 671–72 (1988) (considering a variety of factors in determining that independent counsels are inferior officers).

137 See, e.g., Boumediene, 128 S. Ct. at 2246–47, 2263–74 (finding the writ of habeas corpus essential to the separation of powers, but considering the possibility of an adequate alternative); Clinton v. Jones, 520 U.S. 681, 697–706 (1997) (considering the role of the President and the judiciary and the extent to which participating in court proceedings would burden the President’s official duties); Morrison, 487 U.S. at 685–96 (weighing a range of factors to determine whether the independent counsel law upsets the constitutional balance).

138In this regard, Whitman’s formalistic emphasis on congressional guidance as the sole constitutional consideration in nondelegation challenges and its lack of attention to more functional considerations is particularly unsatisfying, especially given the decision’s functionalist justification for why limitations on delegations are not rigorously enforced. Whitman v. Am. Trucking Ass’ns, 531 U.S. 457, 474–75 (2001) (“[W]e have ‘almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.’” (quoting Mistretta v. United States, 488 U.S. 361, 416 (1989) (Scalia, J., dissenting))).

139In her article on presidential administration, for example, now-Solicitor General Elena Kagan famously argued that, to enhance accountability and improve effectiveness, courts should more readily defer to agency determinations when the President is more directly involved. Kagan, supra note 39, at 2372. Such deference, she maintained, would enhance important constitutional values, including accountability, id. at 2331–39, and effectiveness, id. at 2339–46, without undermining the separation of powers, id. at 2319–31 (accepting the argument that Congress may constitutionally limit presidential control of agency activity, but suggesting courts should generally interpret statues as providing for a presidential role).

140 See FCC v. Fox Television Stations,129 S. Ct. 1800, 1815–16 (2009) (“The independent agencies are sheltered not from politics but from the President, and it has often been observed that their freedom from presidential oversight (and protection) has simply been replaced by increased subservience to congressional direction.”); Calabresi & Prakash, supra note 49, at 583 (“[I]ndirect political control [by Congress] will necessarily exist with any so-called ‘independent’ agency or officer because absent presidential control, congressional oversight and appropriations powers become the only concern for the officers of the allegedly ‘independent’ agencies.”).

141 See John F. Manning, Federalism and the Generality Problem in Constitutional Interpretation, 122 Harv. L. Rev. 2003, 2040–47 (2009) (finding that constitutional decisions derived from background norms are at odds with the understanding of the Constitution as a finely crafted compromise).

142 See Gillian E. Metzger, Response: The Constitutional Legitimacy of Freestanding Federalism, 122 Harv. L. Rev. 98, 103–06 (2009) (“[T]he Court has a longstanding practice of invoking freestanding federalism in resolving disputes about the scope of federal and state powers or immunities.”).

143 See, e.g., Wilkinson v. Austin, 545 U.S. 209, 224–30 (2005) (finding detention procedures provided detainees to be adequate process); Forsyth County v. Nationalist Movement, 505 U.S. 123, 128–37 (1992) (finding parade ordinance invalid under First and Fourteenth Amendments).

144Indeed, administrative law scholars have often criticized the Court on exactly this basis, with a prime example being complaints that the Court’s decisions have served to ossify rulemaking. See, e.g., Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process, 41 Duke L.J. 1385, 1387–96, 1400–03, 1419 (1992) (explaining that the vigorous judicial review “of complex rulemaking” has led to ossification).

145This was the approach taken in Wyeth v. Levine last term, in which the Court indicated that it would determine the weight given to agency assessments of the burden that state laws impose on federal regulatory schemes based on the degree to which those assessments represent an “informed determination[].” Wyeth v. Levine, 129 S. Ct. 1187, 1201 (2009) (“The weight we accord the agency’s explanation of state law’s impact on the federal scheme depends on its thoroughness, consistency, and persuasiveness.”).

146For suggestions along these lines, see Metzger, Ordinary Administrative Law, supra note 78, at 7–12.

147 See, e.g., 5 U.S.C. § 553(a) (2006) (exempting matters involving military and foreign affairs from the scope of the Administrative Procedure Act (APA)); Franklin v. Massachusetts, 505 U.S. 788, 796 (1992) (holding that the President is not an agency under the APA).

148 Massachusetts v. EPA is perhaps the clearest recent example of this phenomenon. See Freeman & Vermeule, supra note 60, at 108–09 (suggesting that the Court in Massachusetts may have attempted to “nudge” agencies away from politicization and strong presidential administration by employing more rigorous scrutiny).

149 See Metzger, Ordinary Administrative Law, supra note 78, at 3 (“What is less often acknowledged . . . is the degree to which constitutional concerns permeate ordinary administrative law . . . .”).

150 See Scott Shane, A.C.L.U. Lawyers Mine Documents for Truth About Detainees and Interrogations, N.Y. Times, Aug. 30, 2009, at A4 (describing surprise at the success of the ACLU’s Freedom of Information Act requests with respect to the government’s detention policies and activities); see also Evan Perez & Deborah Solomon, Treasury Retreats on TARP Watchdog, Wall St. J., Sept. 2, 2009, at A3 (describing Republican lawmakers’ push for the independence of a special inspector general for the Troubled Asset Relief Program).