(Binding the Executive (by Law or by Politics

[أحمد أبو زنط]

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM







REVIEWS




(Binding the Executive (by Law or by Politics




Aziz Z. Huq




The Executive Unbound: After the Madisonian Republic


Eric A. Posner and Adrian Vermeule. Oxford, 2010. Pp 3, 249.






INTRODUCTION




Consider a recent snapshot of our imperial presidency. It is Thursday, September 25, 2008, in the Roosevelt Room of the White House. We are at the heart of the financial meltdown.1 Not two weeks before, investment bank Lehman Brothers filed for bankruptcy, jeopardizing hundreds of creditor counterparties, including major financial institutions. The following day, insurance giant American International Group (AIG) discloses enormous losses on credit default swaps, prompting the Federal Reserve to extend an emergency loan of $85 billion in exchange for a 79.9 percent equity stake in the company. One of the nation’s oldest money market funds, Reserve Primary Fund, experiences a run, collapsing to an unprecedented share price of less than a dollar and “breaking the buck.” It is clear the Treasury and the Federal Reserve need more funds to forestall a general liquidity crunch. In the Roosevelt Room, Treasury Secretary Henry Paulson Jr and Federal Reserve Chairman Ben Bernanke are addressing skeptical congressional leaders and the two presidential candidates, explaining


† Assistant Professor of Law, University of Chicago Law School.


Thanks to Daniel Abebe, Bernard Harcourt, Rick Hills, Trevor Morrison, Eric Posner, and Adrian Vermeule for their insightful and helpful comments, and to Eileen Ho for excellent research assistance. I am especially grateful to Professor Posner for graciously suggesting that I look closely at one of his books. I am pleased to acknowledge the support of the Frank Cicero, Jr Faculty Fund. All errors herein are mine alone.


1 This account draws mainly on Simon Johnson and James Kwak, 13 Bankers: The Wall Street Takeover and the Next Financial Meltdown 157–80 (Pantheon 2010); David Wessel, In Fed We Trust: Ben Bernanke’s War on the Great Panic 188–216 (Crown 2009). See also David M. Herszenhorn, Carl Hulse, and Sheryl Gay Stolberg, Day of Chaos Grips Washington; Fate of Bailout Plan Unresolved, NY Times A1 (Sept 26, 2008); Matthew Karnitschnig, et al, U.S. to Take Over AIG in $85 Billion Bailout; Central Banks Inject Cash as Credit Dries Up, Wall St J A1 (Sept 17, 2008).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


778 The University of Chicago Law Review [79:777


that the “limits of the Fed’s legal authority” have been reached and that legislative action is needed to prevent “a depression greater than the Great Depression.”


2 The meeting ends in disarray, a “partisan free-for-all.”3 And as the meeting breaks up, a desperate Paulson approaches Speaker of the House Nancy Pelosi, and—here’s the kicker—“literally bent down on one knee,” pleads for congressional action.4


It is close to a tenet of faith among constitutional scholars of diverse persuasions that ours is a republic dominated by the executive branch.5 Economies of bureaucratic scale, coupled with the executive’s primacy in responding to new security, economic, and environmental crises, are said to have frayed the Constitution’s delicate interbranch balance of powers.6 As a consequence, it is conventional wisdom that our President is now “imperial,” and Congress “broken.”7 Eric Posner and Adrian Vermeule—hereinafter collectively “PV”—are among the most sophisticated advocates of this dictum. But with a twist. Drawing on political science, game theory models, and the economics of agency relationships, their book The Executive Unbound: After the Madisonian Republic proposes that neither law nor legally constituted institutions (that is, Congress and courts) in practice impose meaningful constraints on the federal executive. This is so, PV say, not only in the heat of emergency but


2 Wessel, In Fed We Trust at 202 (cited in note 1).


3 Id at 215.


4 Herszenhorn, Day of Chaos, NY Times at A1 (cited in note 1).


5 For some recent examples, see Bruce Ackerman, The Decline and Fall of the American Republic 141, 184–85 (Harvard 2010) (describing an “institutional presidency [ ] on the march”); Sanford Levinson and Jack M. Balkin, Constitutional Dictatorship: Its Dangers and Its Design, 94 Minn L Rev 1789, 1810–43 (2010) (characterizing the American presidency as a “constitutional dictatorship”); William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L J 2446, 2449 (2006) (“We live in an era of increasing (and, some would say, increasingly unchecked) presidential power.”); Martin S. Flaherty, The Most Dangerous Branch, 105 Yale L J 1725, 1727 (1996) (“The dominance of executive power ought by now, to lift a phrase from Charles Black, to be a matter of common notoriety not so much for judicial notice as for background knowledge of educated people who live in this republic.”). See also John Yoo, Politics as Law? The Anti-ballistic Missile Treaty, the Separation of Powers, and Treaty Interpretation, 89 Cal L Rev 851, 874–75 (2001) (discussing “the President’s constitutional and structural superiority in conducting foreign affairs,” and the “overwhelming executive dominance in foreign affairs”).


6 See Aziz Z. Huq, Imperial March, Democracy 44, 46–53 (Winter 2008) (summarizing historical trends). For a more celebratory history of the rise of executive power, see generally Steven G. Calabresi and Christopher S. Yoo, The Unitary Executive: Presidential Power from Washington to Bush (Yale 2008).


7 See Arthur M. Schlesinger Jr, The Imperial Presidency 377–419 (Houghton Mifflin 2004); Thomas E. Mann and Norman J. Ornstein, The Broken Branch: How Congress Is Failing America and How to Get It Back on Track 1–13 (Oxford 2006).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 779


also in the ordinary run of administration (pp 4–5).


8 Succinctly put, their thesis is that “major constraints on the executive . . . do not arise from law or from the separation-of-powers framework” (p 4).9 But then, diverging from conventional narratives of executive supremacy, PV caution that the executive branch is not completely “unbound” either in ordinary times or in times of crisis.10 Echoing Madison, they locate a residual “primary controul on the government” in “the people.”11 Rather than legal rules or Congress, it is the strong undertow of democratic sentiment that tugs and binds executive discretion in practice. While The Executive Unbound is largely devoted to developing these descriptive claims with “social-scientific” precision (p 123)—a task PV pursue with verve, aplomb, and considerable force—it also gestures toward a normative claim: executive dominance within the bounds set by popular control is not merely inevitable but for the best.12


And yet . . . why then did Henry Paulson get down on one knee before Nancy Pelosi to ask for Congress to act? The timing and subject matter of the genuflection make his gesture especially puzzling. Paulson’s request concerned a highly technical matter, about which the executive had a clear institutional advantage.13 Increased sophistication in financial technologies of securitization and derivative design importantly impelled Lehman’s and then AIG’s crises.14 Few members of Congress have much inkling of “[t]he efficient market hypothesis, the capital asset pricing model, [and] the


8 “Legislators and courts [ ] are continually behind the pace of events in the administrative state; they play an essentially reactive and marginal role. . . . And in crises, the executive governs nearly alone. . . . Although we pay special attention to times of crisis, our thesis is not limited to those times” (pp 4–5).


9 Emphasis added.


10 The thesis of The Executive Unbound is also not constrained to a particular subject-matter domain, such as foreign affairs or national security law. For a crisp and effectively focused argument that trains on the former domain, see generally Daniel Abebe and Eric A. Posner, The Flaws of Foreign Affairs Legalism, 51 Va J Intl L 507 (2011).


11 Federalist 51 (Madison), in The Federalist 347, 349 (Wesleyan 1961) (Jacob E. Cooke, ed).


12 For a normative argument by one of the authors that courts ought to show greater deference to executive rule making and adjudication in certain cases, see Eric A. Posner and Cass R. Sunstein, Chevronizing Foreign Relations Law, 116 Yale L J 1170, 1204 (2007) (arguing that the executive is best placed to resolve difficult foreign affairs questions requiring judgments of policy and principle, and that the judiciary should defer to the executive based on its foreign policy expertise).


13 Johnson and Kwak, 13 Bankers at 9 (cited in note 1).


14 This increased sophistication was also the basis for regulators’ ex ante confidence that no crisis would likely emerge. See Raghuram G. Rajan, Fault Lines: How Hidden Fractures Still Threaten the World Economy 112–14 (Princeton 2010) (describing the famous “Greenspan put”).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


780 The University of Chicago Law Review [79:777


Black-Scholes option-pricing model.”


15 Compounding legislators’ institutional disadvantage, September 2008 was a moment of crisis. Treasury and Federal Reserve officials discerned a significant probability of catastrophic illiquidity in credit and repo markets. The need to act was believed acute. Under these tight conditions, legislators might be thought to “rationally submit[ ]” to executive demands (p 60). If ever a “blank check”16 was needed, this was surely the moment.


Of course, you know how the story ends. As recounted in The Executive Unbound, Congress did eventually relent and enact new fiscal authorities, although not without a full measure of indecorous wrangling (pp 47–48). As glossed by PV, “Congress pushed back somewhat” but was still “fundamentally driven by events and by executive proposals” (pp 50–51). The theatrics of a single Thursday, their account implies, do not provide an accurate synecdoche for the executive-legislative relationship.


But this moves too fast. For one thing, the executive, as much as Congress, was “driven by events” in the course of the financial crisis.17 For another, the claim that executive dominance characterizes interbranch relations in crisis is inconsistent with the behavior and beliefs of participants in the relevant events. Henry Merritt Paulson Jr, the six-foot-one former chief executive officer of Goldman Sachs, is hardly the sort to bow down lightly. Nor is his legalistic impulse isolated. In the midst of the summer 2011 debt ceiling showdown, President Barack Obama made a similar point:


I’m sympathetic to [the] view that this would be easier if I could do this entirely on my own. It would mean all these conversations I’ve had over the last three weeks, I could have been spending time with Malia and Sasha instead. But that’s not how our democracy works, and as I said, Americans made a decision about divided government that I wish I could undo.18


15 Justin Fox, The Myth of the Rational Market: A History of Risk, Reward, and Delusion on Wall Street 320 (HarperCollins 2009).


16 Hamdi v Rumsfeld, 542 US 507, 536 (2004) (plurality).


17 See David Skeel, The New Financial Deal: Understanding the Dodd-Frank Act and Its (Unintended) Consequences 12–13 (Wiley & Sons 2011) (criticizing the policy responses to the financial crisis as “ad hoc”); Steven M. Davidoff and David Zaring, Regulation by Deal: The Government’s Response to the Financial Crisis, 61 Admin L Rev 463, 466 (2009) (characterizing the government’s response to the financial crisis as animated by a “dealmaking ethos”).


18 Howard Portnoy, Obama: King for a Day?, HotAir Blog (July 23, 2011), online at http://hotair.com/greenroom/archives/2011/07/23/obama-king-for-a-day/ (visited Nov 17, 2011) (describing the town hall event where Obama appeared wistful about the thought of possessing the unilateral power to raise the debt ceiling); Steve Benen, Obama Weighs in on


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 781


A plausible political economy account of executive power, in my view, should be able to explain Paulson’s observed behavior and Obama’s expressed sentiments about “how our democracy works.” It should not write them off ab initio as either irrelevant or insincere.19


Paulson’s genuflection and Obama’s reticence, I will contend here, are symptomatic of our political system’s operation rather than being aberrational. It is generally the case that even in the heart of crisis, and even on matters where executive competence is supposedly at an acme, legislators employ formal institutional powers not only to delay executive initiatives but also affirmatively to end presidential policies.20 Numerous examples from recent events illustrate the point. Congressional adversaries of Obama, for instance, cut off his policy of emptying Guantلnamo Bay via appropriations riders.21 Deficit hawks spent 2011 resisting the President’s solutions to federal debt, while the President declined to short-circuit negotiations with unilateral action.22 Even in military matters, a growing body of empirical research suggests Congress often successfully influences the course of overseas engagements to a greater degree than legal scholars have discerned or acknowledged.23


‘Constitutional Option,’ Political Animal Blog (Wash Monthly July 22, 2011), online at http://www.washingtonmonthly.com/political-animal/2011_07/obama_weighs_in_on_constitutio 031036.php (visited Nov 17, 2011) (quoting President Obama as rejecting unilateral action to issue new debt).


19 See Part II.B (developing the argument in favor of taking such sentiments seriously at length).


20 See Part II.A (developing further examples).


21 See, for example, Ike Skelton National Defense Authorization Act for Fiscal Year 2011, Pub L No 111-383 § 1032, 124 Stat 4137, 4351–53 (prohibiting the expenditure of Pentagon funds on detainee transfers). Underscoring the salience of interbranch agreement in respect to national security initiatives, Professor Jack Goldsmith argues that President Bush’s counterterrorism initiative survived to the extent they were “vetted, altered, blessed—with restrictions and accountability strings attached—by the other branches of the U.S. government.” Jack Goldsmith, Power and Constraint: The Accountable Presidency after 9/11 xii–xiii (W.W. Norton 2012). Elsewhere, Goldsmith asserts a different theory of policy continuity when he claims that “Obama stuck with the Bush policies [because] many of them were irreversibly woven into the fabric of national security architecture.” Id at 27. I concur that it is plausible to think that intragovernmental resistance, arising from agencies’ sunk costs in the development of certain policies or institutional identification with those policies, provides an alternative explanation for the persistence of policies between administrations. It is a task for future empirical and analytic work to disaggregate which of the causes Goldsmith identifies is most important.


22 See Jeffrey Sparshott, Obama Skirts Question on 14th Amendment’s Place in Debt Talks, Wash Wire Blog (Wall St J July 8, 2011), online at http://blogs.wsj.com/washwire /2011/07/06/obama-skirts-question-on-14th-amendments-place-in-debt-talks/ (visited Nov 17, 2011) (quoting President Obama’s statement “I don’t think we should even get to the constitutional issue” implicated by the debt ceiling).


23 See Douglas L. Kriner, After the Rubicon: Congress, Presidents and the Politics of Waging War 285 (Chicago 2011) (arguing that “members of Congress have historically engaged


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


782 The University of Chicago Law Review [79:777


That work suggests that the failure of absolute congressional control over military matters cannot be taken as evidence of “the inability of law to constrain the executive” in more subtle ways (p 5). The conventional narrative of executive dominance, in other words, is at best incomplete and demands supplementing.


This Review uses The Executive Unbound as a platform to explore how the boundaries of discretionary executive action are established. As the controversial national security policies of the Bush administration recede in time, the issue of executive power becomes ripe for reconsideration. Arguments for or against binding the executive are starting to lose their partisan coloration. There is more room to investigate the dynamics of executive power in a purely positive fashion without the impinging taint of ideological coloration.


Notwithstanding this emerging space for analysis, there is still surprising inattention to evidence of whether the executive is constrained and to the positive question of how constraint works. The Executive Unbound is a significant advance because it takes seriously this second “mechanism question.” Future studies of the executive branch will ignore its important and trenchant analysis at their peril.24 Following PV’s lead, I focus on the descriptive, positive question of how the executive is constrained. I do speak briefly and in concluding to normative matters. But first and foremost, my arguments should be understood as positive and not normative in nature unless otherwise noted.


Articulating and answering the question “What binds the executive?”, The Executive Unbound draws a sharp line between legal and political constraints on discretion—a distinction between laws and institutions on the one hand, and the incentives created by political competition on the other hand. While legal constraints usually fail, it argues, political constraints can prevail. PV thus postulate what I call a “strong law/politics dichotomy.” My central claim in this Review is that this strong law/politics dichotomy cannot withstand scrutiny. While doctrinal scholars exaggerate law’s autonomy, I contend, the realists PV underestimate the extent to which legal rules and institutions play a pivotal role in the production


in a variety of actions from formal initiatives, such as introducing legislation or holding hearings that challenge the President’s conduct of military action, to informal efforts to shape the nature of the policy debate [on wars]”); William G. Howell and Jon C. Pevehouse, Presidents, Congress, and the Use of Force, 59 Intl Org 209, 228–29 (2005).


24 This Review does not catalog all of the many ways in which The Executive Unbound makes contributions to the study of the executive branch because it aims to build upon and extend that analysis.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 783


of executive constraint. Further, the political mechanisms they identify as substitutes for legal checks cannot alone do the work of regulating executive discretion. Diverging from both legalist and realist positions, I suggest that law and politics do not operate as substitutes in the regulation of executive authority.


25 They instead work as interlocking complements. An account of the borders of executive discretion must focus on the interaction of partisan and electoral forces on the one hand and legal rules. It must specify the conditions under which the interaction of political actors’ exertions and legal rules will prove effective in limiting such discretion.


Without embarking on the ambitious task of supplying a general theory of such interactions, I will suggest that an accurate political economy of executive restraint must identify a range of mechanisms in which both legal and political elements play roles. The primary aim of this Review is to clear ground for this account by rejecting “law only” and “politics only” explanations in favor of models with space for the interlocking operation of law and politics. To that end, I develop several examples that are suggestive of the potentially complementary operation of legal and political forces. I make no claim that these examples are exhaustive. To the contrary, I suspect that the dynamic interaction of legal rules and political forces takes many different forms depending on the background legal infrastructure and contingent features of the political environment. No brief catalog could capture their heterogeneity.


The caveats and modifications I offer to PV’s descriptive claims about the efficacy of law and politics may additionally have implications for a normative evaluation of executive dominance. If the latter is neither politically inevitable nor precisely calibrated, it cannot be assumed that executive policy choices are always for the best. I also point to recent changes in national politics that may be increasingly undermining the possibility of coproduced legal-political constraints. My positive account of executive constraint hence suggests that the executive may be, as PV’s title suggests, “unbounded,” but not for the reasons that many believe and not in ways that conduce to socially desirable outcomes. This, I conclude, should foster pessimism about the future trajectory of executive-led governance.


The Review proceeds in four parts. Part I sketches PV’s central theme—the strong law/politics dichotomy. Part II challenges PV’s


25 Posner and Vermeule argue similarly in The Executive Unbound that “de facto political constraints [ ] have grown up and, to some degree, substituted for legal constraints on the executive” (p 5) (emphasis added).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


784 The University of Chicago Law Review [79:777


claim that law is irrelevant to limiting executive discretion. Part III then closely parses their account of strong political constraints. My aim in both Part II and Part III is not to carp about PV’s basic rational choice methodology, even though it is not universally shared but often contested. Rather, I aim to develop reasons within that framework for questioning some of their conclusions. Turning from the critical to the constructive, Part IV proposes that rather than acting as substitutes, politics and law may work as complements to coproduce limits on executive actions, although such coordination may be increasingly rare and fragile.


I. THE NEW CRITS


A neophyte consumer of legal scholarship might be forgiven for thinking that one of the terms of admission to law’s ivory tower is a public affirmation that law on its own matters. Yet there is a long line of thinkers, going back to the Legal Realists at the turn of the twentieth century, who “sought to weaken, if not dissolve, the law-politics dichotomy”26 and to infuse legal analysis with social-scientific method.27 In the late 1970s, the Critical Legal Studies movement set itself the task of unpacking the “contradictions” of “mainstream liberal thought” in ways that sapped the formal robustness of legal categories.28 More recently, empirical scholars have aimed to “produc[e] a New Legal Realism—an effort to understand the sources of judicial decisions on the basis of testable hypotheses and large data sets.”29


The Executive Unbound stands squarely in this plural tradition.30 Although it briefly nods to empirical legal studies, it is also usefully read as a descendant of Critical Legal Studies, albeit one that advances a quite different and distinct political and institutional program. Tracking the so-called Crit methodology, PV “refuse to


26 William M. Wiecek, Liberty under Law: The Supreme Court in American Life 187 (Johns Hopkins 1988). See also Brian Z. Tamanaha, Understanding Legal Realism, 87 Tex L Rev 731, 734 (2009) (arguing that skepticism about the rule-bound nature of judging predated the Legal Realists by three decades).


27 See Laura Kalman, Legal Realism at Yale, 1927–1960 97 (North Carolina 1986) (“[T]he overriding concern of the average realist was to make the study of law more ‘realistic.’”).


28 Mark Kelman, A Guide to Critical Legal Studies 3 (Harvard 1987). See also Roberto Mangabeira Unger, The Critical Legal Studies Movement 8–11 (Harvard 1986) (identifying a “critique of formalism” as a central plank of the Critical Legal Studies agenda).


29 Thomas J. Miles and Cass R. Sunstein, The New Legal Realism, 75 U Chi L Rev 831, 831 (2008).


30 See Harvey Mansfield, Book Review, The Inevitable Imperial President, NY Times BR12 (Mar 13, 2011) (“Posner and Vermeule belong to the school of legal realism.”).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 785


accept the ordinary distinction between epiphenomenal law and ‘real’ society.”


31 They also share a goal with Critical Legal Studies: “[T]o identify the crucial structural characteristics of mainstream legal thought as examples of something called ‘liberalism.’”32


Echoing Crit themes, The Executive Unbound mounts a sustained assault on dominant pieties of legal scholarship—the autonomy and relevance of legal and constitutional rules.33 Its central target is “liberal legalism.” This is defined as the view that “representative legislatures govern and should govern . . . [and] that law does and should constrain the executive” (p 3). Liberal legalists, in PV’s telling, emphasize two ways in which law limits executive power. First, law binds via the separation of powers—that is, by fashioning Congress and the courts as institutional counterweights to the executive. Second, it works through framework statutes and by specifying absolute limits in the form of enacted statutes. What the legalist focus on institutions and legal rules fails to discern, PV contend, is that both such mechanisms are ineffectual on the ground (p 7).34 By contrast, it is political mechanisms that do the real work in imposing limits on what the executive can do. Fleshing out the idea of a political mechanism, PV identify a “reelection constraint” on Presidents operating “in a polity dominated by a mass public accustomed to exercising a large degree of democratic control” (p 12).35 This in turn fosters in the White House a need to sustain “popularity and credibility” (p 13). In consequence, opponents of political liberalism such as Weimar- and Nazi-era legal theorist Carl Schmitt exaggerate in their criticism of democratic rule and their advocacy of “decisionism” the claim that power rests in the entity with authority to determine when rules apply or not (pp 32–34, 90–91).36


31 Kelman, A Guide to Critical Legal Studies at 253 (cited in note 28). The Crits, though, are distinctive in their emphasis on the necessary internal contradictions hidden within legal rules, reflecting the pervasiveness of social conflict. See id at 258–62. The Crits were also much more skeptical of social science results than Professors Posner and Vermeule. See id at 167–71.


32 Id at 2.


33 For an extended critique of the dominant mode of liberal legalism in the international context, see Eric A. Posner, The Perils of Global Legalism xii (Chicago 2009) (criticizing legalism as a “view that loses sight of the social function of law and sees it as an end in itself”).


34 “[O]ur main critical thesis is that liberal legalism has proven unable to generate meaningful constraints on the executive” (p 7).


35 In my reconstruction of PV’s account, I take a broader view of the political. See text accompanying note 216.


36 PV have elsewhere recoded much of Schmitt’s argument (stripped of its Catholic theological apparatus) as mid-level institutional insights of a kind that are familiar from political and economic analysis. See Eric A. Posner and Adrian Vermeule, Demystifying Schmitt *1–2 (Harvard Public Law Working Paper No 10-47, Jan 26, 2011), online at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1723191 (visited Nov 17, 2011).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


786 The University of Chicago Law Review [79:777


Before fleshing out the strong law/politics dichotomy further, it is worth explaining briefly how PV prioritize the two strands of their argument. PV present their argument as the equal pursuit of “two main claims”—the fragility of law and the force of politics—and state at the threshold that their six-chapter book will be evenly split between the two theses (p 15). They indeed devote their first three chapters to the negative task of critiquing liberal legalism. But their fifth chapter then concerns the futility of (international) law as a constraint on executive action (pp 156–57).


37 And their final chapter attacks attitudes of suspicion toward the executive, which they call “tyrannophobia,” in order to demonstrate “a central fallacy of liberal legalism: the assumption that the only possible constraints on the executive are de jure constraints” (p 204). In all, five of their six chapters comprise attacks upon liberal liberalism. Only the fourth chapter discusses political constraints (pp 113–14). Viewed as a whole therefore, The Executive Unbound is principally a requiem for legal liberalism, not an ode to robust politics.


Consider first the case PV make against law and legal institutions as bulwarks against the executive. Their central argument rests on a logic of comparative institutional competence. Congress and judges alike, they argue, lack incentives or ability to gather and process information necessary to act quickly or to engage in oversight. Courts suffer from a “legitimacy deficit,” which dampens judicial willingness to intervene (pp 30–31, 57–58). And the separation of powers system can be gamed by an executive using a strategy of “divide and conquer” against the two other branches (pp 19–31).38 The net result is that Congress fails to anticipate crises and then is forced to delegate broad new powers after the fact (pp 43–52), while courts lag far behind executive initiatives.


PV also challenge the notion that framework statutes constrain the executive. Courts exercise a power of review pursuant to general framework statutes such as the Administrative Procedure Act.39 But,


37 This kind of argument has received extensive attention elsewhere. See, for example, Oona A. Hathaway and Ariel N. Lavinbuk, Book Review, Rationalism and Revisionism in International Law, 119 Harv L Rev 1404, 1420–26 (2006), reviewing Jack L. Goldsmith and Eric A. Posner, The Limits of International Law (Oxford 2005).


38 See also pp 118–22 (summarizing studies that show separated powers do not yield optimal economic policies).


39 Pub L No 79-404, 60 Stat 237 (1946), codified as amended in various sections of Title 5. For a specific provision of the Administrative Procedure Act that authorizes judicial review of decisions by executive agencies, see 5 USC § 702.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 787


PV contend, such review has little influence on outcomes (pp 35–37).


40 PV identify a series of exceptions and tractable standards in the doctrinal and statutory structure of administrative law that make law so malleable as to impose no effective resistance to executive action (pp 89–103). PV see law’s plasticity as an “inevitable . . . matter of institutional capacities” (p 105).41 Further, they predict that courts will anticipate the superior ability of the executive to deal with complex technological and economic problems and fall into line (p 31).42 Similarly, a “reluctant” Congress will find legal plasticity “inescapable” (p 108).


Does anything limit the presidency? PV readily concede that the President is not “all-powerful” and the White House “does face some checks even from a generally supine Congress” (p 61). They also recognize “a handful of great [Supreme Court] cases in which judges have checked or constrained discretionary executive action” but style these as the exception, rather than the rule (pp 30–31). And they warn that “the president can exert control only in certain areas” (p 59). They do not, though, closely examine any specific judicial or legislative action to identify the operative mechanisms of constraint. And PV’s concessions on these issues play no large role in their descriptive account. They thus make no claim that law has any systematic function in the political economy of executive constraint. Notwithstanding fleeting caveats, therefore, a fair-minded reader of The Executive Unbound likely finishes the book with the impression that PV are highly skeptical that law plays any meaningful or substantial role in checking the executive.


Instead, PV claim, the main reason Presidents are not “all-powerful” is political checks (p 61).43 PV’s account of political checks is grounded in a view of the President as an agent of the public.44 In this principal-agent model, the public (which is the principal) has imperfect information as to whether the President (the agent) is “well-motivated” in the sense of “choos[ing] the policies that voters


40 For example, PV review evidence of judicial responses to post-9/11 security measures and find that judicial orders account for less than 4 percent of Guantلnamo detainee releases (p 36).


41 Emphasis added. The claim is developed at greater length in Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv L Rev 1095, 1096–98, 1132–36 (2009).


42 “Legislators and judges understand that the executive’s comparative institutional advantages in secrecy, force, and unity are all the more useful during emergencies . . . .” (p 31).


43 PV claim that “Congress’s main weapon for affecting presidential behavior is not the cumbersome and costly legal mechanism of legislation. Rather legislators appeal to the court of public opinion, which in turn constrains the president” (p 61).


44 The theory was initially set forth in Eric A. Posner and Adrian Vermeule, The Credible Executive, 74 U Chi L Rev 865, 874–83 (2007).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


788 The University of Chicago Law Review [79:777


would choose if they knew what the executive knows” (p 130). The public will therefore deny the President rewards such as reelection unless it receives a credible signal that the President is “well-motivated.” Hence, Presidents need to build their “credibility” with the public by demonstrating good motives. The search for credibility induces limits on executive behavior (pp 122–24, 129–33). To maintain credibility, a well-motivated executive cannot rely on demonstrating good outcomes, for there is an imperfect correlation between policy choices and outcomes. Instead, the well-motivated executive must take actions that would be unfeasibly costly if it had undesirable motives (p 123). For example, a well-motivated exe-cutive will share power with political adversaries and disclose information to demonstrate its bona fides,


45 while an ill-motivated executive would find these actions too costly. Paradoxically then, it is the very breadth of presidential discretion that induces a need to sustain popular trust, which in turn leads to actions that limit the exercise of executive power (pp 150–53).46


Credibility is the main mechanism of political control analyzed in The Executive Unbound. But PV hint at others. They point briefly to “a wealthy and highly educated population, whose elites continually scrutinize executive action and tighten the constraints of popularity and credibility” (p 14). Publicity is said to work through a “complex process by which the views of elites, interest groups, ordinary citizens, and others ultimately determine the de facto lines of political authority” (p 78).47 To be sure, elections rely on informed voters (who may not always be available) and can be used only periodically (pp 114–16). And since the enactment of the Twenty-Second Amendment in 1947,48 second-term Presidents have not faced reelection. Nevertheless, PV propose, even second-term Presidents worry about their “policy legacy and their place in history” (p 13). As a consequence, there is a public-regarding friction on even final-period Presidents’ decisions.


45 Possible signals include using the establishment of independent agencies or commissions, bipartisan appointments, actions that cut against partisan priors (the “Nixon goes to China” strategy), information disclosure, multilateral endorsements of contentious foreign military actions, strict liability for damages from executive policies, and “precommitting” to results through statutes (pp 141–50).


46 See also p 13.


47 For a more parsimonious and hence more generalizable account of the inevitable “political and psychological” power of simple majorities, see Adrian Vermeule, The Force of Majority Rule *26–27 (Harvard Public Law Working Paper No 08-48, Oct 27, 2008), online at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1280201 (visited Nov 17, 2011).


48 US Const Amend XXII.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 789


To summarize, the strong law/politics dichotomy at the heart of The Executive Unbound rests on twin claims of law’s fragility and the effective force of politics. Rejecting traditional legal scholarship’s narrow focus on doctrine, PV’s theory predicts that Presidents can and do act forcefully except to the extent they perceive a credibility or publicity benefit from holding back.


49 Congress, the courts, and ex ante legal constraints, by contrast, are epiphenomenal and play little or no role. The account also has a normative sheen. By implication, executive dominance is not merely inevitable but to be welcomed given the presidency’s comparative advantage in policy making and in credibility-induced fidelity to democratic wishes.


II. THE SURPRISING RESILIENCE OF LAW


The Executive Unbound exposes canonical pieties about the efficacy of constitutional rules and statutory limits to corrosive scrutiny. Its account of law and legal institutions as weak forces, however, leaves no room for Paulson’s genuflection, Obama’s resistance to unilateral resolution of the debt crisis, or the current impasse over Guantلnamo. In each of these cases, the logic of comparative institutional advantage points toward strong, even unilateral executive action. In each case, the executive arguably gains little credibility from seeking another branch’s consent. yet in each case, the executive has in fact held back from action ex proprio vigore. And in each case, officials do so apparently on the basis of sincere beliefs about the effective force of law.


This Part develops a case for taking law and institutions more seriously. Specifically, I examine three strands that run through The Executive Unbound’s skepticism toward law to probe their limits. First, PV portray law as historically and presently ineffective. With only minor caveats, they showcase an executive almost never inhibited by ex ante legal rules. Second, PV depict law as lacking the ability to motivate political actors. The fact that one option is legal and another is not therefore is never counted as a reason for picking the first option. Finally, PV contend there is no theoretical account in liberal political thought that explains the efficacy of legal and institutional chains. Law on this account fails not just in practice and in the minds of political actors, but also on the pages of the theorists.


49 Hence, PV argued during the debt ceiling crisis that President Obama should act unilaterally not only because a failure to do so would be “catastrophic” but because he would have had broad political support. See Eric A. Posner and Adrian Vermeule, Obama Should Raise the Debt Ceiling on His Own, NY Times (July 22, 2011), online at http://www.nytimes.com /2011/07/22/opinion/22posner.html (visited Nov 17, 2011).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


790 The University of Chicago Law Review [79:777


This Part lodges exceptions to each of these strands. I begin by highlighting evidence from the presidential studies and legal scholarship that the President is often constrained by other branches and by at least certain laws. Second, I explore evidence of political actors’ normative preferences respecting legality and constitutionality. Finally, I highlight resources in liberal political theory that help explain why legal and institutional constraints are effective. History, theory, and political psychology, I aim to show, provide toeholds for the law by showing how political actors have both normative and instrumental reasons for complying with the law.


I should be clear that my goal here is not to suggest that PV’s skepticism about legal constraints is categorically unwarranted. Concern that Presidents can on occasion play fast and loose with the law is unquestionably grounded in fact. It does not follow from my analysis that law is always, necessarily, or automatically effective—PV persuasively show it is often not for reasons sketched in Part I. My point is rather that law cannot be dismissed so quickly and that PV’s treatment of law as functionally marginal understates its actual salience.50 At least in some nontrivial set of conditions, law is relevant to the imposition of an effective constraint on the executive. It therefore must be awarded a substantial role in any general political economy of the executive branch.


A. Historical Evidence of Executive Constraint via Law


The Executive Unbound paints an image of executive discretion almost or completely unbridled by law or coequal branch. But PV also concede that “the president can exert control only in certain [policy] areas” (p 59).51 They give no account, however, of what limits a President’s discretionary actions. To remedy that gap, this Section explores how the President has been and continues to be hemmed in by Congress and law. My aim here is not to present a comprehensive account of law as a constraining mechanism. Nor is my claim that law is always effective. Both as a practical matter and as a result of administrative law doctrine, the executive has considerable authority to leverage ambiguities in statutory text into warrants for discretionary action.52 Rather, my more limited aspiration here is to


50 See text accompanying note 43 (describing PV’s caveats).


51 See also Posner and Vermeule, 74 U Chi L Rev at 889 (cited in note 44) (“We neither make, nor need to make, any general empirical claim that Congress has no control over executive discretion.”).


52 See pp 94–109 (identifying mechanisms for generating discretion in the face of legal constraint and contending that legislators and judges allow such discretion for “quite practical


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 791


show that Congress and law do play a meaningful role in cabining executive discretion than The Executive Unbound credits. I start with Congress and then turn to the effect of statutory restrictions on the presidency.


Consider first a simple measure of Presidents’ ability to obtain policy change: Do they obtain the policy changes they desire? Every President enters office with an agenda they wish to accomplish.53 President Obama came into office, for example, promising health care reform, a cap-and-trade solution to climate change, and major immigration reform.54 President George W. Bush came to the White House committed to educational reform, social security reform, and a new approach to energy issues.55 One way of assessing presidential influence is by examining how such presidential agendas fare, and asking whether congressional obstruction or legal impediments—which could take the form of existing laws that preclude an executive policy change or an absence of statutory authority for desired executive action—is correlated with presidential failure. Such a correlation would be prima facie evidence that institutions and laws play some meaningful role in the production of constraints on executive discretion.


Both recent experience and long-term historical data suggest presidential agenda items are rarely achieved, and that legal or institutional impediments to White House aspirations are part of the reason. In both the last two presidencies, the White House obtained at least one item on its agenda—education for Bush and health care for Obama—but failed to secure others in Congress. Such limited success is not new. His famous first hundred days notwithstanding, Franklin Delano Roosevelt saw many of his “proposals for reconstruction [of government] . . . rejected outright.”56 Even in the midst of economic crisis, Congress successfully resisted New Deal initiatives from the White House. This historical evidence suggests that the diminished success of presidential agendas cannot be


reasons”); Ackerman, Decline and Fall 87–89 (cited in note 5) (arguing that the executive has developed new institutional tools to “give [a] constitutional imprimatur to presidential power grabs”).


53 See Paul C. Light, The President’s Agenda: Domestic Policy Choice from Kennedy to Clinton 2–3 (Johns Hopkins 3d ed 1999) (defining the agenda as a signal that “indicates what the President believes to be the most important issues facing his administration”).


54 Matt Welch and Nick Gillespie, What’s Next, Mr. President—Cardigans?, Wash Post B9 (July 19, 2009).


55 Matthew Eshbaugh-Soha and Tom Miles, George W. Bush’s Domestic Policy Agenda, 29 Am Rev Pol 351, 356–57 (2008).


56 Stephen Skowronek, The Politics Presidents Make: Leadership from John Adams to Bill Clinton 32 (Belknap rev ed 1997).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


792 The University of Chicago Law Review [79:777


ascribed solely to the narrowing scope of congressional attention in recent decades; it is an older phenomenon. Nevertheless, in more recent periods, presidential agendas have shrunk even more. President George W. Bush’s legislative agenda was “half as large as Richard Nixon’s first-term agenda in 1969–72, a third smaller than Ronald Reagan’s first-term agenda in 1981–84, and a quarter smaller than his father’s first-term agenda in 1989–92.”


57 The White House not only cannot always get what it wants from Congress but has substantially downsized its policy ambitions.


Supplementing this evidence of presidential weakness are studies of the determinants of White House success on Capitol Hill. These find that “presidency-centered explanations” do little work.58 Presidents’ legislative agendas succeed not because of the intrinsic institutional characteristics of the executive branch, but rather as a consequence of favorable political conditions within the momentarily dominant legislative coalition.59 Again, correlational evidence suggests that institutions and the legal frameworks making up the statutory status quo ante play a role in delimiting executive discretion.


But attention to the White House’s legislative agenda may be misleading. Perhaps the dwindling of legislative agendas is offset by newly minted technologies of direct “presidential administration.”60 The original advocate of this governance strategy has conceded, however, that presidential administration is available only when “Congress has left [ ] power in presidential hands.”61 Where there is no plausible statutory or constitutional foundation for a White House agenda-item, or where there is a perceived need for additional congressional action in the form of new appropriations or the like, Presidents cannot act alone.


The notion of a legislatively constrained presidential agenda is consistent with two canonical political science accounts of the contemporary presidency. Richard Neustadt, perhaps the most influential presidential scholar of the twentieth century,


57 Paul C. Light, Fact Sheet on the President’s Domestic Agenda *2 (Brookings Institute Oct 12, 2004), online at http://www.brookings.edu/papers/2004/1012governance_light.aspx (visited Nov 17, 2011).


58 See Jon R. Bond and Richard Fleisher, The President in the Legislative Arena x (Chicago 1990).


59 See id at 117 (concluding that it is “the distribution of partisan and ideological forces [among Congressmen that] sets the basic parameters of presidential success or failure in Congress”). This evidence reinforces the inference that it is not merely the shrinking of congressional agendas that drives a smaller presidential agenda.


60 Elena Kagan, Presidential Administration, 114 Harv L Rev 2245, 2282–90 (2001).


61 Id at 2251.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 793


encapsulated the Constitution’s system as one of “separated institutions sharing powers” in which “a President will often be unable to obtain congressional action on his terms or even . . . halt action he opposes.”


62 Writing in 1990, Neustadt concluded that the President “still shares most of his authority with others and is no more free than formerly to rule by command.”63 Neustadt’s finding of a weak presidency rested in part on his discernment of political constraints. But he also stressed “Congress and its key committees” as necessary partners in the production of policy.64 Neustadt thus identified institutions, as much as public opinion, as impediments to the White House.


In harmony with Neustadt’s view, Stephen Skowronek’s magisterial survey of presidential leadership suggests Presidents are not free to ignore or sideline Congress. Skowronek points out that “[i]t is not just that the presidency has gradually become more powerful and independent over the course of American history, but that the institutions and interests surrounding it have as well.”65 His complex argument (much simplified) situates presidential authority within a cyclical pattern of political “regime” creation, maintenance, and disintegration.66 In this cycle, the presidency is primarily a destructive force. Chief executives affiliated with past regimes have fewer tools at their disposal than oppositional leaders who “come[] to power with a measure of independence from established commitments and can more easily justify the disruptions that attend the exercise of power.”67 Executive discretion, in this account, is a function of a President’s location in the cycle of historical change. It is not a necessary attribute of the institution.


Skowronek also argues that Congress maintains and enforces prior regimes’ policy commitments against presidential innovation. He finds congressional abdication to be “virtually unknown to the modern presidency.”68 To the contrary, Skowronek contends, Congress has become more effective over time. Thomas Jefferson in the early 1800s, working with an “organizationally inchoate and politically malleable” legislature, had greater discretion than Ronald


62 Richard E. Neustadt, Presidential Power and the Modern Presidents: The Politics of Leadership from Roosevelt to Reagan 29, 32 (Free Press rev ed 1990) (emphasis omitted).


63 Id at 199 (emphasis added).


64 See id at 197, 199.


65 Skowronek, The Politics Presidents Make at 31 (cited in note 56).


66 See id at 34–52.


67 Id at 35.


68 Id at 418.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


794 The University of Chicago Law Review [79:777


Reagan in the 1980s.


69 By President Reagan’s time in office, the “governmental norms and institutional modalities” used to resist presidential initiatives had secured sufficient political capital to become resilient to presidential efforts at change.70 Until then, political movements proposing greater presidential authority also tended to advocate “some new mechanisms designed to hold [presidential] powers to account.”71 Skowronek provides a useful corrective to the assumption that historical change occurs only at one end of Pennsylvania Avenue. Echoing Neustadt’s analysis, his bottom line is that the contemporary executive remains “constrained by Congress”72 in ways that meaningfully hinder achievement of presidential goals.73


Nevertheless, neither Neustadt nor Skowronek articulate the precise role of law in congressional obstruction of presidential goals. Perhaps observed executive reticence is merely a result of political calculations, consistent with PV’s core hypothesis. But the evidence that the limits on executive authority tend to arise when Congress or existing law preclude a discretionary act suggests that institutions and statutes do play a meaningful role. Such correlations do not, however, establish the precise mechanisms whereby laws and institutions impose frictions on the employment of executive discretion.


Alternatively, perhaps the Neustadt and Skowronek accounts can be explained solely in terms of Congress’s negative veto in bicameralism and presentment, which is anticipated by the White House and so delimits the scope of presidential agendas. This would suggest that Congress’s power is asymmetrical: it can block some


69 See Skowronek, The Politics Presidents Make at 418 (cited in note 56).


70 See id. See also Keith E. Whittington and Daniel P. Carpenter, Executive Power in American Institutional Development, 1 Persp Polit 495, 508 (2003) (“In response [to unilateral executive action], Congress has fundamentally restructured itself in ways that would not have occurred in the absence of a more powerful executive branch.”).


71 Stephen Skowronek, The Conservative Insurgency and Presidential Power: A Developmental Perspective on the Unitary Executive, 122 Harv L Rev 2070, 2079 (2009). See also Kagan, 114 Harv L Rev at 2251, 2320 (cited in note 60) (emphasizing also the continuing possibility of congressional controls).


72 Whittington and Carpenter, 1 Persp Polit at 508 (cited in note 70) (describing, for example, how Congress has “developed entirely new institutions to challenge presidential budgetary decisions”).


73 Hence, in other work, Skowronek and Karen Orren explain how Congress in the early 1970s rewrote federal forest management law in order to restrict President Nixon’s ability to use the impoundment power or the new Office of Management and Budget for deregulatory ends, and “to promote Congress as the final arbiter of agency priorities.” See Karen Orren and Stephen Skowronek, The Search for American Political Development 168–69 (Cambridge 2004).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 795


executive initiatives but do little midstream to regulate the use of discretion powers already possessed by the presidency. Consistent with this interpretation, The Executive Unbound stresses the failure of framework laws passed after the Nixon presidency to regulate war and emergency powers (pp 86–87).


74 If the executive can so easily find work-arounds, PV explain, it follows that Congress also has less incentive to pass such laws. In the long term, the incentives for Congress to enact statutory limits on presidential authorities will accordingly atrophy.


There is some merit to this story. But in my view it again understates the observed effect of positive legal constraints on executive discretion. Recent scholarship, for example, has documented congressional influence on the shape of military policy via framework statutes. This work suggests Congress influences executive actions during military engagements through hearings and legislative proposals.75 Consistent with this account, two legal scholars have recently offered a revisionist history of constitutional war powers in which “Congress has been an active participant in setting the terms of battle,” in part because “congressional willingness to enact [ ] laws has only increased” over time.76 In the last decade, Congress has often taken the initiative on national security, such as enacting new statutes on military commissions in 2006 and 2009.77 Other recent landmark security reforms, such as a 2004


74 PV list the War Powers Resolution, the National Emergencies Act, the International Emergency Economic Powers Act, and the Inspector General Act as examples. Their analysis of security-related statutes, however, is incomplete. At least until 2001, the Foreign Intelligence Surveillance Act of 1978 (FISA), Pub L No 95-511, 92 Stat 1783, codified at 50 USC § 1801 et seq, proved effective in regulating federal electronic surveillance outside the criminal investigation context. The Freedom of Information Act (FOIA), Pub L No 89-487, 80 Stat 250 (1966), codified as amended at 5 USC § 552, which established an enforceable right of access to executive branch information, also likely influenced government behavior. FOIA litigation resulted in disclosures of significant documentary evidence of torture and abusive treatment of detainees. While both FISA and FOIA have their limits, and have been violated, neither is a wholesale failure.


75 See note 23. Some of the best-known examples concern American funding for the Nicaraguan Contras. See, for example, Department of Defense Appropriations Act, 1985 § 8066(a), Pub L No 98-473, 98 Stat 1837, 1935 (prohibiting expenditure of funds on aid to Nicaraguan Contras); Joint Resolution of Oct 3, 1984 § 106(c), Pub L No 98-441, 98 Stat 1699, 1700–01; Intelligence Authorization Act for Fiscal Year 1984 § 108, Pub L No 98-215, 97 Stat 1473, 1475; Department of Defense Appropriation Act, 1983 § 793, Pub L No 97-377, 96 Stat 1833, 1865.


76 David J. Barron and Martin S. Lederman, The Commander in Chief at the Lowest Ebb—A Constitutional History, 121 Harv L Rev 941, 947 (2008) (emphasis added).


77 See Military Commissions Act of 2009 (MCA), Pub L No 111-84, title XVIII, 123 Stat 2574, codified at 10 USC § 948 et seq; Military Commissions Act of 2006, Pub L No 109-366, 120 Stat 2600, 2602–03, codified as amended at 10 USC § 948a et seq.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


796 The University of Chicago Law Review [79:777


statute restructuring the intelligence community,


78 also had only lukewarm Oval Office support.79 Measured against a baseline of threshold executive preferences then, Congress has achieved nontrivial successes in shaping national security policy and institutions through both legislated and nonlegislated actions even in the teeth of White House opposition.80


The same point emerges more forcefully from a review of our “fiscal constitution.”81 Article I, § 8 of the Constitution vests Congress with power to “lay and collect Taxes” and to “borrow Money on the credit of the United States,” while Article I, § 9 bars federal funds from being spent except “in Consequence of Appropriations made by Law.”82 Congress has enacted several framework statutes to effectuate the “powerful limitations” implicit in these clauses.83 The resulting law prevents the President from repudiating past policy commitments (as Skowronek suggests) as well as imposing barriers to novel executive initiatives that want for statutory authorization.84


Three statutes merit attention here. First, the Miscellaneous Receipts Act of 184985 requires that all funds “received from customs, from the sales of public lands, and from all miscellaneous sources, for the use of the United States, shall be paid . . . into the treasury of the


78 Intelligence Reform and Terrorism Prevention Act of 2004, Pub L No 108-458, 118 Stat 3638.


79 See Michael Warner, Legal Echoes: The National Security Act of 1947 and the Intelligence Reform and Terrorism Prevention Act of 2004, 17 Stan L & Pol Rev 303, 310 (2006) (“It is true that President Bush himself was not the engine of intelligence reform, but it is also likely that without his efforts on its behalf, the Intelligence Reform Act would never have reached a vote in Congress.”).


80 The same has been said of federal responses to the 2008 financial crisis. See Davidoff and Zaring, 61 Admin L Rev at 465 (cited in note 17) (“Although the government never . . . acted as if it felt very constrained by the law that limited its actions, we think that its legal constraints help to explain a great deal of the government response.”).


81 Kenneth W. Dam, The American Fiscal Constitution, 44 U Chi L Rev 271, 271–72 (1977).


82 US Const Art I, § 8, cl 1, 2; US Const Art I, § 9, cl 7.


83 Kate Stith, Congress’ Power of the Purse, 97 Yale L J 1343, 1345 (1988).


84 Congress’s power of the purse is not a perfect mechanism of control. To control agency slack, Congress may have to choose between either overfunding an agency and allowing some waste, or underfunding an agency at the cost of not achieving policy goals entirely. See Michael M. Ting, The “Power of the Purse” and Its Implications of Bureaucratic Policy-Making, 106 Pub Choice 243, 264–65 (2001) (describing the “double-edged sword” of fiscal control). See also Christopher R. Berry, Barry C. Burden, and William G. Howell, The President and the Distribution of Federal Spending, 104 Am Polit Sci Rev 783, 786–87 (2010) (finding evidence that suggests the White House, through the Office of Management and Budget, exercises considerable post-legislative control of fiscal flows).


85 Act of Mar 3, 1849, 9 Stat 398, codified as amended at 31 USC § 3302(b).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 797


United States.”


86 It ensures that the executive cannot establish off-balance-sheet revenue streams as a basis for independent policy making. Second, the Anti-Deficiency Act,87 which was first enacted in 1870 and then amended in 1906,88 had the effect of cementing the principle of congressional appropriations control.89 With civil and criminal sanctions, it prohibits “unfunded monetary liabilities beyond the amounts Congress has appropriated,” and bars “the borrowing of funds by federal agencies . . . in anticipation of future appropriations.”90 Finally, the Congressional Budget and Impoundment Control Act of 197491 (Impoundment Act) channels presidential authority to decline to expend appropriated funds.92 It responded to President Nixon’s expansive use of impoundment.93 Congress had no trouble rejecting Nixon’s claims despite a long history of such impoundments.94 While the Miscellaneous Receipts Act and the Anti-Deficiency Act appear to have succeeded, the Impoundment Act has a more mixed record. While the Supreme Court endorsed legislative constraints on presidential impoundment,95 President Gerald Ford increased impoundments through creative interpretations of the law.96 But two decades later, Congress concluded the executive had too little discretionary spending authority and expanded it by statute.97


86 Act of Mar 3, 1849 § 1, 9 Stat at 398. The act is subject to a number of exceptions, none of which are relevant here. See Stith, 97 Yale L J at 1365–70 (cited in note 83).


87 Act of July 12, 1870, 16 Stat 230, codified as amended in various sections of Title 31.


88 Act of Feb 27, 1906 § 3, 34 Stat 49.


89 See Act of July 12, 1870 § 7, 16 Stat at 251 (providing that no governmental department shall expend in one year more than the appropriations made by Congress for that year).


90 Stith, 97 Yale L J at 1371 (cited in note 83) (stating that the Anti-Deficiency Act prevents unauthorized prior executive obligations from undermining Congress’s power of the purse).


91 Pub L No 93-344, 88 Stat 297, codified as amended 2 USC §§ 681–92.


92 Impoundment Act §§ 1001–17, 88 Stat at 332–39. This Act “began the use of statutory framework laws in this area, including establishing the concurrent budget resolution, setting the stage for the reconciliation process, and ultimately providing points of order and other internal enforcement provisions to increase congressional authority over the federal purse.” Elizabeth Garrett, The Purposes of Framework Legislation, 14 J Contemp Legal Issues 717, 723 (2005).


93 See Louis Fisher, Presidential Spending Power 158–59 (Princeton 1975) (describing President Nixon’s claims to an impoundment power); id at 176 (“The Nixon impoundments were unprecedented in their scope and severity.”).


94 Whittington and Carpenter, 1 Persp Polit at 507 (cited in note 70).


95 See Train v City of New York, 420 US 35, 41 (1975) (finding no presidential authority to withhold funds under the Clean Water Act).


96 Fisher, Presidential Spending Power at 200 (cited in note 93).


97 See Line Item Veto Act, Pub L No 104-130, 110 Stat 1200 (1996), codified at 2 USC §§ 691–92, abrogated in Clinton v New York, 524 US 417 (1998).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


798 The University of Chicago Law Review [79:777


Moreover, statutory regulation of the purse furnishes a tool for judicial influence over the executive. Judicial action in turn magnifies congressional influence. A recent study of taxation litigation finds evidence that the federal courts interpret fiscal laws in a more pro-government fashion during military engagements supported by both Congress and the White House than in the course of unilateral executive military entanglements.


98 Although the resulting effect is hard to quantify, the basic finding of the study suggests that fiscal statutes trench on executive discretion not only directly, but also indirectly via judicially created incentives to act only with legislative endorsement.99


To be sure, a persistent difficulty in debates about congressional efficacy, and with some of the claims advanced in The Executive Unbound, is that it is unclear what baseline should be used to evaluate the outcomes of executive-congressional struggles. What counts, that is, as a “win” and for whom? What, for example, is an appropriate level of legislative control over expenditures? In the examples developed in this Part, I have underscored instances in which a law has been passed that a President disagrees with in substantial part, and where there are divergent legislative preferences reflected in the ultimate enactment. I do not mean to suggest, however, that there are not alternative ways of delineating a baseline for analysis.100


In sum, there is strong evidence that law and lawmaking institutions have played a more robust role in delimiting the bounds of executive discretion over the federal sword and the federal purse than The Executive Unbound intimates. Congress in fact impedes presidential agendas. The White House in practice cannot use


98 Nancy Staudt, The Judicial Power of the Purse: How Courts Fund National Defense in Times of Crisis 77–86, 106–07 (Chicago 2011) (finding that the Supreme Court decided taxation cases “in a manner that strongly supported the elected branches of government, thereby increasing the size of the fiscal pie” during World War I, World War II, and the wars in Afghanistan and Iraq, but not during the Korean, Vietnam, and First Gulf Wars).


99 It is also worth noting that the absence of conspicuous executive failures to get laws passed is not especially probative. The system of bicameralism and presentment created by Article I, § 7, creates conditions in which “players realize that their preferences may have to be compromised to guarantee the cooperation of other players as required by the constitutional structure.” William N. Eskridge Jr, Philip P. Frickey, and Elizabeth Garrett, Cases and Materials on Legislation: Statutes and the Creation of Public Policy 77 (Thomson West 4th ed 2007). Under those conditions, the executive’s request of Congress will itself be a function of what Congress is likely to accept. Presidents are likely to have good information about how Congress will respond to proposals, and little incentive to seem weak by pushing for a new law that will be denied.


100 I am grateful to Professor Trevor Morrison for emphasizing to me the significance of the baseline problem.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 799


presidential administration as a perfect substitute. Legislation implementing congressional control of the purse is also a significant, if imperfect, tool of legislative influence on the ground. This is true even when Presidents influence the budgetary agenda


101 and agencies jawbone their legislative masters into new funding.102 If Congress and statutory frameworks seem to have such nontrivial effects on the executive’s choice set, this at minimum implies that the conditions in which law matters are more extensive than The Executive Unbound suggests and that an account of executive discretion that omits law and legal institutions will be incomplete.


B. The Motivational Status of Law


But why should Presidents attend to statutory constraints or Congress in the first place? What stops Henry Paulson from proceeding with the bailout without waiting for new appropriations? Or President Obama from sua sponte issuing new debt or transferring Guantلnamo detainees? The political economy developed in The Executive Unbound suggests that executive branch officials have no reason to heed legal and institutional constraints absent the possibility of credibility gains. PV also suggest that legislators and judges defer to the executive because of the latter’s superior institutional competence (pp 107–08).103 Recognizing the “inevitable” (p 103), they stay their hand rather than needlessly expend effort. This account of executive dominance, however, rests on an incomplete theory of political actors’ and judges’ motivations.104


PV’s rendition of the relevant motivations rests on rational choice foundations. Rational choice models take individuals as the


101 Some presidential control is a function of the Budget and Accounting Act of 1921, 42 Stat 20, codified as amended in various sections of Title 31, (vesting the President with proposal power respecting many parts of the federal budget). See Dam, 44 U Chi L Rev at 278 (cited in note 81) (stating the Act was “designed to substitute central presidential planning for the prior practice” of individual agency submissions to Congress).


102 See generally Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies, 1862–1928 (Princeton 2001) (finding, in the late nineteenth and early twentieth centuries, that mid-level administrators lobbied Congress to secure administrative reorderings).


103 PV further argue that legislators have “pragmatic reasons” to leave the executive with broad discretion (pp 107–08).


104 Consider Jon Elster, Explaining Social Behavior: More Nuts and Bolts for the Social Sciences 107 (Cambridge 2007) (“[E]xplanation[s] of [ ] behavior must often appeal to [ ] beliefs about the motivations of others.”).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


800 The University of Chicago Law Review [79:777


central unit of analysis.


105 They assume individuals “form rational beliefs, including beliefs about the options available to them” and then take actions that maximize preferences “given [those] beliefs.”106 Rational choice explanations come in “thick” and “thin” forms.107 Thin forms make no assumptions about the content of individuals’ preferences; thick accounts do.108 The strong law/politics dichotomy rests on “thick” rational choice assumptions.109 Political actors and judges are not only utility maximizers, their utilities also have defined content. Specifically, they have preferences over first-order policy outcomes, but not over second-order goals such as legality and constitutionality.110


This distinction between first-order and second-order preferences is not explicitly stated in The Executive Unbound. But it is omnipresent. The sole reason the President recognizes constraints is to obtain credibility that yields further “power” to achieve particular policy ends (p 153). Legislators capitulate before executive initiatives because they recognize them to be “inescapable” (p 108). Judges “remain quiet” because they recognize the “sharp pragmatic limits” on what they can do (pp 35–37). In all these arguments, political actors and judges are characterized as acting on the basis of expected policy outcomes. The possibility that their choices will reflect normative preferences for legality and constitutionality with a “dimension of ‘oughtness’”111 does not enter the analysis.112


This account of first-order preferences, which underwrites the law/politics dichotomy, embodies controversial assumptions. Notice, at the threshold, that arguments from inevitability or inescapability


105 See Donald P. Green and Ian Shapiro, Pathologies of Rational Choice: A Critique of Applications in Political Science 16 (Yale 1994).


106 Elster, Explaining Social Behavior at 191 (cited in note 104). See also Jon Elster, Introduction, in Jon Elster, ed, Rational Choice 1, 5 (NYU 1986) (“The normatively proper decision criterion [in rational choice theory] under risk is to choose the option that maximizes expected utility.”).


107 See Dietrich Rueschemeyer, Usable Theory: Analytic Tools for Social and Political Research 30 (Princeton 2009); Green and Shapiro, Pathologies of Rational Choice at 17–18 (cited in note 105).


108 See Green and Shapiro, Pathologies of Rational Choice at 17–18 (cited in note 105).


109 In this regard, the dichotomy is consistent with much recent political science work, which also rests on “unambiguously thick-rational assumptions.” See id at 19.


110 For this reason, I do not believe it is adequate to defend the use of thick preferences based on the need for a parsimonious analytical framework. It is true, to be sure, that excluding normative preferences from the analysis makes the latter more tractable. But when that exclusion preordains the answer to the question whether law or politics is more effective, I think it is an unwarranted simplification.


111 Rueschemeyer, Usable Theory at 72 (cited in note 107).


112 Note that my argument here does not concern the limits of rationality, such as those imposed by bounded rationality. Rather it trains on the stipulated content of preferences.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 801


cannot be literally true either for the courts or the legislative branch. It is not impossible for judges to issue timely preliminary injunctions. Nor is Congress necessarily disabled from quick action, as its first-blush response to 9/11 demonstrates. Rather, the inevitability argument relies on an implicit, unstated claim that judges and legislators accept comparative institutional competence arguments in favor of executive-branch primacy.


113 Courts and Congress, that is, are said to refrain from acting because they recognize that “institutional capacities” make it “inevitable” for the executive to take the lead (p 105).114 But it is not at all clear whether judges and legislators accept the “essentially normative” claim that “our nation would be safer . . . if judges [or Congress] appropriately deferred to their [ ] presidents.”115 What judges and legislators believe is an empirical question, a question on which The Executive Unbound adduces no evidence. Absent an empirical foundation, it nonetheless seems implausible (at least to me) to assert that federal judges and legislators have uniformly internalized a controversial logic that teaches them their own impotence.116


Equally peculiar, the strong law/politics dichotomy omits normative preferences respecting legality and constitutionality from political actors’ calculus. It thus rests on a strong assumption about the narrowly consequential nature of executive branch actors’ utility function. This is of concern for three reasons. First, a model that makes the predicate assumption that political actors do not have preferences over legality or constitutionality will always find political restraints to be more effective than legal ones. It is not clear law can ever explain fully official behavior if political actors have preferences over policy outcomes, but not over the legality of the methods used to obtain those outcomes. H.L.A. Hart famously argued that law rests ultimately on the fact that “officials of the system” view it as the source of “common standards of official behaviour” against which they “appraise critically their own and each other’s deviations as lapses.”117 If Hart’s claim is correct, officials’ “acceptance” of


113 For an argument that at least in the national security context, claims of comparative institutional competence founded on the American Separation of Powers are surprisingly fragile, see Aziz Z. Huq, Structural Constitutionalism as Counterterrorism, 100 Cal L Rev (forthcoming 2012).


114 See also p 31.


115 William G. Howell, Presidential Power in War, 2011 Ann Rev Pol Sci 89, 101.


116 On the other hand, PV’s claim may have in practice the circular quality of a self-fulfilling prophecy. That is, judges and legislators may come to believe that they are not institutionally qualified not out of direct experience but because they are repeatedly told so by influential commentators such as PV.


117 H.L.A. Hart, The Concept of Law 116–17 (Oxford 2d ed 1994).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


802 The University of Chicago Law Review [79:777


normative standards is the sociological fact upon which a modern legal system necessarily rests.


118 Absent such normative preferences, law has no grasp upon official behavior. In The Executive Unbound, it is categorically excluded from the domain of possible causes. By bracketing off normative preferences, the book thus stacks the explanatory deck against law.


Second, the omission of normative preferences about legality is in tension with the historical record.119 Ample evidence shows executive-branch officials to have normative preferences about legality and constitutionality. Deliberation on legal and constitutional questions within the executive branch is highly structured along channels that are reportedly entrenched.120 Recent insider accounts of national security lawmaking hence underscore thick “cultural norms” respecting the law within the executive branch, although they can also be read to suggest that the commitment to legality was occasionally uneven.121 Even the Bush administration, which has been accused of a cavalier attitude to the law, appeared to insist on the legality and constitutionality of its most controversial actions at some cost.122 There is also an extensive literature documenting how lawyers within the Justice Department take account of the normative force of law even when their clients within the executive branch are more cavalier.123


118 Id at 117. See also id at 6.


119 Consider Stephen Holmes, Passions and Constraint: On the Theory of Liberal Democracy 3 (Chicago 1995) (criticizing the “empirically implausible view that human beings are, for the most part, instrumentally rational and naturally self-regarding”).


120 See Executive Order No 12146, 3 CFR 409, 411 (1980) (providing for coordinated legal review of constitutional questions within the executive). See also Trevor W. Morrison, Stare Decisis in the Office of Legal Counsel, 110 Colum L Rev 1448, 1470–92 (2010).


121 See Jack Goldsmith, The Terror Presidency: Law and Judgment inside the Bush Administration 37 (Norton 2007). Goldsmith also explains how the availability of executive branch lawyers to argue “without any citation of authority” for sweeping executive power was a “godsend.” Id at 98.


122 See Benjamin A. Kleinerman, The Discretionary President: The Promise and Peril of Executive Power 4–5 (Kansas 2009) (describing and criticizing efforts to maintain legality during the Bush administration). It is especially worth noting that external accounts of military lawyers’ involvement in counterterrorism operations support the idea that law plays a role in their preferences. Goldsmith, Power and Constraint at 135–46 (cited in note 21) (discussing the influence of lawyers in targeting decisions).


123 See, for example, Trevor W. Morrison, Book Review, Constitutional Alarmism, 124 Harv L Rev 1688, 1707–30 (2011), reviewing Bruce Ackerman, Decline and Fall (cited in note 5); Randolph D. Moss, Executive Branch Legal Interpretation: A Perspective from the Office of Legal Counsel, 52 Admin L Rev 1303, 1305–06, 1308–09 (2000). I am assuming here that such insider accounts of executive legality are accurate. But there is insufficient independent evidence to ascertain whether this is so, or to ascertain whether political superiors will manipulate the occasions for advice seeking or the sources of advice to obtain sought-after outcomes (although some of Professor Morrison’s recent empirical work does support the


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 803


Of course, it is possible that all such anecdotal evidence reflects an optimistic hindsight bias on the part of insiders seeking to burnish their own credentials. I doubt this.124 It would indeed be surprising if federal officials did not generally take the law seriously given the normative force accorded to constitutional and legal norms in contemporary American society.125 All federal officers—not just lawyers within the Department of Justice—also swear or affirm a mandatory oath “to support this Constitution” before exercising their powers.126 Certainly it is conceivable that no federal official taking this oath has meaningful preferences over the constitutionality of his or her actions. This skeptical conclusion would be surprising, though, in light of the weak evidence that such oaths are routinely ignored.127


Alternatively, it may be that expressions of legalistic preferences are held only by lower-level officials, while senior policy makers have no illusions about the weak effect of the law. But recall that some of the examples of law talk I canvassed in the introduction came from senior policy makers, such as the President and the secretary of the treasury.128 To conclude that all use of legalism by senior officials is merely cheap talk without some substantial evidence on that score seems again incautious. This is especially so since both the President and the secretary of the treasury arguably paid a price in terms of nonattainment of policy preferences by sticking to their constitutional guns.


Third, where The Executive Unbound does take into account normative preferences, it does so by assimilating them to purely instrumental judgments about consequences. For example, PV claim courts will stay their hand because they lack “legitimacy.” Judges know they are ill-equipped to second-guess executive policy judg-ments, and so do not act for fear of losing public support (pp 30–31).129 This equates legitimacy with efficacy. It assumes judgments of legitimacy are correlated to policy outcomes. But that equation is inconsistent with available evidence. Studies of legitimacy do not


conclusions drawn here). See Morrison, 124 Harv L Rev at 1718 (cited in note 123). To some extent, therefore, my argument here rests on empirical points that are necessarily impressionistic. I am grateful to Professor Vermeule for pressing me on this last point.


124 To be clear, I draw here on no executive-branch experience of my own.


125 See, for example, Sanford Levinson, Constitutional Faith 9–53 (Princeton 1988) (examining the Constitution as the source of a “civil religion”).


126 US Const Art VI, § 3.


127 See note 123. See also Andrew D. Leipold, Rethinking Jury Nullification, 82 Va L Rev 253, 259 (1996).


128 See notes 4, 18, and accompanying text.


129 PV argue that courts defer to the executive because of a “legitimacy deficit” (p 30).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


804 The University of Chicago Law Review [79:777


show that views of, for example, the Supreme Court are a function of outcomes. To the contrary, support for the Court “has little to do with ideology or partisanship” but “is grounded in broader commitments to democratic institutions and processes.”


130 And external legitimacy judgments of law enforcement bodies, even in the fraught context of national security, are not driven by perceptions of efficiency but by ideals of fairness and procedural justice.131 Scholars of all ideological stripes tend to endorse the notion that “a reputation for restraint and commitment to the rule of law” will “legitimate the extraordinary powers the President must exercise in the long term” against national security threats.132 Cross-national studies of legitimacy also identify a complex bundle of legitimacy predictors, including participation rights, welfare rights, and accountability.133


The Executive Unbound’s view of legitimacy is in any event symptomatic of a more diffuse skepticism of normative preferences. No doubt this captures the standpoint of some official actors, who really do take the perspective of a Holmesian bad man. But that seems inadequate as a more general description of contemporary political actors’ beliefs and motivations. Foolishly or not, American officials often appear to hold strong views about legality and the Constitution. A positive political economy of executive constraint and discretion is surely incomplete without an accounting of those preferences.


C. Political Theory and Legal Constraint


A third strand of The Executive Unbound’s skepticism about legal constraints is its critique of the Madisonian theory of separation of powers (pp 19–31). Liberal legalism is said to be inadequate


130 James L. Gibson and Gregory A. Caldeira, Citizens, Courts, and Confirmations: Positivity Theory and the Judgments of the American People 61 (Princeton 2009). See also Gregory A. Caldeira and James L. Gibson, The Etiology of Public Support for the Supreme Court, 36 Am J Pol Sci 635, 644–45 (1992) (reporting no relationship between diffuse support for the Court and specific policy goals).

[أحمد أبو زنط]

131 See Aziz Z. Huq, Tom R. Tyler, and Stephen J. Schulhofer, Why Does the Public Cooperate with Law Enforcement? The Influence of the Purposes and Targets of Policing, 17 Psych, Pub Pol, & L 419, 426–38 (2011) (presenting evidence about the predictors of public attitudes to both crime and counterterrorism); Tom R. Tyler, Stephen J. Schulhofer, and Aziz Z. Huq, Legitimacy and Deterrence Effects in Counterterrorism Policing: A Study of Muslim Americans, 44 L & Society Rev 365, 379–86 (2010) (presenting analogous evidence concerning Muslim Americans and counterterrorism).


132 Goldsmith, Power and Constraint at 42 (cited in note 21).


133 See Bruce Gilley, The Right to Rule: How States Win and Lose Legitimacy 42–43 (Columbia 2009).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 805


because the theory upon which it rests does not have within it adequate resources to explain the constraining effect of law and institutions. Yet a narrow focus on the work of James Madison may slight the resources of political theory. A broader review of the political theory corpus surfaces instrumental reasons for the powerful to conform to legal rules that supplement the normative effects discussed above.


To begin, it is not clear Madison was mainly concerned with the scope of executive authority. His imagination seized by states’ experiences after independence, Madison perceived popular majorities acting through the legislative branch as the principal source of destabilizing political risk.134 He anxiously contrasted the “weight of the legislative authority” against the “weakness of the executive.”135 Worse, Madison mustered only an imperfect inkling of how the federal executive would work after the Founding. As PV rightly observe, the text of Article II failed to resolve many design questions, leaving it uncertain whether the presidency would be much more than “a ministerial position” (p 184). Madison and other Framers also profoundly misunderstood how chief executives would be selected. They assumed the Electoral College would rarely yield a majority behind a single presidential candidate, and so it would usually be the House of Representatives that selected the chief executive.136 It is hardly surprising under those conditions that Madisonian political science derives an imperfect account of legal constraints on the contemporary executive.


But other strands of political theory now available today do provide explanations for why powerful political actors, even absent proximate adverse consequence for violating the law, still have cause to conform their conduct with ex ante legal rules. In a comprehensive account of liberalism’s view of law as a constraint on government, Stephen Holmes historicizes the notion that even a sovereign with “strictly unconditional” power can further its own interests by imposing ex ante constraints via other institutions or laws.137 That idea goes back to the sixteenth-century theorist of monarchy Jean Bodin.138 Bodin demonstrated how “[c]onstitutional


134 See Federalist 48 (Madison), in The Federalist 332, 333 (cited in note 11) (“The legislative department is every where extending the sphere of its activity and drawing all power into its impetuous vortex.”).


135 Federalist 51 (Madison) at 350 (cited in note 11).


136 See Jack N. Rakove, Original Meanings: Politics and Ideas in the Making of the Constitution 265–66 (Knopf 1996).


137 See Holmes, Passions and Constraint at 106–08 (cited in note 119).


138 See id at 7, 106.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


806 The University of Chicago Law Review [79:777


constraints may be an indirect technique for building effective state institutions and reinforcing governmental power.”


139 Beginning from the premise that no monarchy is “self-sufficient” in its ability to secure compliance from officials and subjects, Bodin argued that legal constraints bridge the gap.140 To sustain obedience, the prince recognizes rights and shares authority to secure cooperation and mitigate resistance.141 Restated in more familiar contemporary terms, Bodin argued that powerful actors benefited from legal compliance in situations of repeat play.142 By distributing the labor of governance, diversifying information sources, and co-opting resistance, constitutional and legal rules conduce to stability.143 This observation is supported by empirical work finding that “nominally democratic” institutions within an autocracy can be successfully employed to “solicit cooperation and to neutralize the threat of rebellion from forces within society.”144 Even an all-powerful sovereign for this reason has instrumental reasons for compliance with the law and employment of putatively power-diffusing institutions.


Whereas Bodin’s monarchs were sometimes seen as standing above the law, the federal Constitution that creates the presidency both empowers its occupant through quadrennial elections and also limits the officeholder’s formal powers. The text of the Constitution, that is, bundles enabling rules and constraining rules together as a take-it-or-leave-it package. Oath rules require officials to endorse the Constitution wholesale, not retail.145 While some parts of the constitutional text have fallen into desuetude,146 no political actor openly claims the right to pick and choose among constitutional


139 Id at 133.


140 Id at 117 (arguing that the monarchy is “rel[iant] on prescription, habit, [and] custom” to maintain control).


141 See Holmes, Passions and Constraint at 116–17 (cited in note 119).


142 See Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of Constitutional Commitment, 124 Harv L Rev 657, 711 (2011).


143 See Holmes, Passions and Constraint at 6 (cited in note 119); Levinson, 124 Harv L Rev at 711 (cited in note 142). This anticipates an argument of Max Weber, who argued that ordinary administration “requires that human conduct be conditioned to obedience towards those masters who claim to be bearers of legitimate power.” Max Weber, Politics as a Vocation, in H.H. Gerth and C. Wright Mills, eds, From Max Weber: Essays in Sociology 77, 80 (Oxford 1946). Because force alone cannot sustain compliance with the law, legitimacy is necessary to the production of stability. See id at 78–79.


144 Jennifer Gandhi and Adam Przeworski, Authoritarian Institutions and the Survival of Autocrats, 40 Comp Polit Stud 1279, 1280–81 (2007).


145 See Levinson, 124 Harv L Rev at 695 (cited in note 142) (“[B]undling should dial up the positive political feedback effects of decisionmaking institutions, making them more strongly self-reinforcing on average than discrete policy outcomes.”).


146 See, for example, US Const Art IV, § 4 (“The United States shall guarantee to every State in this Union a Republican Form of Government.”).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 807


provisions.


147 A President cannot categorically repudiate the Constitution or the laws that arise under it without repudiating the foundations of his or her own authority. He or she is thus bound tighter than Bodin’s sovereign.


To be sure, this bundling effect does not wholly preclude a political actor today from taking advantage of empowering elements in the Constitution while attempting to suppress its constraining features. It does though impose a tax on efforts to leverage constitutional power without constitutional constraints, if only via the “civilizing force of hypocrisy.”148 It also means that credible executive claims to authority must be advanced in a field of shared understandings about the law and the Constitution in which the President has no monopoly on semantic meaning. Mere presidential diktat cannot often oust enduring norms that have accrued wide acceptance and are backed by others’ asset- or office-specific epistemic and normative commitments. To the extent the legal culture maintains resources to distinguish good from bad arguments, executive efforts to unbundle constitutional license from constitutional limitation will be met with resistance.


D. Summary


In principal part, The Executive Unbound comprises a sustained assault on the efficacy of law as a limit on executive discretion. It finds thin evidence of compliance with law because it is law. This Part has parsed out three important elements within that assault—the claim that law is not observed to work on the ground; the claim that it provides no motivation heeded by political actors; and the claim that it lacks a theoretical basis. I have attempted to demonstrate here that each of these strands can be countered. Congress can impede the President from achieving sought-after goals, most profoundly by exercise of its fiscal powers. Executive actors have both normative and instrumental reasons to take legality and constitutionality into account. And notwithstanding Madison’s blinders, political theory elaborates purely instrumental grounds for compliance with law. The end result of my analysis may not be a demonstration that law is


147 It is not immediately clear why American political culture has developed this all-or-nothing view of constitutional fidelity even if the near-universal acceptance of the view is hard to gainsay. It may be that the comprehensive character of observed constitutional commitments is contingent but has proved essential to the maintenance of persisting beliefs in the Constitution’s binding character. That is, in the absence of an all-or-nothing view of the Constitution, the latter would cease to operate as an effective focal point for governance.


148 Jon Elster, Arguing and Bargaining in Two Constituent Assemblies, 2 U Pa J Const L 345, 413 (2000).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


808 The University of Chicago Law Review [79:777


always or pervasively effectual. But it does show that law plays a tangible role in determining the bounds of executive action.


The Executive Unbound also has a normative thread, counseling in favor of accepting executive dominance on grounds of the executive’s comparative institutional advantage. The arguments marshaled here counsel for some caution respecting that claim. A broad executive dominance thesis is in tension with the incentive structures of political actors and the entrenched distribution of fiscal powers in particular. If Presidents rely on Congress to achieve their agenda, they must engage in negotiation and compromise on policy choices. This may include log-rolling across different policy domains. Deal-making of this kind will have the result that observed policy choices reflect nothing more than the specific bargain reached at a given moment in time in light of the existing distribution of political preferences. Moreover, Presidents may take positions that reflect not their optimal policy selection but their strategic anticipation of Congress’s or the judiciary’s responses. Under those conditions, there is no strong reason to believe that the executive will always press for the socially desirable policy choice either at the inception or close of the policy-making process.


In sum, without rejecting the important weaknesses in legal constraints that PV identify, it is possible to think that law plays a substantial role in limiting executive discretion. This role, which is underpinned by both normative and instrumental motives, seems largest in fiscal matters but can also be discerned on security questions, too. This suggests a political economy of executive constraint should have some substantial place in it for law.


III. THE FRAGILITY OF POLITICS


This Part turns to the second element of the strong law/politics dichotomy: the thesis that political forces bind the executive in ways legal rules cannot. The “political” mechanisms identified by PV are organized around two concepts: credibility and popularity. Presidents want credibility and popularity, PV argue, and these preferences induce the executive branch to share authorities. Political incentives as a result “at least block the most lurid forms of executive abuse” in ways legal constraints cannot (p 5).149 In this Part, I argue that neither credibility nor popularity mechanisms can


149 The Executive Unbound does not define the term “abuse.” In correspondence, Professor Posner informs me that the term applies to deviations from the median voter’s preference rather than violations of constitutional rights as defined by text and legal precedent.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 809


generate stable effects on executive behavior standing on their own. I focus here not solely on the question whether an executive under political constraints will diverge from median popular preferences, but also on whether it will violate deeply held deontological values, such as those embodied in generally recognized constitutional entitlements under the Bill of Rights. Considering the effect of political bonds upon both genres of “abuse” suggests that the political mechanisms limned by PV work best (or only) when they interact with legal limits on executive authority. The possibility of such complementary interactions will be taken up at greater length in Part IV.


As a threshold matter, it is worth noting that the account of political constraint is more loosely sketched in The Executive Unbound than the finely drawn critique of law’s efficacy. The former occupies one chapter, whereas the discussion of legal constraints occupies several chapters. To some extent then, the arguments in this Part are an effort to take seriously objections to PV’s arguments while at the same time sketching in a more substantial way the political mechanisms on which the asserted law/politics dichotomy rests. At the same time, my approach differs from PV’s in one important regard. In my judgment, their analysis is marked by an unwarranted evaluative dualism. Whereas The Executive Unbound dons the actuary’s green eyeshade in examining law, it reaches for a rose-tinted monocle to deal with politics. To compensate and to approximate better a broadly acceptable political economy of executive constraint, I have found it especially important to press skeptically upon PV’s posited political mechanisms in order to ensure the analysis involves a comparison of like with like.


My focus in this Part is PV’s descriptive and positive claims. My responsive arguments are also intended to be understood as positive and not normative in nature. But it is again worth noting that the frailty of political mechanisms developed here also may undermine strong normative implications of the executive dominance thesis. That is, if freestanding political constraints on the executive are weak, there is less reason to think that the policy selections of a president unfettered by law will necessarily be optimal. And indeed, the analysis of this Part suggests that the presidency will not be reined in by either a desire for credibility or a need for popularity. That result, and its troublesome normative implications, set the stage for the normative analysis at the close of Part IV.


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


810 The University of Chicago Law Review [79:777


A. Credibility


The notion of a “political” constraint on executive discretion is ambiguous.150 It is possible to imagine both ex ante and ex post effects. For example, the circumstances of national politics could tie executive policy choices to congressional preferences ex ante if the presidential selection system throws up only candidates closely tied to their legislative party’s agenda. This might have been the case had the House of Representatives had a larger routine role in presidential selection.151 Alternatively, national politics might impose ex post constraints on executive discretion. Politicians seeking reelection thus respond to median voters or to interest groups capable of influencing median voters.152


The Executive Unbound emphasizes ex post political checks on executive discretion. Its first credibility-related mechanism rests on the idea that Presidents will be “restrained” when doing so maximizes attainment of such first-order preferences through reelection.153 Constraint thus emerges as a byproduct of the agency relationship between executives and the electorate. Presidents are distrusted by voters. To secure the electorate’s trust, they voluntarily enter into forms of “self-binding.” Because this imposes higher costs on ill-motivated actors, well-motivated executives can use such actions to signal their virtuousness to the electorate (p 137). By the alchemy of signaling theory—a model developed first to understand employment markets154—self-serving motivations are transub-stantiated into bonds against the misuse of executive authority.


This signaling-based mechanism is contestable. I sketch four criticisms here. First, PV’s model assumes there are two sorts of Presidents. Some are well-motivated and “choos[e] the policies that


150 In a trivial sense, all “political” constraints are also legal constraints insofar as national politics is nested in a larger enabling framework of regulations, statutes, and constitutional rules. PV acknowledge as much (pp 4–5). Furthermore, as developed in Part II, there may be instrumental reasons for hewing to what the law commands. I mean to set aside those possibilities in this Part and examine the possibility of freestanding constraints on executive power emerging through political contestation alone.


151 See note 136 and accompanying text.


152 See David R. Mayhew, Congress: The Electoral Connection 13 (Yale 1974) (developing the thesis of electoral responsiveness).


153 The credibility argument is largely an argument about the President, rather than an argument about the executive branch. Hence, endowing agencies with independence is a way of enhancing credibility (p 142). This leaves unanswered the question of what then constrains an independent agency with broad discretion, such as the Federal Reserve during the financial crisis (pp 58–59) (suggesting, tentatively, that “the executive is checked, in some sense, by its own internal divisions”).


154 See Michael Spence, Job Market Signaling, 87 Q J Econ 355, 361–68 (1973) (developing the model).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


2012] Binding the Executive 811


voters would choose if they knew what the executive knows” (p 130). Others will not. If voters are capable of reliably selecting well-motivated executives, the latter will not need to distinguish themselves from ill-motivated executives. A premise of the signaling model is therefore not merely heterogeneity in presidential type but also electors’ inability to select good types reliably. This is surely right to some extent. For every Lincoln there is a Nixon. The model predicts that only well-motivated Presidents will engage in self-binding. If ill-motivated Presidents could perfectly mimic such behavior, self-restraint would no longer have a sorting effect. Credibility effects, therefore, are predicted most often for well-motivated Presidents and least often for ill-motivated Presidents (p 152).


155 If the public selects an ill-motivated President, she has no incentive to exercise constraint. Rather, she should extract maximal rents during the first term in the Oval Office in the expectation of being turfed out four years later.156 Thus, precisely when it is most needed, credibility does the least work. When it is least needed, it is expected to have the greatest effect. This hardly seems ideal.


Second, in a world in which neither law nor Congress constrains the executive, it is not clear why Presidents even need credibility. Self-binding for the sake of building credibility makes sense only if there is something gained by having credibility. Thus, the proposed constraining mechanism implies an exchange: the President gives up some control over one domain of policy and thereby gains something else of value. The “something else” is assumed to be correlated to expected electoral advantage. But why should this be so? It is not clear why a gesture demonstrating good intentions without generating positive policy outcomes would generate increased electoral support. All this is quite aside from the fact that electoral mechanisms simply do not apply to second-term Presidents.


On PV’s account, credibility is valuable instead because both first- and second-term occupants of the Oval Office “need public support . . . to prod Congress to provide the President with funds for his programs and statutory authorization when necessary” (p 132).157 But this explanation is puzzling. Recall that a premise of The Executive Unbound is that “major constraints on the executive . . . do


155 PV elaborate on how signaling sorts well-motivated and ill-motivated Presidents when they note that “where the benefits really do exceed the costs, the well-motivated executive will employ the mechanisms, whereas the ill-motivated executive is less likely to do so” (p 152).


156 Or, to conjure Nixon’s shade again, by fixing the second-term presidential election.


157 Emphasis added. Elsewhere, the argument lapses into ambiguous generalities, such as the claim that credibility is desirable because it induces “voters and legislators . . . to confer authority” on the President (p 151).


06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM


812 The University of Chicago Law Review [79:777


not arise from law or from the separation-of-powers framework” (p 4).


158 The idea that Presidents need credibility in order to secure new statutory authorities is in tension with the claim that the scope of available legal authority is irrelevant to executive discretion. Stated more generally, Presidents’ desire for credibility is explicable only if there is some domain of action in which the executive does not exercise broad control as a consequence of effective legal or constitutional bindings. The strength of credibility-related mechanisms is a positive function of the size of that policy domain.159 The looser legal constraints are, the less downstream value credibility has—because the President has less need for new grants of discretion. In short, the need for political credibility makes sense only if there are nonpolitical (i.e., legal or institutional) constraints on the presidency—yet these have been characterized as irrelevant.



Equally puzzling is the assertion in The Executive Unbound that presidential “powers cannot effectively be exercised if a sufficiently large supermajority of the public believes that the President lies or has nefarious motives” (p 153).160 Again, on the terms of PV’s own model, it is not clear why this would be so. Congress is assumed to be too fragmented and clumsy to oppose Presidents. And the public cannot directly recall a President in the fashion that some state governors can be recalled.161 So why a President would believe she needs persistent supermajorities (let alone majorities) to govern in the short-term is not immediately clear.162 Unless opposition takes an

158 Emphasis added.

159 Indeed, PV offer a series of five examples in which Presidents appeared to believe they needed legislative action for a policy to be successful (pp 124–28).

160 PV also say that “the unchecked president is generally too weak to adopt abusive policies” (p 153). Credibility might be relevant to overcoming political opposition, but if so, we need to know why legally unbound executives care about political opposition that takes neither institutional nor legal form.

161 See, for example, Cal Const Art II, § 15.

162 An obvious example of the difficulties of public opposition bereft of institutional channels is the public opposition to the 2003 Iraq War, which took the form of tens of thousands in the streets but could not muster any meaningful political change. Empirical data on public trust here does not help the argument. Marc Hetherington finds that “decreasing trust leads to substantially more negative evaluations of both the incumbent president and Congress.” Marc J. Hetherington, The Political Relevance of Political Trust, 92 Am Polit Sci Rev 791, 791 (1998). But PV do not explain why the “warmer feelings” and “larger store of support,” id at 803, generated by trust matter given their model of broad executive discretion (p 122–23). Further, Hetherington’s data suggests that declining political trust affects attitudes toward both the President and Congress. See id at 795. Declining trust also induces lower levels of participation more generally, and increased support for outsider candidates. See Marc J. Hetherington, The Effect of Political Trust on the Presidential Vote, 1968–96, 93 Am Polit Sci Rev 311, 311, 321 (1999). Declining participation may loosen political bonds on the President.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 813

institutionalized or legalized form, effective public action by latent majorities or supermajorities is relatively costly and rare.

163

In short, a theory of executive constraint based on credibility mechanisms needs supplementing with a separate account of some other kind of constraint. Realistically, it needs an account of legal and institutional constraints. While accounting many of PV’s arguments about the law, I have endeavored to redeem such an account from their criticism in Part II. But the point to note here is that in order to accept the credibility mechanism posited by PV one must engage in some serious surgery to the threshold claim in The Executive Unbound that law and legal institutions impose no “major constraints” on the executive. The credibility mechanism requires an account of how the executive is constrained by law.

The third objection to the credibility mechanism trains on its connection to the goal of minimizing costly executive behavior. On PV’s model, self-binding has a temporal dimension.164 It requires that Presidents make a trade, exchanging some “control over policy choices” now for “more discretion in the future” (p 151). At T1 the executive gives a little; at T2 it gets a little. But why should such temporal trade-offs limit the scope for executive “abuses,” understood either as variations from median voter preferences or violations of deontological floors? The effect on either sort of “abuse” will depend on whether the risk of abuse is greater at T1 or at T2. To see this, imagine that a President relinquishes control over, say, questions pertaining to the diplomatic recognition of other nations today, and thereby gains credibility with the electorate.165 The President then cashes out this “credibility” at T2 in decisions to send citizens alleged to be terrorists back to countries where they are vulnerable to torture. The basic point here is simple: A practice of intertemporal exchanges of policy discretion may result in more, as well as less, “abuse.” Indeed, to the extent the executive has a large range of policies within its control, it can pick and choose between

163 Large public protests in southern European countries against austerity measures are an interesting counterexample. At the time of this writing, it remains to be seen whether those measures are at all successful. See, for example, Rachel Donadio and Niki Kitsantonis, Thousands in Greece Protest New Austerity Bill, NY Times A10 (Oct 20, 2011); Rachel Donadio, Italians Strike to Protest Berlusconi’s Austerity Plan, NY Times A4 (Sept 7, 2011). For a general account of the circumstances in which latent public majorities can be effective, see R. Douglas Arnold, The Logic of Congressional Action (Yale 1990).

164 This is often seen as key to definitions of trust. See Russell Hardin, Trust 18 (Polity 2006) (“The most natural and commonplace representation of the incentive structure of a trust relationship between individuals is as an exchange that cannot be consummated on the spot but in which the second party is required to fulfill after the first.”).

165 See Zivotofsky v Clinton, 132 S Ct 1421, 1425 (2011).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

814 The University of Chicago Law Review [79:777

policies on which such control will be diluted. There is no reason to expect it will generally give up control of policies for which the risk of abuse is especially grave.

The intertemporal dimension of PV’s credibility mechanism also makes imposing demands on voters’ cognitive capacities. It is realistic to imagine that the President makes not only one, but several policy decisions at T1, and then several more policy decisions at T2. A credibility mechanism allows the executive to obtain slack tomorrow by giving up control over some of those policy choices today in exchange for greater leeway over some policy choices tomorrow. Voters hence make a comparison between portfolios of policy choices at two points in time accounting for the possibility of discretion being traded between past and future. Voters must presumably keep a running tally of how much control executives have foregone and how much extra discretion they have claimed over time. This seems to credit the electorate with greater cognitive capacity than seems plausible.

Finally, the credibility mechanism is insensitive to the fact that many derogations of constitutional rights are carried out by the executive with the support of large majorities, while many advances in fundamental rights are a product of executive initiative alone. A desire for credibility, standing alone, does not extinguish the executive’s willingness to accede to the undesirable wishes of the majority while increasing its incentive to expand human liberties. To make this point more explicit, notice that many policies now universally condemned on libertarian or humanitarian grounds had broad political support in their day. This was true, for instance, of the removal of the Cherokee from Georgia during President Jackson’s term in office.166 It was true of the internment of Japanese Americans during World War II.167 Both times Congress approved the President’s actions. The same public support might, on close examination, also be found for many post-9/11 measures that have resulted in significant liberty deprivations of persons unconnected with terrorism. By contrast, consider unilateral presidential initiatives that led to improved race-related policies in times of war.168 Both Presidents Truman and Eisenhower also resisted “witch-

166 Lucas A. Powe Jr, The Supreme Court and the American Elite, 1789–2008 77–80 (Harvard 2009) (describing congressional approval of Jackson’s Cherokee policy).

167 Geoffrey R. Stone, Perilous Times: Free Speech in Wartime from the Sedition Act of 1798 to the War on Terrorism 296 (Norton 2004) (noting in relation to the internment that “public opinion played a key role in the thinking of both the military and the president”).

168 See Mary L. Dudziak, Law, War, and the History of Time, 98 Cal L Rev 1669, 1690 n 122 (2010) (collecting sources).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 815

hunting” anti-Communist probes led by Cold War legislators, albeit with mixed success.

169 It is far from clear that these now lauded executive initiatives had popular backing at the time. Abuses may thus be a result either of executive initiatives or of presidential acquiescence. To reduce the net level of abuses, it is necessary to find mechanisms that apply asymmetrically to harmful presidential initiatives and to executive branch resistance against popular pressure for rights curtailment. Ex ante promulgation of entrenched individual rights has this asymmetrical structure. The technologies of self-restraint on which PV rely do not.

One possible response to this concern would contend that policies adopted with majority support cannot, by definition, be abusive.170 I am not convinced this is a desirable approach to the analysis of governance practices. There is no reason to exclude categorically from normative concern harmful policies animated by the force of majority approval. To the contrary, these have long been seen as a central problem for liberal theorists and constitutional designers.171 There is also a thread of argument in The Executive Unbound to the effect that worries about executive abuse are overblown, especially around post-9/11 counterterrorism.172 By implication only weak mechanisms of constraint are desirable. The threshold empirical premise is, to say the least, highly controversial. It rests on contested premises about the magnitude of terrorism threats, the most effective means of minimizing such threats, and the scale of historical erroneous rights deprivations.

To summarize, The Executive Unbound proposes political bonds as substitutes for legal constraints on executive power. Its first proposed political mechanism, however, is ineffective when most needed and bites hardest when least desired. It also requires a domain of legal constraint that is said not to exist. Even if that condition is met, it is not clear how the quest for credibility can induce less variance from the median voter’s preferences or fewer rights violations. The credibility argument is thus an insubstantial limit on executive power standing on its own.

169 See Stone, Perilous Times at 337–38, 385–86 (cited in note 167).

170 PV characterize Presidents as well-motivated if they “choos[e] the policies that voters would choose if they knew what the executive knows” (p 130).

171 See, for example, John Stuart Mill, On Liberty, in Alan Ryan, ed, On Liberty and The Subjection of Women 7, 10–11 (Penguin 2006).

172 For example, PV suggest that the most important risk is that the executive will not be trusted in times of crisis, and they descriptively claim that abuse has been infrequent (pp 122–24, 204).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

816 The University of Chicago Law Review [79:777

B. Popularity

The credibility mechanism is entangled in The Executive Unbound with a second and separate control mechanism founded on public opinion. There is, PV argue, something called “the court of public opinion” wherein executive claims are tried and either vindicated or found wanting (p 61). Appealing to that tribunal of electoral sentiment, Presidents need “public support” for their actions (p 61). PV identify elections as key popularity-based mechanisms of constraint. They also properly caution that elections “are far from perfect,” because they are only periodic, are hampered by voter ignorance, and fail to account for the possibility of majoritarian tyranny (p 115). But, PV imply, they are the best mechanism available and cannot meaningfully be supplemented with laws or institutions. To be sure, they concede, elections impose no direct constraint on second-term Presidents who are barred by the Twenty-Second Amendment from standing again for the office of chief executive. But for second-term Presidents “a concern with the judgment of history pushes [them] to make the decisions that future generations . . . will approve” (p 131).

Unfortunately, this lightly sketched argument from popularity raises more questions than it resolves. The popularity mechanism has a poorly specified subject (the “public”) and a causal foundation that is a touch opaque. At times, the argument seems to boil down to a kind of economic determinism. PV might be read to say that the mere existence of “an educated and leisured population, and the regular cycle of elections” disciplines the executive (p 204). Read literally, this suggests GDP growth somehow cashes out mechanically as a more disciplined executive branch. This seems an insufficient explanation for decade-by-decade fluctuations in civil liberties in the United States over the last half-century. It also fails to single out specific mechanisms through which economic growth affects improved liberty, or to show how mechanisms in fact operate in a given national context, or whether they are muffled by other dynamics.173

173 To be sure, there is some relationship between levels of gross economic development and political structure: above a certain level of development, it appears that democratic transitions stabilize and do not revert to autocracies. See Adam Przeworksi, Democracy and Economic Development, in Edward D. Mansfield and Richard Sisson, eds, The Evolution of Political Knowledge: Democracy, Autonomy, and Conflict in Comparative and International Politics 300, 308 (Ohio State 2004) (“While the paths to democracy are varied, the survival of democratic regimes depends on a few easily identifiable factors. Foremost among them is the level of development, as measured by per capita income.”). But the effect works in the other direction, too: dictatorships are also stabilized by economic development and thus persist once

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 817

There are three more granular objections to the popularity mechanism. Call these the objections from collective action costs, information costs, and political psychology. In each case, there is an impediment to “public opinion” standing alone playing any meaningful role in limiting executive choice. In the first two cases, importantly, the obstacles to effective political action are mitigated through the use of laws and legal institutions. Fresh review of PV’s popularity mechanism thus hints at some kind of complementary interaction between legal and political mechanisms—a theme elaborated further in Part IV.

Consider first the question of collective action costs. The Executive Unbound repeatedly states the executive is “helpless” in the face of public opposition (p 153). But if the speed and agility of the presidency compare favorably to the torpor of legislative action, then it is hard to see how the public could pose a threat to executive dominance. A diffuse and heterogeneous public acting alone will predictably find it very costly to coordinate on a judgment about the executive’s actions and to respond in a way that expresses disapproval.174 The Executive Unbound does not elaborate on the precise mechanisms through which “public opinion” overcomes the costs of collective action (except for elections). Of special note, it does not address the simple point that legal rules and government institutions, such as Congress and the courts, may play a role in channeling and organizing public action in a way that mitigates collective action costs.

Even casual observation suggests that unilateral executive responses to crises do not betray the guiding hand of public judgment. Consider the “ad hoc” Treasury and Federal Reserve responses to the financial crisis, from the deal to save Bear Stearns to the bailout of Citibank and other financial institutions that Paulson sought in September 2008.175 To the extent the public had a view, it seemed to be opposed to taxpayer-funded bailouts176—and it was

a certain level of democracy is reached. See id at 303. Even if this evidence suggests the United States is not about to collapse into Stalinist terror, it says little about more micro fluctuations in respect for democratic process or fundamental rights. Further, other studies emphasize “the threat of revolution by the citizens” and the presence of “inter-group inequality” rather than gross economic development as inducements toward democratization. See Daron Acemoglu and James A. Robinson, Economic Origins of Dictatorship and Democracy 29, 35–37 (Cambridge 2006).

174 See Mancur Olson Jr, The Logic of Collective Action: Public Goods and the Theory of Groups 2 (Harvard 1965).

175 See Skeel, The New Financial Deal at 12–13 (cited in note 17).

176 Jeanne Sahadi, Low Marks for Paulson, Bailout, CNNMoney (Oct 22, 2008), online at http://money.cnn.com/2008/10/21/news/economy/paulson_poll/index.htm (visited Nov 16, 2011).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

818 The University of Chicago Law Review [79:777

roundly ignored. To the contrary, the government “bent over backward to make the [initial bailouts] attractive for the banks, [first] charging below-market interest and eschewing any significant ownership” and then, to avoid public sanction, adopting “more complex,” less transparent subsidies.

177 Or consider PV’s observation that one effect of increased critical attention to Guantلnamo detentions was a shift in detention operations to Bagram, Afghanistan, which is almost completely off the public radar (p 37). This too might be fairly interpreted as a strategy of avoiding public opprobrium rather than responding to the public’s concerns.178

The second objection to the publicity mechanism concerns information. One argument that PV tender against legislative control is Congress’s want of information (p 26).179 A reliance on public sanctions implies that members of the public are better informed, in the aggregate, than members of the legislature about executive policy choices. Yet it requires something of a leap of faith to posit that Main Street does better than Capitol Hill on this score. Decades of political-science research finds scant evidence that the public is even minimally informed about American politics, let alone about specific national issues.180 Even if the public is in the aggregate informed, that knowledge is likely to be highly diffused and therefore extremely costly to collect, analyze, and act upon in the absence of an aggregation mechanism. Technological advances, rather than increasing the capacity for public oversight, may have diminished the public’s ability to act. Empirical studies suggest that voters face serious difficulties in processing the increasingly large glut of information now available through the growing range of electronic media.181 This is not to say that public pressure will necessarily be ineffectual. It is rather to suggest that prioritizing the

177 Johnson and Kwak, 13 Bankers at 168 (cited in note 1).

178 For a model of the President’s asymmetric ability to influence public opinion in emergencies, see Robert M. Entman, Projections of Power: Framing News, Public Opinion, and U.S. Foreign Policy 9–13 (Chicago 2004). Entman suggests this presidential authority is at its lowest when the events in question “neither readily fit nor obviously clash with habitual, well-established mental associations.” Id at 18.

179 PV observe that “[i]n many cases, Congress lacks the information necessary to monitor discretionary policy choices” (p 26).

180 The canonical study is Angus Campbell, et al, The American Voter 543 (John Wiley & Sons 1960) (finding that “[a] substantial portion of the public . . . is almost completely unable to judge the rationality of government actions”). That finding has been duplicated in later studies. See Michael X. Delli Carpini and Scott Keeter, What Americans Know about Politics and Why It Matters 62 (Yale 1996); Eric R.A.N. Smith, The Unchanging American Voter 190 (California 1989).

181 See Doris A. Graber, Processing Politics: Learning from Television in the Internet Age 46 (Chicago 2001).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 819

diffuse public as a check on executive discretion over and above the role played by institutional actors, such as legislators and judges, triggers a substantial burden of explanation. Rather than struggling with that burden, it may be more realistic to look to laws and political institutions as coordinating mechanisms in the production of public opinion.

182

To be sure, PV recognize the steep information asymmetry between the executive and the public. They offer two remedies to bridge the gap. First, they argue that “whistleblowers can easily find an audience” for revelations of abuse (p 209). This claim reflects an unwarranted confidence. Currently, “whistleblowers are vulnerable to criminal prosecution, adverse personnel actions, and in the case of government lawyers, professional ruin.”183 In the national security domain, journalists report that the flow of information has slowed under the putatively more transparent Obama administration,184 while prosecutorial efforts and onerous pretrial treatment of whistleblowers have become respectively more vigorous and more severe.185 Under these conditions, there is no reason to believe that criminal penalties and other negative consequences do not in fact deter most whistleblowers, leaving the public largely in the dark about many executive branch initiatives. Ironically then, legal rules of the kind PV identify as inefficacious in cabining executive discretion may be all too effective in torpedoing the publicity mechanism on which they elsewhere rely for political constraint.

Second, PV contend that the “terrific competition” for the Oval Office will conduce to disclosure of presidential misdeeds (p 119). It

182 Consistent with this view, Professor Jack Goldsmith has argued that “courts, members of Congress and their staff, human rights activists, journalists and their collaborators, and lawyers and watchdogs” together “constitute a vibrant presidential synopticon [i.e., device for watching the government].” Goldsmith, Power and Constraint at 207 (cited in note 21). Notice that most of the instruments on which Goldsmith relies in turn depend on legal rules, such as laws that force information disclosure, or institutional status, to be effective.

183 Mika C. Morse, Note, Honor or Betrayal? The Ethics of Government Lawyer–Whistleblowers, 23 Georgetown J L Ethics 421, 421 (2010). In the national security context, whistleblowers could be prosecuted under the broad Espionage Act of 1917, Act of June 15, 1917, 40 Stat 217, codified as amended in various sections of Titles 22 and 50. Consider Geoffrey R. Stone, Top Secret: When Our Government Keeps Us in the Dark 10–14 (Rowman & Littlefield 2007) (arguing for the constitutional protection of some acts of disclosure but acknowledging the existence of doctrinal uncertainty).

184 Dana Priest and William M. Arkin, Top Secret America: The Rise of the New American Security State xxi (Little, Brown 2011).

185 For an account of one recent whistleblower prosecution, see Jane Mayer, The Secret Sharer: Is Thomas Drake an Enemy of the State?, New Yorker 46 (May 23, 2011). For an account of one suspected leaker’s treatment, see Elisabeth Bumiller, Pentagon to Move Suspect in Leaks, NY Times A12 (Apr 19, 2011).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

820 The University of Chicago Law Review [79:777

is hard to see why. A threshold problem is that the period in which such competition arises—presidential electoral campaigns—is a period in which “few voters change their minds.”

186 Even those who say they enter the campaign period with an open mind “end up voting in a way that could readily have been predicted before the campaign on the basis of their religious affiliation, socioeconomic status, or occupation.”187 The claim that political opponents reliably “disclose negative information about those in office” relevant to abusive conduct is also in tension with the empirical evidence that finds no public learning occurs during election campaigns (p 119). Moreover, political opponents are subject to the same informational limitations as Congress and the public, constraints that are difficult and costly to overcome. As a result, political opponents may be better off not embarking on costly and risky searches for abuse and instead resorting to more familiar negative campaigning strategies of character assassination and distorting voting records.188 Nor, it should be emphasized, will electoral competition induce some candidates to make credible commitments that effectively bind their hands while in office. Such “precommitment politics” are rare, even if they would otherwise be desirable.189

Information costs impede the mooted popularity mechanism in another way: voters face barriers to signaling effectively disapproval

186 See James A. Gardner, What Are Campaigns For? The Role of Persuasion in Electoral Law and Politics 86–87 (Oxford 2009). See also Thomas M. Holbrook, Do Campaigns Matter? 157 (SAGE 1996) (arguing that the function of election campaigns is to “move public opinion toward the expected outcome” based on the ex ante equilibrium of voters’ preferences). Consistent with these findings, other empirical studies suggest that an increasing segment of the American public is becoming ideologically polarized along racial, religious, and economic class lines in ways that are resilient to short-term influences. See Alan I. Abramowitz, The Disappearing Center: Engaged Citizens, Polarization, and American Democracy 62–83 (Yale 2010) (presenting data on race and religion); id at 130–38 (presenting data on increasing ideological polarization over time); Jacob S. Hacker and Paul Pierson, Winner-Take-All Politics: How Washington Made the Rich Richer—And Turned Its Back on the Middle Class 146–49 (Simon & Schuster 2010) (“Evangelicals notwithstanding, economic issues divide the parties more sharply along class lines than in the past.”).

187 Gardner, What Are Campaigns For? at 87 (cited in note 186).

188 The shift of campaign funding from candidates to so-called SuperPACs largely postdates the publication of The Executive Unbound, and its authors cannot be blamed for failing to predict that sea-change in electioneering. Whether the shift will lead to more substantive or merely nastier campaign seasons remains to be seen at the time of this writing. I am not optimistic.

189 See Saul Levmore, Precommitment Politics, 82 Va L Rev 567, 569 (1996) (“One explanation for the failure of political candidates to take the leap from familiar campaign promises to more serious precommitments is that it is not easy to precommit.”). Levmore sees both a “positive puzzle” and a “normative disappointment” in the observed absence of precommitment politics. Id at 570. If that positive puzzle with respect to Presidents were to dissolve, then PV’s argument would stand on different ground.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 821

of discrete policies. PV argue that electoral competition is a main way to convey disapproval of the executive (pp 119–20). But “representative elections result in aggregation or bundling across the entire range of issues or potential issues.”

190 They yield muddy signals.191 This is especially so in presidential elections, where the bundle of issues at stake in any given election is larger and more complex than in a federal legislative or state contests. Voters appear to be respond to the bundling problem in presidential elections by focusing consistently on a single factor. Public approval of Presidents is correlated to economic trends, and the related presence or absence of economic “disturbance” in “the private lives of millions of Americans.”192 Voters focus on macroeconomic health as a proxy for presidential success despite “the vast gap between expectations and the [President’s] capacity to deliver” in economic matters.193 By contrast, there is little empirical reason to think that public opinion will be primarily concerned with “abuses” of government power, however the term is defined. To the extent the latter considerations are salient, they may be washed out at the polls by the force of macroeconomic circumstances over which Presidents have little control.194 By contrast, the legislators and judges that PV spurn have

190 Sherman J. Clark, A Populist Critique of Direct Democracy, 112 Harv L Rev 434, 469 (1998). See also Jacob E. Gersen, Unbundled Powers, 96 Va L Rev 301, 324–28 (2010) (discussing the effects of the bundling of government functions on accountability).

191 For example, they do not account for intensity of preferences. See Clayton P. Gillette, Plebiscites, Participation, and Collective Action in Local Government Law, 86 Mich L Rev 930, 968 (1988) (arguing that direct forms of democracy are superior in this regard because turnout functions as a proxy for intensity).

192 Neustadt, Presidential Power and the Modern Presidents at 81 (cited in note 62) (observing that the approval ratings of President Truman and President Eisenhower declined when government action or inaction negatively affected the lives of many Americans). Gardner, What Are Campaigns For? at 86–87 (cited in note 186) (reporting studies that “repeatedly” find that “influences unrelated to the actual campaigns—the health of the national economy, for example, or recent job growth” influence outcomes in elections across the board).

193 Forrest McDonald, The American Presidency: An Intellectual History 466 (Kansas 1994) (describing the restrictions the President faces when dealing with the economy because, despite interest rate and money supply levers, the President “cannot raise or lower taxes, cannot hire or fire people, [and] cannot reward productivity or punish inefficiency”).

194 One might add to this the plethora of cognitive pitfalls and distortions voters face. Professor Posner has insightfully explored when and how cost-benefit analysis should deal with those distortions. See Matthew D. Adler and Eric A. Posner, Implementing Cost-Benefit Analysis When Preferences Are Distorted, in Matthew D. Adler and Eric A. Posner, eds, Cost-Benefit Analysis: Legal, Economic, and Philosophical Perspectives 269, 306–07 (Chicago 2001) (listing “treatment of disinterested preferences,” “preferences distorted by lack of information,” “adaption,” and “objective badness” as some of the problems faced in cost-benefit analysis, and tentatively arguing that uninformed preferences should be considered “at least to some extent”).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

822 The University of Chicago Law Review [79:777

better information, a greater number of opportunities to act on its basis, and more fine-grained tools, all of which enable them to express and then to act on disapproval of specific executive policies. They are also unlikely to focus on macroeconomic outcomes alone.

The bundling problem can also preclude effective electoral condemnation of the misuse of emergency powers. Consider in this regard the contrasting roles that President Obama’s handling of Guantلnamo and his economic policies will likely play in the 2012 presidential race. This point can be made especially vivid with a transnational comparison to a situation in which a formal emergency was followed proximately by polls. The 1975–77 Emergency in India was characterized by “widespread detention [and] horrific acts of torture” of the ruling Congress Party’s political opponents.195 What dominated the subsequent election campaign, however, was Congress’s use of involuntary sterilization.196 That particularly emotionally charged issue drowned out public consideration of other, arguably more serious, forms of abuse. (The Congress Party responsible for the torture, killings, and forced sterilizations was returned to power in short order when its political opposition fell apart,197 showing how exogenous political dynamics can blunt political checks.) Even when the extensive use of emergency powers is promptly set before voters, the exercise of electoral choice over decision makers with complex, manifold portfolios means that voting might generate no meaningful signal about “public opinion” respecting abuses.198

195 Aziz Z. Huq, Uncertain Law in Uncertain Times: Emergency Powers and Lessons from South Asia, 13 Constellations 89, 100 (2006) (querying how to determine what measures, exactly, had caused such disapproval by the electorate).

196 See id at 101 (noting that the 1977 election hinged on only the question of sterilization).

197 See id at 94–95.

198 For similar reasons, I am also skeptical of the claim made by PV that concerns about their future reputation will check second-term Presidents (p 131). Even apart from that problem, the aftermath of the Bush administration hardly supports the view that senior officials who made serious policy errors will suffer for it. To take one perhaps controversial example, it is far from clear that any senior official has acknowledged the gravity of misjudgments about the Iraq war or the harm that flowed from those errors. Rather, senior officials’ memoirs do their best to sweep aside responsibility for past mistakes. See, for example, Michiko Kakutani, Leaving Regrets to Others, Cheney Speaks, NY Times C1 (Aug 26, 2011):

Mr. Cheney writes that “the liberation of Iraq” was “one of the most significant accomplishments of George Bush’s presidency”—never mind the failure to find the weapons of mass destruction that were cited as a chief reason for the invasion, or a botched occupation that allowed an insurgency to metastasize for years.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 823

There is a third and final criticism to be lodged against the popularity argument. Unlike the criticisms developed above, it is more narrowly tailored to PV’s argument: The Executive Unbound is structurally self-defeating because an acceptance of its claims about the optimal distribution of political authority would corrode the conditions of political constraint that make those premises feasible.

This argument needs some unpacking. Political constraints, argue PV, work only “[a]s long as the public informs itself and maintains a skeptical attitude toward . . . government officials” (p 209). That is, both the arguments about credibility and popularity have an embedded assumption that the public is willing to examine critically administration policies. An uninformed or credulous public, by contrast, provides no check.

But another core claim of The Executive Unbound is that a “skeptical attitude” toward the executive is unwarranted, and even irrational. PV repeatedly claims that the executive is better situated to make optimal policy decisions than other branches—indeed, this claim of comparative institutional advantage is an integral part of its critique of law (pp 107–08).199 As a result, PV stress that “[t]he risk that the public and legislators will fail to trust a well-motivated president is just as serious as the risk that they will trust an ill-motivated president” (p 123). They express skepticism about how serious the abuse of executive power has been in the US.200 The Executive Unbound’s analysis of “tyrannophobia” also poses a challenge to their premise of public skepticism. PV argue that skepticism of the executive does not deter even “low-level executive abuses,” and “it is equally plausible that overheated rhetoric limits beneficial grants of power to the executive” (p 203).

Put these latter strands together, and you get a powerful normative case for political quietism in the face of an “extensive” and “power[ful]” presidency (p 207). Rather than counseling public supervision of the executive, PV suggest it is positively irrational to “maintain[ ] a skeptical attitude toward . . . government officials”

199 The claim is also made in PV’s earlier work. See Eric A. Posner and Adrian Vermeule, Terror in the Balance: Security, Liberty, and the Courts 29–38 (Oxford 2007) (making the argument in the national security context); Adrian Vermeule, Judging under Uncertainty: An Institutional Theory of Legal Interpretation 205–23 (Harvard 2006) (making the argument in the context of judicial review of agency action).

200 PV acknowledge that there have been “[l]ow-level executive abuses” in recent memory (p 203). This seems to me an understatement.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

824 The University of Chicago Law Review [79:777

(p 209).

201 The public should relax, cease to worry about an unchecked executive, and accept whatever bailouts, black sites, and boondoggles flow from unfettered executive discretion.202 This conclusion is, to say the least, in some tension with the skepticism PV view as necessary for political checks to work.203

In summary, attention to the “social-scientific microfoundations” of the proposed publicity and credibility mechanisms of executive constraint elicits more skepticism than confidence (p 123). Diffuse publics face large and oft preclusive transaction costs. To the extent the executive does respond to public opinion, it can leverage its asymmetrical access to information to obscure objectionable policies rather than ending them. Neither political opposition nor elections does much to cure such evasion. Political contests, unlike the adversarial trial, are not reliable vehicles for surfacing factual truths. Without legal institutions to structure opposition, laws to force policy-related disclosures, and individual rights to protect dissent, it is not clear popularity can do the work assigned it by The Executive Unbound.

C. Summary and Normative Implications

This Part has developed several grounds for skepticism of The Executive Unbound’s positive account of political constraints. Those

201 Consistent with that dictum, PV have in other work expressed great confidence in the bona fides of government action. See, for example, Posner and Vermeule, Terror in the Balance at 21, 26–28 (cited in note 199).

202 Consider PV’s argument that it is “pointless” to bewail the decline of legal constraints (p 209).

203 It is worth noting one further objection that, at least in my view, should not be given much weight. This is the standard objection from public choice theory against any inference of motive or intentionality of groups such as the “public” or a “supereducated elite whose members . . . care about . . . civil liberties” (p 202). As Kenneth Arrow famously demonstrated, nontyrannical decision rules for collective action are pervasively susceptible to instability. Unconstrained employment of a majority-vote rule, for example, often leaves a collective unable to settle on a stable choice. See Kenneth A. Shepsle, Analyzing Politics: Rationality, Behavior, and Institutions 53–84 (2d ed 2010). Stability in collective decision making is induced only by adopting mechanisms that limit agendas, narrow choices, or prorogue debate. See Kenneth A. Shepsle and Barry R. Weingast, Structure-Induced Equilibrium and Legislative Choice, 37 Pub Choice 503, 507 (1981) (arguing that “institutional restrictions on the domain of exchange induce stability, not legislative exchange per se”) (emphasis omitted). In large democracies, this means the results of collective choice are importantly a function of the basic rules of the democratic game—that is, what the public “thinks” is a function of what the law allows them to “think.” There is some truth to this, but it is easily overstated. PV could fairly reply that often the “public” is confronted with binary policy choices—should we invade Iraq? Bail out Lehman Brothers? In such cases, it is quite plausible to think the public has a clear majority view immune to cycling. And nothing in their argument is inconsistent with the role statutory and constitutional law can be observed to play in structuring electoral choice.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 825

arguments, however, also bear on normative claims made on behalf of broad executive discretion. If public efforts to constrain the President by demanding credibility or through the exercise of “public opinion” tend to fail, Presidents do not face serious political sanctions for straying from the public’s wishes. And if legal-institutional bonds on executive discretion must be set aside as too fragile, the President has only weak reasons to conform actions to the democratic principal’s preferences.

Ex ante, it is hard to see why a polity would agree to create a plebiscitary office with such destabilizing potential. Likely political opponents of the presidency, for example, would rationally fear the entrenching effect of White House tenure. Claims that “the presidency already has intrinsic interest for the public” will sound hollow to the public it is meant to reassure (p 204). As the polity develops, it is hard to see how a plebiscitary office subject to only weak democratic controls will generate desirable policy outcomes. However expert the officeholder is—and I bracket here any reasons for skepticism of executive competence—the institution lacks incentives to conform to democratic wishes, and to avoid either rent-seeking or abusive policies. Under these conditions, it seems more likely that public and oppositional political elites would seek ways to tie down Presidents by law, if not by politics. That is, an executive “bound” somewhat by law may not only be observable, but also inevitable in practice.

IV. LAW AND POLITICS AS COMPLEMENTS

To recap briefly, my argument to this point has been largely skeptical of both the treatment of “law” and “politics” as parts of the strong law/politics dichotomy undergirding The Executive Unbound. First, I have argued that while many of The Executive Unbound’s criticisms of the liberal legalist positions find their mark—such that there remains something of a “positive puzzle”204 of how mere law has effective force—legal and institutional constraints still appear to play a nontrivial role in the processes whereby the executive is checked. Second, I have developed reasons for thinking that the political mechanisms that PV invoke as chaining presidential power are, at least standing on their own, more akin to loose links than durable chains.

This Part advances a preliminary synthesis of the strong law/politics dichotomy with the objections aired in Parts II and III. It

204 Levinson, 124 Harv L Rev at 662 (cited in 142).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

826 The University of Chicago Law Review [79:7777).

[أحمد أبو زنط]

presents in very general form a different political economy of executive constraint as an alternative to either the “law alone” view of liberal legalists or the “politics alone” position of the New Crits. My general thesis is that, while standing alone, legal and political mechanisms each yield only fragile constraints on government, when they work together they can prove effective. Legal chains are weak, as PV stress, in part because they are not self-executing. They require political will to prevail. Political forces, I have argued, are weak due inter alia to transaction and information costs. But these impediments can be mitigated through law. Standing alone, each of the two mechanisms (legal and political) may well fail. But if they can be harnessed in tandem, they may yield a greater likelihood of effective constraint on executive discretion because they mitigate each other’s weaknesses. Legal and political mechanisms thus should not be conceptualized as substitutes (p 15).

205 Executive constraint instead is coproduced by the operation of both legal and political forces working as complements.206

A caveat is in order before this thesis can be pressed. My aim is to reorient debate about executive power toward a new set of inquiries rather than to define with precision a fixed domain of mechanisms that produce checks on executive discretion. Mapping comprehensively the intersections of legal and political constraining mechanisms would be an enormous undertaking. It would entail (among other things) an account of the normative status of law in American culture, a detailed investigation of the diverse enforcement mechanisms Congress has over the years invented to rein in the executive, and a cartography of the diverse private interests that stand ready in any given policy domain to press for executive constraint. I cannot cover all these matters here. As a result, the discussion that follows will properly be criticized as illustrative, rather than exhaustive. Nevertheless, by supplying numerous examples of how political and legal mechanisms interlock to yield observable cases of executive constraint—and, as a corollary, by suggesting that in the absence of such complementary action no constraint emerges—this Part develops an agenda for further, more systematic research.

205 PV note in their introduction that “the substitution of political for legal constraints” is a core claim of the book (p 15).

206 To be clear, my claim is not that all politics are conveyed through legal channels such as elections and legally defined offices. Nor is my claim that legal constraints are epiphenomenal of underlying politics. Rather, absent legal tools, political forces would not constrain.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 827

To see the utility of reorienting attention to the interaction of legal and political mechanisms, notice how the respective frailties and strengths of legal and political mechanisms of executive constraint play out. Objections that legal rules are mere parchment barriers lose much of their force when a political coalition stands behind that law or constitutional principle. The Madisonian architecture of separated powers in which each branch defends its own prerogatives may be fragile, but congressional coalitions can and do nonetheless push to enforce specific legal constraints for electoral or normative reasons. That political coalitions should marshal their capital behind specific laws is not surprising: statutes and other rules often embody, in a temporally durable form, the political commitments of a given political coalition. Statutes, moreover, can be used as focal points for organizing by that coalition to enable resistance against the disruptive initiatives of new Presidents and emergent factions.

The interaction of law with political commitments can be observed in granular form within the executive. American political culture has developed in a path-dependent fashion that invests legal rules with normative significance. With notable outliers on both sides of the aisle, most officials express sincere normative preferences for legality and constitutionalism.207 Legality has its own standing constituency. That lobby is cultivated by law schools and by law professors who profess to teach what the Constitution and federal statutes command. This has consequences for the breadth of presidential discretion because lawyers trained in this tradition currently staff the length and breadth of the executive branch. These lawyers’ preferences for legality impose frictions on executive initiatives that run afoul of legal limits. At the limit, these lawyers are likely aware of the leverage they have by dint of the fact that their noisy exit would inflict damage on the government’s policy projects. As one scholar of executive branch legal interpretation has recently observed, good-faith lawyering by attorneys within the Justice Department and other components of the executive will constrain presidential options so long as “the press, the lawyerly public, and Congress [ ] care about preserving the traditional structures of executive-branch legal interpretation.”208

207 See, for example, Goldsmith, The Terror Presidency at 158–60 (cited in note 121) (describing the decision to withdraw a 2002 Office of Legal Counsel opinion that authorized a variety of coercive forms of interrogation).

208 Trevor W. Morrison, Libya, “Hostilities,” the Office of Legal Counsel, and the Process of Executive Branch Legal Interpretation, 124 Harv L Rev F 62, 74 (2011). But see Eric Posner

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

828 The University of Chicago Law Review [79:777

Outside government, legal rules and legally constituted institutions also furnish something of a solution to some of the transaction cost and information cost problems for groups within the general public wishing to challenge government policy decisions.209 Laws and legal institutions reduce the coordination and informational costs for the public associated with identifying undesirable acts by the government in three ways. First, legislators and judges, as holders of institutional power and as generally recognized experts in matters of legality, are vested with publicly recognized authority to draw attention and to seek remediation.210 Compared to ordinary citizens, they also have greater access to media platforms. They can thus better disseminate information, priming the electorate to attend to issues and thereby lowering the costs of collective action.211 Second, legislators and judges have credible tools to threaten the executive ranging from politically damaging criticism to enactment of new rules further binding the President.212 Finally, specific legal rules can mitigate informational and coordination problems beyond their function as focal points. The First Amendment, most obviously, ensures that information in the public sphere may be transmitted freely; prior restraints against national media are de facto unknown today. In the national security domain, remedies such as habeas corpus have served as informational platforms by revealing details of executive policies that contradict official narratives.213 Ex ante legal prohibitions also make public coordination easier because violation of legal rules is evidence that an executive act is undesirable, harmful, or unjustified. Whatever justifications the executive offers, the public knows that a

and Adrian Vermeule, Libyan Legal Limbo: Why There’s Nothing Wrong with Obama Ignoring Some of His Own Legal Advisers on Libya, Slate (July 5, 2011), online at http://www.slate.com/articles/news_and_politics/jurisprudence/2011/06/libyan_legal_limbo.html (visited Nov 11, 2011).

209 The idea that constitutional rules can serve as focal points for political mobilization was developed by David A. Strauss, Common Law Constitutional Interpretation, 63 U Chi L Rev 877, 910–11 (1996). For experimental evidence that focal points can enable coordinated action absent any prospect of sanctions or rewards, see Richard H. McAdams and Janice Nadler, Coordinating in the Shadow of the Law: Two Contextualized Tests of the Focal Point Theory of Legal Compliance, 42 L & Socy Rev 865 (2008).

210 See Daniel Sutter, Enforcing Constitutional Constraints, 8 Const Pol Econ 139, 142 (1997) (identifying the need for sincere experts to lower information costs in the public enforcement of constitutional rules).

211 Id at 145 (identifying free-rider problems in enforcement of constitutional rules).

212 Id (recognizing cost of enforcing constitutional rules).

213 See Aziz Z. Huq, What Good Is Habeas?, 26 Const Commen 385, 422–24 (2010). For an analysis of the potential epistemic effects of jurisdictional design in the national security context, see Aziz Z. Huq, Forum Choice for Terrorism Suspects, 61 Duke L J 1415, 1484–87 (2012).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 829

previous political coalition already weighed the costs and benefits of a policy and found a categorical bar appropriate.

Observed from fifty-thousand feet, therefore, legal and political mechanisms seem to correct for each other’s shortfalls within Congress, inside the executive, and across the larger public sphere. Putting these observations together, it is plausible to posit that it will be a combination of contemporaneous political support and the availability of a plausible constraining reading of the law that in tandem has the best chance of narrowing the executive’s range of discretionary choices.214 There is not a dichotomy between law and politics but instead a symbiotic relationship. All things being equal, the executive is most likely to be constrained when both legal and political mechanisms are in play. And it is least likely to be constrained when neither obtains.215

At the same time, legal mechanisms and political forces take many forms. On the one hand, constitutional norms, statutory prohibitions (whether enforceable in court or not), and regulatory demands each supply a different sort of resource. On the other hand, many forms of political action may be salient to executive constraint. These range from citizens’ diffuse orientation toward the law and a general absence of legal cynicism to the existence of a robust legislative opposition willing to challenge executive power and a flourishing of private actors in the media and civil society devoted to making public abuses of state authority. Accordingly, it seems to me that a focus on a “reelection constraint” is unnecessarily narrow (p 13). There is no reason, in thinking about mechanisms of

214 I should be clear that my argument is not one of “bilateralism,” which is the claim that a court will endorse an emergency action when it has the backing of both the executive and the legislature. See Samuel Issacharoff and Richard H. Pildes, Between Civil Libertarianism and Executive Unilateralism: An Institutional Process Approach to Rights during Wartime, 5 Theoretical Inquiries L 1, 5 (2004). I have argued elsewhere that this claim is empirically unfounded and normatively ambiguous. See Aziz Z. Huq, Against National Security Exceptionalism, 2009 S Ct Rev 225, 254–56. The claim advanced here is distinct because it does not concern what Congress has done in the past, but what political support exists at the moment a law is tested.

215 In a recent article, Professor Daryl Levinson advances the similar claim that “rights and votes also appear to function as complements.” Daryl J. Levinson, Rights and Votes, 121 Yale L J 1286, 1350 (2012). Levinson’s argument focuses on the ways in which individuals’ possession of a legal right can help obtain political participation, or vice versa. Id at 1350–55. This seems right, but also distinct from the claim advanced here. My focus is broader than mere voters, and the independent variable with which I am concerned is constraint on government power—a factor not in play directly in Levinson’s analysis.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

830 The University of Chicago Law Review [79:777

executive constraint, to limit the focus to actions taken with an eye to directly influencing electoral outcomes.

216

There are, as a result of this heterogeneity, many ways in which politics and law can interact. Further, my use of the term “complements” does not imply some orderly or predictable coproduction of constraint. Rather, joint legal-political limits on executive discretion are forged or shattered based on transient and highly variable political and institutional conditions. Far more work would be required to develop a predictive theory of executive constraint with a domain that extended any way across the variegated political history of the United States.217 My aim here is more modest. It is to provide a set of illustrations of the complementary action of interacting legal and political mechanisms in contradistinction to the single-strand proposals of the liberal legalists and the New Crits.

To that end, consider the following matrix. It presents in simplified form four possible combinations of legal and political constraints. The table is simplified because the existence of a legal or a political constraint is not a categorical variable. Both legal and political constraints are matters of degree. They might depend, among other things, on the subject matter at issue, the substantive right or interest at stake, and the nature and extent of political mobilizations. But Table 1 conveys the gist of the problem.

216 By way of example, consider the efforts of interest groups to raise awareness of issues, ranging from the dignitary interests of torture victims or unborn fetuses to the hazards of global warming or childhood vaccines. Such interest groups may sometimes care about electoral politics, but their reach and ambitions are often far more varied and wide ranging. Or consider the effects of legal scholars who write for widely read non-specialist publications or even blogs about issues of federal government policy. I count these activities as political even if they have no tight connection to electoral outcomes.

217 Hence, my argument is not reducible to H.L.A. Hart’s observation that the efficacy of law rests on a social rule of recognition. See text accompanying note 117. Hartian social acceptance is necessary, but not sufficient, for the legal-political coproduction of executive constraint.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 831

TABLE 1. POSSIBLE COMBINATIONS OF LEGAL AND POLITICAL CONSTRAINTS ON EXECUTIVE DISCRETION

No Legal Constraints

Legal Constraints

No Political Constraints

Air Force responds to 9/11

War Powers resolutions

Political Constraints

Cheney Energy Taskforce

Closing Guantلnamo

Take each quadrant in turn. In the top-left quadrant, when law does not forbid a discretionary executive action and insubstantial or nugatory political forces oppose it, the President’s freedom to act is at its peak. The decision to shoot down planes suspected of being hijacked on September 11 is an obvious example of this situation. Conversely, when both law and political forces array against the executive (in the bottom-right quadrant), discretion is most likely to be choked off. Most recently, law and political forces have aligned against executive initiatives respecting transfers from the Guantلnamo base.218 They also impeded Treasury Secretary Paulson’s projected bailout of financial institutions, and the unilateral issuance of debt by President Obama to end the debt standoff.219 In all these cases, political opponents of a presidential action had credible legal arguments against the President. In each case, government officials effectively conceded the validity and force of normative preferences in favor of legality. And in each case, the legality of given policy options was pivotal to public debate.

The intermediate cases—the top-right and bottom-left quadrants—are ones in which either legal or political mechanisms, but not both, stand in the way of a presidential policy choice. Here, my prediction is not that legal constraints alone or political constraints alone never bite on presidential authority. Where a dispute has low stakes, it surely may be that the cost of disregarding a law or imperiling a political relationship may outweigh the benefit of a particular policy action. But constraint is most likely to pinch if either the legal or political landscape changes. When a constituency exists for constraint, these situations will be unstable. Outcomes may depend on whether it is the President or the President’s opponents who manage to shift matters to a “law plus politics” or a “neither law

218 See note 21.

219 See notes 4, 22, and accompanying text.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

832 The University of Chicago Law Review [79:777

nor politics” equilibrium. Hence, in cases where there is political but not legal support for a constraint, the President’s opponents may add a legal constraint by working to have new legislation enacted—moving the situation into the bottom-right quadrant. The ban on Guantلnamo transfers is a case in point. Alternatively, where ex ante legal constraints exist, political entrepreneurs may use them as a focal point to increase the political opposition to presidential action. If the President’s opponents fail, the situation collapses into the top-left quadrant.

Two examples in Table 1 illustrate some of these possibilities. A first example concerns the National Energy Policy Development Group established under Vice President Richard Cheney’s auspices in 2001 to “develo[p] . . . a national energy policy designed to help the private sector.”220 The closed-door task force generated significant congressional opposition. But legislators were unable to force disclosure of its deliberations.221 In 2002, the congressionally controlled General Accounting Office filed an unusual civil action against the Vice President seeking disclosures about the task force. The suit was dismissed on justiciability grounds.222 A subsequent suit by private parties seeking disclosures also failed.223 Hence, political opposition alone proved insufficient to change executive—in part because of an absence of legal support. The White House prevailed not simply because its political opposition proved too frail but because those opponents were unable to establish a platform in the law to sustain their opposition. Had the lawsuits come out the other way, it is possible to imagine the political tide may have turned further against the White House.

The second example is one in which a legal rule existed but political will to enforce it did not. The 1973 War Powers Resolution224 imposes temporal limits on the executive’s use of military force without authorization from Congress.225 As PV observe, it has not been viewed as a success. Exemplary violations include the 1999

220 Cheney v United States District Court for the District of Columbia, 542 US 367, 373 (2004).

221 See Jeff Gerth, Accounting Office Demands Energy Task Force Records, NY Times A20 (July 19, 2001).

222 See Walker v Cheney, 230 F Supp 2d 51, 65–70 (DDC 2002) (holding that the comptroller general had not had a sufficient institutional interest to merit standing). The decision was not appealed.

223 See In re Cheney, 406 F3d 723, 730 (DC Cir 2005) (en banc) (holding that statutory disclosure requirements under the Federal Advisory Committee Act did not extend to the energy taskforce).

224 Pub L No 93-148, 87 Stat 555, codified as amended at 50 USC §§ 1541–48.

225 See War Powers Resolution § 5(b), 87 Stat at 556, codified at 50 USC § 1544(b).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 833

Kosovo intervention and the 2011 Libya air campaign (p 86).

226 In the latter case, conflicting House votes disapproving of the military deployment and refusing to order its end signal the absence of consensus even within the lower house.227 That is, the presence of a legal norm has not been sufficient to induce coordinated political action. It is worth noting that in some cases the White House has decided to comply with the Resolution.228 Accounts of war powers often omit this history of compliance and thus may be missing evidence of a more subtle interaction between law and politics in the absence of outright interbranch confrontations.

Rather than choosing between legal and political mechanisms, in short, a positive account of executive constraint can profitably focus on their interaction. On this view, law and politics may have something more than a strictly additive relationship. By filling each other’s lacunae, legal and political mechanisms may yield more than the mathematical sum of their stand-alone effects. But they also may not. The political economy of executive constraint thus demands an investigation of the conditions under which political coalitions can successfully invoke legal tools to secure limits on executive discretion.

In concluding, I turn from the agenda for positive political analysis to some normative implications of this alternative framing. Specifically, if the executive is indeed best constrained by the coordinated effect of law and politics, and such coordinated action is a contingent and occasional phenomenon, troubling normative questions arise. There are two grounds for concern.

First, the capacity of our political system to generate meaningful checks on the national executive may be waning. National legislative politics in the United States are characterized by growing polarization between the two main parties. Legislative caucuses for both parties are more ideologically coherent than they were a generation ago; ideological overlap between the parties has

226 See Abraham D. Sofaer, The War Powers Resolution and Kosovo, 34 Loyola LA L Rev 71, 71–79 (2000) (describing the debate over the Kosovo intervention); Charlie Savage, 2 Top Lawyers Lose Argument on War Power, NY Times A1 (June 18, 2011) (describing internal debates over the Libyan air campaign). See also Jacob S. Hacker and Oona A. Hathaway, Our Unbalanced Democracy, NY Times A21 (Aug 1, 2011) (repeating the longstanding criticism that Congress is “unwilling or unable to defend its constitutional and statutory power to authorize a war” with respect to Libyan operations).

227 See Richard F. Grimmett, War Powers Resolution: Presidential Compliance 1–2 (Congressional Research Service June 24, 2011), online at http://www.fas.org/sgp/crs/natsec /RL33532.pdf (visited Nov 18, 2011).

228 See id at 13–18.

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

834 The University of Chicago Law Review [79:777

vanished.

229 The combination of ideologically homogenous and distinct parties with bicameralism and presentment predictably generates legislative gridlock.230 Either one side or the other can be relied upon to leverage vetogates in the legislative process. This affects not only fiscal matters—as the 2011 debt ceiling debate illustrated—but also impedes the possibility of effective political checks on the executive. In the absence of a major partisan realignment, the conditions for neither significant political nor legal constraint may be met.231 More subtly, deepening ideological commitments may render voters less receptive to new information. Revelation of executive abuse—whether it is the exploitation of Chrysler’s creditors or Guantلnamo detainees—may consequently be less likely to influence a chief executive’s credibility or popularity. It may not be law’s weakness but the short-term deliquescence of American politics that drives changes to the scope of executive power in coming decades.

Second, the exercise of executive discretion, especially in emergencies, may have troubling distributional consequences. If politics plays a role in shaping executive constraint, it is to be expected that the distribution of losses among social groups will be a function of those groups’ influence in national politics. In the current structure of national politics, “affluent people have considerable clout, while the preferences of people in the bottom third of the income distribution have no apparent impact on the behavior of their elected officials.”232 Information about national politics is similarly unevenly distributed within the electorate.233 As a result, “policy outcomes strongly reflect the preferences of the most

229 See Abramowitz, The Disappearing Center at 11–13 (cited in note 186) (summarizing data); Delia Baldassarri and Andrew Gelman, Partisans without Constraint: Political Polarization and Trends in American Public Opinion, 114 Am J Sociology 408, 442 (2008) (describing how “[p]arty voters . . . are likely to convey more extreme preferences to their party leaders,” leading to party polarization).

230 PV blame gridlock on the separation of powers (p 197). This seems only partly right. It is the interaction between bicameralism and the current party system that conduces to a particularly high likelihood of stalemate.

231 A major partisan realignment, to be sure, is hardly impossible. On the frequency of major party realignments in the United States, see John H. Aldrich, Why Parties? A Second Look 299–304 (Chicago 2011).

232 Larry M. Bartels, Unequal Democracy: The Political Economy of the New Gilded Age 285 (Princeton 2008). See also Hacker and Pierson, Winner-Take-All Politics at 41–72 (cited in note 186).

233 See Carpini and Keeter, What Americans Know about Politics at 176–77 (cited in note 180) (finding that political ignorance “is most likely to be found among those who arguably have the most to gain from effective political participation: women, blacks, the poor, and the young”).

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

2012] Binding the Executive 835

affluent but bear virtually no relationship to the preferences of poor or middle-income Americans.”

234 In moments of crisis, when the politically ill-connected have the least opportunity to compensate for their impoverished leverage through time-consuming political mobilization, the effects of asymmetrical influence may be especially pronounced. Emergencies often impose sudden, large losses on the polity though their immediate impact (for example, Hurricane Katrina) or indirectly due to a costly government response (for example, 9/11). In a world in which political mechanisms bear heavily on executive responses, losses will be allocated disproportionately to those with the least political influence, exacerbating their ex ante disenfranchisement.235

Returning to that fraught Thursday in September 2008, consider again the Troubled Asset Relief Program236 (TARP) created by Congress in response to Treasury Secretary Paulson’s pleas.237 TARP provides a useful platform for analysis here because its implementation was largely a matter of executive discretion rather than legislative direction. Although losses on TARP were not as great as feared,238 the primary beneficiaries of the intervention were “creditors and counterparties” of financial institutions.239 By contrast, elements of TARP meant to serve broad swaths of the public, such as its mortgage restructuring program, were later deemed at best a “qualified” success.240 TARP also aggravated the “too big to fail” problem, leaving the financial industry more concentrated and more confident of its implicit taxpayer-funded subsidy.241 The risk of a

234 Martin Gilens, Inequality and Democratic Responsiveness, 69 Pub Op Q 778, 778 (2005). See also id at 790 (presenting graphical representations of the differential influence of the top and bottom deciles of the population by income).

235 Consider Johnson and Kwak, 13 Bankers at 175 (cited in note 1) (noting that responses to the financial crisis have socialized losses while keeping gains private).

236 Emergency Economic Stabilization Act of 2008, Pub L No 110-343, 122 Stat 3765, codified as amended in various sections of Title 12.

237 See Jonathan G. Katz, Who Benefited from the Bailout?, 95 Minn L Rev 1568, 1580–84 (2011) (describing several of TARP’s elements).

238 See id at 1588 (estimating that of the $700 billion appropriated by Congress, less than $400 billion was spent, and more than $200 billion was repaid).

239 See id at 1612.

240 See id at 1585 (citation omitted). See also Johnson and Kwak, 13 Bankers at 179 (cited in note 1) (noting that President Obama’s Homeowner Affordability and Stability Plan, which would have allowed judicial reductions in mortgages, was “kill[ed] . . . in the Senate, with the banks and their industry associations refusing even to negotiate”).

241 See Peter Boone and Simon Johnson, The Next Financial Crisis: It’s Coming—and We Just Made It Worse, New Republic 24, 26 (Sept 23, 2009) (“If [major banks] . . . get into serious financial trouble, the Fed can be counted on to lend them essentially unlimited amounts at effectively zero interest rates.”); Johnson and Kwak, 13 Bankers at 208–13 (cited in note 1)

06 HUQ BKR (DO NOT DELETE) 5/25/2012 12:50 PM

836 The University of Chicago Law Review [79:777

future crisis, which will likely be remedied from the public purse, is thus greater at the time of this writing than it was in early 2008.

The political foundations of executive constraint, in short, may be fraying as greater legislative polarization and political inequality create political dynamics in which political elites are increasingly unlikely to converge on opposition to a President. This in turn will change both the quantity of constraint and the downstream distributive effects of executive initiatives.

CONCLUSION

The general tenor of The Executive Unbound is optimistic. Increased presidential authority is a boon. When it is not, there are political checks in reserve. The book’s normative implication is quietist. Effort to constrain the executive is pointless and unnecessary. But the account developed in this Review suggests more cause for pessimism about both the current state of affairs and future prospects for national governance. Just as liberal legalists hoped for too much from the law, so PV’s aspirations for politics may be disappointed. Rather than emphasizing law alone or politics alone, I have argued that the executive is constrained most by complementary action of legal and political mechanisms. My primary goal has been to posit an alternative description of executive restraint and to suggest that much more work needs to be done specifying the precise mechanics whereby legal and political mechanisms interact to produce checks under diverse background institutional and political circumstances.

In closing I have also suggested that my alternative account of executive checks has normative implications. This account at a minimum implies that the current optimality of executive action cannot be assumed. And it hints that talk of an “unbounded” executive may be increasingly descriptively correct because national politics is becoming more polarized and more elite-driven. The burden of this emerging political economy of executive power, I suspect, will fall hardest on those who can least afford to bear its costs. Inevitable or not, this to me hardly seems cause for celebration.

(arguing that a breakup of large financial institutions is the only solution to the problem of moral hazard); Skeel, The New Financial Deal at 11 (cited in note 1