Constitutional Stickiness

Constitutional Stickiness Ozan O. Varol∗ Continuity is a striking hallmark of the constitutional world. Empirical evidence shows that many constitutional amendment and replacement processes counterintuitively produce relatively little change in substance. During constitutional makeovers, existing provisions frequently “stick,” even where they are arbitrary, suboptimal, or anachronistic. This paradox, which I call “constitutional stickiness,” has been neglected in the scholarship. American constitutional theorists have largely assumed that Article V’s high threshold for amendment is the primary culprit for lack of formal constitutional change and that significant alterations might follow with a lower threshold. With mounting calls by the states for a constitutional convention, this assumption has also prompted concerns about a “runaway convention” that could drastically alter the substance of the U.S. Constitution. This Article challenges that assumption. Drawing on rational-choice theory and behavioral law and economics, it provides the first theoretical analysis of constitutional stickiness in descriptive and normative terms. Even with low amendment thresholds, the constitutional status quo exerts significant historical weight and the constitutional starting points constrain future choices in specific and systemic ways. The existing constitutional configurations therefore often depend, quite arbitrarily, on the historical starting point, rather than a rational assessment of all alternatives. As a result, relatively insignificant events in a country’s early ∗ Copyright © 2016 Ozan O. Varol. Associate Professor of Law, Lewis & Clark Law School. For helpful comments, I thank Erin Delaney, Rosalind Dixon, David Fontana, Tom Ginsburg, Vicki Jackson, Richard Kay, Jeremy Kidd, David Landau, David Law, Russell Miller, Christopher Schmidt, Mark Tushnet, Kathy Varol, Mila Versteeg, and the participants at workshops and conferences held at the Clough Center for the Study of Constitutional Democracy at Boston College Law School, the Center for the Constitution at James Madison’s Montpelier (the Montpelier Roundtable in Comparative Constitutional Law), as well as participants in the 2014 SEALS Annual Meeting. For outstanding research assistance, I thank Kelsey Benedick, Katherine Davis, Elisabeth Rennick, Philip Thoennes, and Lynn Williams. I am also grateful to the Editors of the UC Davis Law Review (particularly Gokalp Gurer) who provided expert editorial assistance. 899 900 University of California, Davis [Vol. 49:899 constitutional history can have an enormous impact, whereas more dramatic events that occur later — such as a revolution or a major constitutional convention — are much less consequential than assumed. Ultimately, the Article aims to reorient the normative focus of constitutional scholarship to oft-neglected temporal and sequential considerations. TABLE OF CONTENTS INTRODUCTION ................................................................................... 901 I. A THEORY OF CONSTITUTIONAL STICKINESS ............................. 909 II. THE CAUSES OF CONSTITUTIONAL STICKINESS.......................... 915 A. Rational Choice ................................................................. 915 1. Benefits of Constitutional Change ............................. 917 2. Costs of Constitutional Change ................................. 925 a. Costs of Drafting ................................................... 925 b. Costs of Negotiation and Ratification ..................... 928 c. Implementation Costs ............................................ 931 B. Behavioral Law and Economics ......................................... 937 1. Status Quo Bias ........................................................... 938 2. Anchoring Bias ........................................................... 947 3. Availability Heuristic .................................................. 948 4. Hedonic Adaptation ................................................... 951 5. Excessive Veneration of the Constitution .................. 952 III. NORMATIVE IMPLICATIONS OF CONSTITUTIONAL STICKINESS ... 954 CONCLUSION....................................................................................... 960 2016] Constitutional Stickiness 901 INTRODUCTION American constitutional theory suffers from a normative bias. Scholars have been preoccupied with analyzing the normative implications of various constitutional provisions at particular moments of interest and how to alter them for the better.1 Implicit in these normative explorations is an assumption of a world where constitution-makers can produce meaningful changes, correct errors in the original design, and attain certain ideals — as long as they are unencumbered by a high bar for constitutional amendment. This assumption has produced a prevailing view that Article V’s high threshold for constitutional amendment is the primary culprit for lack of formal constitutional change and that significant alterations might follow with a lower threshold.2 With mounting calls by the states for a 1 See David S. Law, Constitutions, in THE OXFORD HANDBOOK OF EMPIRICAL LEGAL RESEARCH 376, 376 (Peter Cane & Herbert M. Kritzer eds., 2010) [hereinafter Constitutions] (noting that “constitutional scholarship has long been predominantly normative in character”); Richard H. Fallon, Jr., Constitutional Constraints, 97 CALIF. L. REV. 975, 977 (2009) (noting the normative orientation of constitutional theorists); John Ferejohn, The Politics of Imperfection: The Amendment of Constitutions, 22 LAW & SOC. INQUIRY 501, 526 (1997) (noting that normative scholarship can “create false visions of constitutional possibilities”); Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of Constitutional Commitment, 124 HARV. L. REV. 657, 659 (2011) (“Constitutional lawyers and theorists have . . . focus[ed] instead on normative issues . . . .”). 2 See Bruce Ackerman, The Emergency Constitution, 113 YALE L.J. 1029, 1031, 1077 (2004) (noting that Article V represents “a formidable obstacle course” to amendment); John Ferejohn & Lawrence Sager, Commitment and Constitutionalism, 81 TEX. L. REV. 1929, 1954 (2003) (“[Article V’s] resistance to change is what cements the external, institutional commitments of the Constitution in place.”); William E. Forbath, The Politics of Constitutional Design: Obduracy and Amendability — A Comment on Ferejohn and Sager, 81 TEX. L. REV. 1965, 1979-80 (2003) (“The overbearing obduracy of the unamended Article V diminished the jurisgenerative politics of the New Deal moment, and deprived us of new constitutional texts — texts on which citizens and courts could have rested claims to extend and deepen our constitutional commitments in ways we would prize.”); Aziz Z. Huq, The Negotiated Structural Constitution, 114 COLUM. L. REV. 1595, 1665 (2014) (noting that political actors are unable to amend the U.S. Constitution due to “exorbitant transaction costs” stemming from Article V); Michael J. Klarman, What’s So Great About Constitutionalism?, 93 NW. U. L. REV. 145, 149 (1998) (“The supermajority requirements of Article V unduly privilege the status quo . . . .”); Sanford Levinson, How the United States Constitution Contributes to the Democratic Deficit in America, 55 DRAKE L. REV. 859, 874 (2007) (arguing that Article V’s requirements make constitutional amendment “almost impossible by the difficulties placed in its path”); David A. Strauss, The Irrelevance of Constitutional Amendments, 114 HARV. L. REV. 1457, 1459 (2001) (“It is only a slight exaggeration to say that [informal amendments] are the only means of change we have.”). 902 University of California, Davis [Vol. 49:899 constitutional convention,3 this assumption has also prompted concerns about a “runaway convention” that could drastically alter the substance of the U.S. Constitution.4 This conventional wisdom is appealing in principle. After all, constitution-makers have the authority to rewrite the constitution or some of its provisions from scratch and pick from a rich menu of options in global constitutional models.5 Indeed, in cases of constitutional replacement, constitution-makers are specifically charged with replacing the existing constitution with a new one, presumably because the status quo suffers from significant shortcomings. Without a high threshold for constitutional amendment — such as that imposed by Article V — substantial, substantive changes should follow. Yet even a casual survey of constitutional replacement and amendment processes around the globe reveals that many counterintuitively produce relatively little change in constitutional substance.6 That anecdotal observation is confirmed by the empirical 3 On some accounts, Michigan became in March 2014 the thirty-fourth state to request a constitutional convention, which meets the two-thirds majority required by Article V. Stephen Dinan, Balanced Budget Convention Gains Steam as Congressman Calls for Official Evaluation, WASH. TIMES (Apr. 1, 2014), http://www.washingtontimes. com/news/2014/apr/1/balanced-budget-convention-gains-steam-congressman. Although many states, including Michigan, have sought a convention for the limited purpose of proposing a balanced budget amendment, scholars disagree on whether states have the authority to restrict the subject matter of constitutional conventions. Compare Walter E. Dellinger, The Recurring Question of the “Limited” Constitutional Convention, 88 YALE L.J. 1623-24 (1979) (arguing that limited subject matter conventions are not authorized by Article V), with Michael B. Rappaport, The Constitutionality of a Limited Convention: An Originalist Analysis, 28 CONST. COMMENT. 53, 56 (2012) (arguing that states can call for limited or unlimited conventions). 4 See, e.g., Gerald Gunther, The Convention Method of Amending the United States Constitution, 14 GA. L. REV. 1, 4 (1979) (finding unpersuasive the claim that a convention limited to a single narrow subject “won’t get out of hand”); Michael B. Rappaport, Reforming Article V: The Problems Created by the National Convention Amendment Method and How to Fix Them, 96 VA. L. REV. 1509, 1528-31 (2010) (“A runaway convention is not merely a theoretical possibility . . . .”); see also Laurence H. Tribe, Issues Raised by Requesting Congress to Call a Constitutional Convention to Propose a Balanced Budget Amendment, 10 PAC. L.J. 627, 635 (1979) (contending that “[h]owever democratic an Article V Convention might be in theory, such a convention would inevitably pose enormous risks of constitutional dislocation”). 5 See Jon Elster, Constitutionalism in Eastern Europe: An Introduction, 58 U. CHI. L. REV. 447, 468-71 (1991) [hereinafter Constitutionalism]. 6 See, e.g., Arend Lijphart, Democratization and Constitutional Choices in CzechoSlovakia, Hungary and Poland: 1989–1991, 4 J. THEORETICAL POL. 207, 208 (1992) (noting that changes to “fundamental constitutional structure” are rare in established democracies); Adam M. Samaha, Dead Hand Arguments and Constitutional 2016] Constitutional Stickiness 903 evidence, which shows that constitutions exhibit high serial similarity across time.7 On average, an 81% match exists in constitutions preand post-replacement, and a 97% match between a constitution preand post-amendment.8 Even in transitions from one regime type to another or in constitution-making processes following exogenous shocks such as a revolution or war — when one might expect tectonic constitutional shifts9 — the resulting constitutional changes are often relatively minor.10 The recent post-revolutionary constitution-making process in Egypt, for example, produced a constitution that looks remarkably similar to its predecessor.11 In the United States, as Alison Interpretation, 108 COLUM. L. REV. 606, 627 (2008) (noting the persistence of bicameralism in state legislatures in the United States even after Supreme Court decisions weakened the justification for state senates). 7 See ZACHARY ELKINS ET AL., THE ENDURANCE OF NATIONAL CONSTITUTIONS 56-57 (2009); Tom Ginsburg & James Melton, Innovation in Constitutional Rights 6 (draft for presentation at N.Y.U. Workshop on Law, Econ. & Pol., 2013) [hereinafter Innovation], available at http://www.law.nyu.edu/sites/default/files/upload_documents/November% 2019%20Ginsburg%20Melton%20Innovation%20in%20Constitutional%20Rights%20.pdf (“The set of institutions adopted in the very first constitution in a country’s history tends to be very sticky . . . .”); see also Tom Ginsburg & James Melton, Does the Constitutional Amendment Rule Matter at All? Amendment Cultures and the Challenges of Measuring Amendment Difficulty 16 (Coase-Sandor Inst. for Law and Econ., Working Paper No. 682, 2014) [hereinafter Amendment Cultures], available at http://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2432520 (finding that “the vast majority of constitutional amendments result in few [constitutional] changes”). 8 ELKINS ET AL., supra note 7, at 57. As Elkins et al. define those terms, an amendment occurs when constitutional designers follow the amendment rule in the existing constitution and a replacement occurs when the constitutional designers disregard it. Id. at 55. 9 See, e.g., András Sajó, Preferred Generations: A Paradox of Restoration Constitutions, in CONSTITUTIONALISM, IDENTITY, DIFFERENCE, AND LEGITIMACY: THEORETICAL PERSPECTIVES 335, 341 (Michel Rosenfeld ed., 1994) (“After a major revolution, there is neither a serious willingness nor any real possibility to return to the status quo ante . . . .”). 10 See, e.g., Kathleen Thelen, How Institutions Evolve: Insights from Comparative Historical Analysis, in COMPARATIVE HISTORICAL ANALYSIS IN THE SOCIAL SCIENCES 20910 (James Mahoney & Dietrich Rueschemeyer eds., 2003) [hereinafter How Institutions Evolve] (“[F]requently, particular institutional arrangements are incredibly resilient and resistant even in the face of huge historic breaks . . . .”); Thandika Mkandawire, Crisis Management and the Making of “Choiceless Democracies,” in STATE, CONFLICT, AND DEMOCRACY IN AFRICA 119, 125 (Richard Joseph ed., 1999) (“Democratic states that are built on the ruins of authoritarian rule often retain some of the previous state’s institutions, which linger on due to social inertia and structural rigidities.”). 11 See, e.g., John Mukum Mbaku, Providing a Foundation for Wealth Creation and Development in Africa: The Role of the Rule of Law, 38 BROOK. J. INT’L L. 959, 1042-45 (2013) (arguing that the recent revolutions in Egypt and Tunisia produced only “regime changes” and did not significantly alter the governmental structure); Samer 904 University of California, Davis [Vol. 49:899 LaCroix explains, the Constitution of the Confederate States (which seceded from the Union) displayed surprising continuity with the U.S. Constitution, including some of the most contested constitutional provisions of the pre-war period.12 I call this phenomenon “constitutional stickiness.” Drawing on rational-choice theory and behavioral law and economics, this Article offers the first theoretical analysis of how the path of constitutional history constrains future constitutional paths in specific and systemic ways. Studies of path dependence have spanned numerous other fields, such as technological standards,13 the economy,14 political science,15 sociology,16 the common law,17 intellectual-property Atallah, Egypt’s ‘New’ Constitution: Repeating Mistakes, AL JAZEERA (Jan. 14, 2014, 9:33 AM), http://www.aljazeera.com/indepth/opinion/2014/01/egypt-new-constitution-repeatingmistakes-201411255328300488.html (arguing that the 2014 Egyptian Constitution “repackage[s] the old” constitution). 12 Alison L. LaCroix, Continuity in Secession: The Case of the Confederate Constitution, in NULLIFICATION AND SECESSION (Sanford Levinson ed., forthcoming Feb. 2015) (manuscript at 2), available at http://ssrn.com/abstract=2571358 (noting that the Confederate Constitution “duplicated some of the most contested language of the pre-war period, including several clauses that suggested a relatively powerful central . . . level of government with a robustly powered Congress”). 13 See, e.g., Paul A. David, Clio and the Economics of QWERTY, 75 AM. ECON. REV. 332, 332-33 (1985) [hereinafter Clio]. But see S.J. Liebowitz & Stephen E. Margolis, Path Dependence, Lock-in, and History, 11 J.L. ECON. & ORG. 205, 205-06 (1995) (criticizing the application of path dependence to technological standards). 14 See, e.g., W. BRIAN ARTHUR, INCREASING RETURNS AND PATH DEPENDENCE IN THE ECONOMY 33-34 (1994). 15 See, e.g., Scott E. Page, Path Dependence, 1 Q.J. POL. SCI. 87, 88 (2006); Paul Pierson, Increasing Returns, Path Dependence, and the Study of Politics, 94 AM. POL. SCI. REV. 251, 251 (2000) [hereinafter Increasing Returns]; Kathleen Thelen, Historical Institutionalism in Comparative Politics, 2 ANN. REV. POL. SCI. 369, 384 (1999) [hereinafter Historical Institutionalism]. 16 See, e.g., James Mahoney, Path Dependence in Historical Sociology, 29 THEORY & SOC’Y 507, 507-08 (2000). 17 See, e.g., Oona A. Hathaway, Path Dependence in the Law: The Course and Pattern of Legal Change in a Common Law System, 86 IOWA L. REV. 601, 603 (2001); see also Clayton Gillette, Lock-in Effects in Law and Norms, 78 B.U. L. REV. 813, 813-814 (1998); Oliver Wendell Holmes, Jr., The Path of the Law, 10 HARV. L. REV. 457, 457-58 (1897); Matthew Tokson, Judicial Resistance and Legal Change, 82 U. CHI. L. REV. 901, 902-04 (2015). In a famous passage, Holmes explained the path dependency of the common law: The life of the law has not been logic: it has been experience. The felt necessities of the time, the prevalent moral and political theories, intuitions of public policy, avowed or unconscious, even the prejudices which judges share with their fellow-men, have had a good deal more to do than the syllogism in determining the rules by which men should be governed. 2016] Constitutional Stickiness 905 institutions,18 corporate governance and contracting,19 adjudication,20 law and economics,21 and social norms,22 among others. Yet, despite the salience of path dependence in constitutions, the academic commentary has surprisingly failed to examine it in systematic and theoretical terms.23 This Article fills that scholarly void. The study of constitutional stickiness reveals important insights as well as troubling conclusions for contemporary constitutional theory. Constitutional provisions carry significant historical weight, and the starting point often constrains future choices. Even where the initial constitutional choice is suboptimal or anachronistic, that choice has a profound effect on current behavior. As a result, a constitutional provision that has gained an arbitrary historical advantage may be OLIVER WENDELL HOLMES, JR., THE COMMON LAW 1 (1881). 18 See, e.g., Paul A. David, Intellectual Property Institutions and the Panda’s Thumb: Patents, Copyrights, and Trade Secrets in Economic Theory and History, in GLOBAL DIMENSIONS OF INTELLECTUAL PROPERTY RIGHTS IN SCIENCE AND TECHNOLOGY 19, 23, 28 (Mitchel B. Wallerstein et al. eds., 1993). 19 See, e.g., Lucian Arye Bebchuk & Mark J. Roe, A Theory of Path Dependence in Corporate Ownership and Governance, 52 STAN. L. REV. 127, 129-30 (1999); Marcel Kahan & Michael Klausner, Path Dependence in Corporate Contracting: Increasing Returns, Herd Behavior, and Cognitive Biases, 74 WASH. U. L.Q. 347, 348 (1996); Michael Klausner, Corporations, Corporate Law, and Networks of Contracts, 81 VA. L. REV. 757, 762-64 (1995); Amir N. Licht, The Mother of All Path Dependencies: Toward a Cross-Cultural Theory of Corporate Governance Systems, 26 DEL. J. CORP. L. 147, 14950 (2001). 20 See, e.g., Richard H. Fallon, Jr., The “Conservative” Paths of the Rehnquist Court’s Federalism Decisions, 69 U. CHI. L. REV. 429, 434-36 (2002); Lawrence Friedman, Path Dependence and the External Constraints on Independent State Constitutionalism, 115 PENN ST. L. REV. 783, 789 (2011); Michael J. Gerhardt, The Limited Path Dependency of Precedent, 7 U. PA. J. CONST. L. 903, 905-06 (2005); Alain Marciano & Elias L. Khalil, Optimization, Path Dependence and the Law: Can Judges Promote Efficiency?, 32 INT’L REV. L. & ECON. 72, 73 (2012); Richard A. Posner, Past-Dependency, Pragmatism, and Critique of History in Adjudication and Legal Scholarship, 67 U. CHI. L. REV. 573, 583-88 (2000). 21 See, e.g., Mark J. Roe, Chaos and Evolution in Law and Economics, 109 HARV. L. REV. 641, 643-44 (1996). 22 See, e.g., Dan M. Kahan, Gentle Nudges v. Hard Shoves: Solving the Sticky Norms Problem, 67 U. CHI. L. REV. 607, 607-08 (2000). 23 There are a few articles that mention in passing, with no systematic analysis, the possibility of path dependence in constitutional structures. See, e.g., H. Kwasi Prempeh, Presidential Power in Comparative Perspective: The Puzzling Persistence of Imperial Presidency in Post-Authoritarian Africa, 35 HASTINGS CONST. L.Q. 761, 819-20 (2008) (noting that executive-legislative relations in contemporary Africa “reflect strong elements of path dependency”); Samaha, supra note 6, at 625-27 (suggesting that culture and path dependence can serve as sources of constitutional stability); Lawrence B. Solum, Constitutional Possibilities, 83 IND. L.J. 307, 312-14 (2008) (citing the difficulty of replacing constitutions as an example of path dependency). 906 University of California, Davis [Vol. 49:899 retained in future design processes. An alternative constitutional provision may not be adopted even when its benefits are obvious. Conversely, the objective flaws of an existing constitutional provision may not necessarily lead to its amendment or replacement. Constitutional provisions may thus reflect arbitrary historical selection, rather than a comprehensively rational assessment of the costs and benefits presented by alternatives.24 Constitutions may end up where they are because they happened to start there, not because that starting point enjoys a distinct normative advantage. Constitutional stickiness calls into question the prevailing orthodoxy25 in the United States that Article V’s high threshold for constitutional amendment is the primary culprit for lack of formal constitutional change. As a result of the rational-choice and behavioral mechanisms discussed below, constitutional provisions can — and often do — stick even absent significant external restraints on constitution-makers, such as high amendment thresholds. This analysis suggests that any constitutional changes that either a lower amendment threshold or a major constitutional convention may generate would be much less significant than the legal scholarship assumes. Constitutional stickiness can be especially problematic since Constitution 1.0 of most countries is a beta version that can benefit from a comprehensive makeover.26 Even if the designers of Constitution 1.0 happen to stumble upon optimal provisions from the start, over time, those optimal provisions may become anachronistic and suboptimal as society evolves.27 As a result of constitutional stickiness, however, these increasingly obsolete constitutional provisions can become increasingly difficult to alter. After repeated commitments to the existing constitutional path, “the road not taken,” to quote the poem by Robert Frost, grows increasingly distant and difficult to reach.28 Constitutional stickiness also amplifies the salience of earlier events and dampens the significance of later ones.29 Counterintuitively, relatively insignificant events in a country’s early constitutional history can have an enormous impact, whereas more dramatic events that See Gillette, supra note 17, at 817. See Samaha, supra note 6. 26 ELKINS ET AL., supra note 7, at 15. 27 Id. 28 See Paul Pierson, Not Just What, but When: Timing and Sequence in Political Processes, 14 STUD. AM. POL. DEV. 72, 75 (2000) [hereinafter Not Just What]. 29 See id. 24 25 2016] Constitutional Stickiness 907 occur later may be much less consequential.30 It is therefore a mistake to assume that large-scale events — such as the Arab Spring or a major constitutional convention in the United States — will necessarily produce large constitutional waves. But constitutional stickiness is not necessarily undesirable from a normative perspective. In some cases, there will be multiple constitutional paths that produce an optimal result. That a constitutional trajectory has been locked into one path does not necessarily render it inefficient or suboptimal. In addition, constitutional stickiness can generate significant societal benefits, such as promoting constitutional stability, facilitating consensus building during constitutional design, and impeding constitutional change towards a less optimal configuration. But ultimately, as this Article argues, constitutional stickiness can also become a straitjacket that leads to the retention of inferior constitutional configurations where the optimal outcome is their amendment or replacement. In addition, constitutional stickiness is neither inevitable nor irreversible. If it were, the United States would not have experienced significant constitutional transformations as it transitioned from the Articles of Confederation to the U.S. Constitution, nor would South Africa as it adopted a democratic constitution after the end of apartheid. Although constitutional stickiness is a strong force to contend with, it is sometimes possible to reverse the initial constitutional path, go back, and take an alternative path. Moreover, even where the initial path cannot be completely reversed, changes might still occur, but in a bounded fashion constrained by the path initially forged. My primary aim in this Article is to explain stickiness where it occurs, not to argue that constitutions inevitably exhibit that phenomenon. I also consider the conditions under which constitutions are likely to be less sticky and analyze the relative stickiness of different constitutional alterations. This analysis sheds light on the sources of potential disruptions that can foment largescale constitutional changes.31 This Article’s focus is on the stickiness of the provisions in the written, formal constitution (i.e., the “large-c” constitution).32 To be See id. See id. at 78 (discussing large scale institutional development in general political processes through “self-reinforcing processes” that have “distinct trajectories” because of different historical events at their inception). 32 See Law, Constitutions, supra note 1, at 377 (contrasting the large-c constitution to the small-c constitution, the latter of which refers to the “de facto, unwritten, uncodified, or informal constitutions”). 30 31 908 University of California, Davis [Vol. 49:899 sure, constitutional provisions can also be transformed informally through interpretation by relevant judicial, political, and social actors, without any formal changes to the constitutional text.33 For example, the authorities of the federal government in the United States have expanded significantly since the adoption of the U.S. Constitution with little corresponding change to the constitutional text. Likewise, the right to privacy, which is not expressly recognized in the constitutional text, was adopted through interpretation by the U.S. Supreme Court.34 These informal changes can obviate the need for formal constitutional amendments, reducing the significance of textual stickiness. Despite the significance of informal alterations, the formal constitution is undoubtedly important. Setting aside normative debates about judicial philosophy and the appropriateness of “updating” constitutions through interpretation, not all constitutional provisions are amenable to updating through informal means. Constitutional drafters may want to adopt provisions that require specificity — for example, the establishment of an electoral system. In a world where constitutions are becoming increasingly more detailed and specific, it is not possible to informally update many provisions absent brazen disregard for the constitutional language.35 Even with respect to less specific constitutional provisions, the text provides the starting point for interpretation and either constrains or influences the available interpretive paths.36 The Article proceeds in three parts. Part I sets out a basic theory and definition for constitutional stickiness. Part II turns to its causes and, drawing on both rational-choice theory and behavioral law and economics, explains the confluence of factors that can produce 33 See GABRIEL L. NEGRETTO, MAKING CONSTITUTIONS: PRESIDENTS, PARTIES, AND INSTITUTIONAL CHOICE IN LATIN AMERICA 18 (2013) (“Adaptation . . . can take place by the introduction of formal alterations, by old rules being interpreted in new ways, or by the development of informal rules and practices that transform the meaning of existing institutions.”). 34 See Griswold v. Connecticut, 381 U.S. 479, 485-86 (1965). 35 See Brannon P. Denning & John R. Vile, The Relevance of Constitutional Amendments: A Response to David Strauss, 77 TUL. L. REV. 247, 275 (2002) (“If the nation is to continue with a written constitution that contains the specificity of some of the provisions of the existing document, there will be times when, absent flagrant disregard for constitutional language, some amendments will be required as defects become apparent, or changes are desired.”). 36 In fact, many of the rational-choice and behavioral mechanisms discussed in this Article can also cause stickiness in judicial interpretations of the Constitution. See also supra note 17 (summarizing the scholarship on path dependency in judicial decisions). 2016] Constitutional Stickiness 909 constitutional stickiness. As to rational choice, where the costs of transition to alternative constitutional configurations are high relative to benefits, suboptimal constitutional configurations can endure despite rational behavior by constitution-makers. I then apply behavioral research to consider how cognitive limitations and biases that lead individuals to act irrationally can contribute to constitutional stickiness. Specifically, I analyze the status quo bias, anchoring bias, availability heuristic, hedonic adaptation, and excessive veneration of the constitution. Part III turns from the descriptive to the normative and analyzes the normative implications of constitutional stickiness. I. A THEORY OF CONSTITUTIONAL STICKINESS The layout of the modern computer keyboard is called “QWERTY” by reference to the first six letters of its top line. This layout first appeared in typewriters. Before the adoption of the QWERTY arrangement, the typewriter suffered from a defect: the keys would jam if the user typed too quickly.37 The QWERTY layout was designed specifically to slow down typing speed so as to prevent mechanical key blockage.38 In addition, for marketing purposes, the letters that comprise the word “TYPEWRITER” were placed on the first line of the keyboard to permit salespersons to efficiently demonstrate how the machine operates by quickly typing the brand name — Typewriter.39 These purposes were obviated by later technological developments, such as the ball-point typewriter and computer keyboards, which do not cause mechanical key blockage.40 Yet, despite the apparent availability of superior, more efficient, and more ergonomic layouts, the QWERTY arrangement has “stuck.”41 The initial adoption of the QWERTY keyboard prompted manufacturers to create software and hardware compatible with that keyboard layout, which in turn reinforced it.42 In addition, the increasing numbers of QWERTYSee David, Clio, supra note 13, at 333. See Jürgen Beyer, The Same or Not the Same — On the Variety of Mechanisms of Path Dependence, 5 INT’L J. SOC. SCI. 1, 2 (2010). 39 See David, Clio, supra note 13, at 333. 40 Beyer, supra note 38, at 2; Marciano & Khalil, supra note 20, at 76. 41 See Beyer, supra note 38, at 1-2; Posner, supra note 20, at 583; see also Marciano & Khalil, supra note 20, at 76 (“[T]he most appealing and seemingly more efficient alternative to QWERTY is the Dvorak Simplified Keyboard (DSK).”). But see S. J. Liebowitz & Stephen E. Margolis, The Fable of the Keys, 33 J.L. & ECON. 1, 2-3 (1990) [hereinafter Fable of Keys] (disputing the supposed inefficiency of the QWERTY arrangement). 42 See David, Clio, supra note 13, at 335-36. 37 38 910 University of California, Davis [Vol. 49:899 trained typists encouraged employers to buy QWERTY machines, and the value of QWERTY training for employees increased with every employer who did so.43 This is a classic illustration of path dependence. The concept was first introduced by economic historians interested in trajectories of technological development.44 Although the term has assumed multiple meanings,45 in common parlance, path dependence refers to “an outcome or decision [that] is shaped in specific and systematic ways by the historical path leading [up] to it.”46 Path dependence, as I use the term here, is more than a broad, vague claim that “history matters.”47 I use the term in a narrower sense where each stage of historical development constrains the next stage in the temporal sequence and stimulates movement in the same direction.48 The past narrows the choices available in the future and links decision making through time.49 Consider an example from evolutionary biology.50 Evolution, like constitutions, must contend with existing configurations. Although evolutionary processes result in the “survival of the fittest,”51 the See id. See Thelen, How Institutions Evolve, supra note 10, at 219. 45 See Tom Ginsburg et al., Libertarian Paternalism, Path Dependence, and Temporary Law, 81 U. CHI. L. REV. 291, 292-93 (2014). 46 Hathaway, supra note 17, at 604. As Margaret Levi explains: 43 44 Path dependence has to mean, if it is to mean anything, that once a country or region has started down a track, the costs of reversal are very high. There will be other choice points, but the entrenchments of certain institutional arrangements obstruct an easy reversal of the initial choice. Perhaps the better metaphor is a tree, rather than a path. From the same trunk, there are many different branches and smaller branches. Although it is possible to turn around or to clamber from one to the other — and essential if the chosen branch dies — the branch on which a climber begins is the one she tends to follow. Margaret Levi, A Model, a Method, and a Map: Rational Choice in Comparative and Historical Analysis, in COMPARATIVE POLITICS: RATIONALITY, CULTURE, AND STRUCTURE 28 (Mark Irving Lichbach ed., 1997). 47 Hathaway, supra note 17, at 604; Pierson, Increasing Returns, supra note 15, at 252 (distinguishing the broader and narrower conceptions of path dependence). 48 Pierson, Increasing Returns, supra note 15, at 252. 49 See DOUGLASS C. NORTH, INSTITUTIONS, INSTITUTIONAL CHANGE, AND ECONOMIC PERFORMANCE 98-99 (James Alt & Douglass North eds., 1990). 50 In providing this example, I do not mean to suggest that evolution in biology and law are identical phenomenon. The example is meant to illustrate the existence of similar path-dependence mechanisms in diverse fields. 51 CHARLES DARWIN, THE ORIGIN OF SPECIES 94 (Charles W. Eliot ed., 1909). 2016] Constitutional Stickiness 911 species that do survive are far from perfect.52 The panda’s thumb is a prominent example. The “thumb” functions as a thumb, but it is not a thumb from an anatomical perspective.53 The panda’s thumb is an enlarged bone that developed from the radial sesamoid, a small component of the wrist.54 The design of the thumb is clumsy and, to quote Stephen Gould, “wins no prize in an engineer’s derby.”55 It cannot move like a thumb, but it presents a workable solution to a recurring problem in the life of a panda: it allows the panda to strip leaves from bamboo shoots, its primary food source.56 If one were designing the panda from scratch, the optimal design of the thumb might look very different. But the panda’s thumb as it exists today was the best that evolution could do given the limited options provided to it by the historical development of the panda’s anatomy.57 Similar arbitrary starting points can also constrain later choices in constitutional design. Consider the Necessary & Proper Clause in Article I, Section 8 of the U.S. Constitution. Significant as it is today, the framers did not precisely know the Clause’s purpose.58 The Constitutional Convention records do not state a precise justification for the Clause.59 Nor do the records otherwise indicate that the Clause resulted from a bitter compromise between the Federalists and the Anti-Federalists.60 The inclusion of the Clause can be traced to Charles Pinckney’s suggestion that Congress be authorized “to make all laws for carrying the foregoing powers into execution.”61 But the Committee on Detail did not explain its acceptance of Pinckney’s suggestion nor its addendum of the phrase “which shall be necessary and proper” into Pinckney’s proposal.62 The Constitutional Convention appears to have perceived these additions as “mere rhetorical flourish” and unworthy of debate, which suggests that the Convention was unaware of the significance that the Clause would See Hathaway, supra note 17, at 616. See STEPHEN JAY GOULD, THE PANDA’S THUMB 22-23 (1980). 54 Id. at 22. 55 Id. at 24. 56 Id. at 21-24. 57 See id. at 22-24. 58 Mark Graber, Unnecessary and Unintelligible, in CONSTITUTIONAL STUPIDITIES, CONSTITUTIONAL TRAGEDIES 43, 46 (William N. Eskridge, Jr. & Sanford Levinson eds., 1998). 59 See id. 60 See id. 61 Id. 62 Id. 52 53 912 University of California, Davis [Vol. 49:899 later assume.63 In Federalist No. 44, James Madison essentially admits that the Clause was pointless, writing, “Had the Constitution been silent on this head, there can be no doubt that all the particular powers requisite as means of executing the general powers would have resulted to the government, by unavoidable implication.”64 The Necessary & Proper Clause might be considered the legal analogue of the panda’s thumb, a once-insignificant part that evolved to serve a momentous function. Although the panda lacks the luxury of redesigning its thumb, constitutions can be rewritten from scratch. Yet starting points in constitutional design can, and often do, exert substantial historical weight on subsequent choices. I call this phenomenon “constitutional stickiness.” Without constitutional stickiness, each constitutional choice would be made in isolation65 and previous constitutional choices would have no effect on current options.66 With constitutional stickiness, however, past constitutional configurations constrain later decisions, displaying what mathematicians call a “sensitive dependence on initial conditions.”67 Constitution-makers are humans, and humans make errors with regularity. Some of these errors can even generate payoffs. A polity can learn from the undesirable consequences of suboptimal constitutional provisions and amend them. But stickiness may impede constitutional change and freeze in place norms that will continue to generate undesirable consequences. In other words, constitutional stickiness can inhibit the changes that should result as constitution-makers learn from their predecessors’ errors. Although stickiness particularly resists the removal or amendment of existing constitutional provisions, the addition of new provisions can also generate stickiness because addition alters the status quo. For example, the addition of a federal system of government changes what was a non-federal state to a federal state. Likewise, the grant of additional authorities to the executive branch can alter the existing constitutional power balance between the legislature and the See id. THE FEDERALIST NO. 44 (James Madison), available at http://avalon.law.yale.edu/ 18th_century/fed44.asp. 65 See Page, supra note 15, at 107. 66 See id. 67 See JON ELSTER, ULYSSES AND THE SIRENS: STUDIES IN RATIONALITY AND IRRATIONALITY 47 (1979) (“[P]references are always relative to a past history of choices.”); Liebowitz & Margolis, Fable of Keys, supra note 41, at 3; Page, supra note 15, at 91, 107. 63 64 2016] Constitutional Stickiness 913 executive. Stickiness can therefore resist all types of constitutional alterations — whether by removal, amendment, or addition. Constitutional stickiness is a variable phenomenon and can be divided into three gradations in decreasing order of strength. First, constitutional stickiness, at its sturdiest, can make it prohibitively difficult to alter an initial constitutional choice. Second, even where the initial constitutional choice can be altered, that choice can still constrain the options available to future constitution-makers.68 In other words, the initial choice can limit or narrow available future paths. Constitutional change might continue, but bounded by the path initially forged. Constitution-makers may take their bearings from the existing constitutional norms, with the past provisions serving as a baseline for departure and shaping their mental models.69 For example, despite the significant changes that the U.S. Constitution introduced, it also retained many features from the Articles of Confederation.70 The Interstate Compact Clause, the Privileges and Immunities Clause, and the Extradition Clause, among others, all had their origins in the Articles of Confederation.71 Likewise, the Constitution gave Congress all of the authorities it had under the Articles of Confederation, and neither the Virginia Plan nor the New Jersey Plan sought to remove them.72 In describing the authorities of Congress, the Constitution also adopted the language and structure of the Articles of Confederation.73 The post-World War II German Constitution, otherwise known as the Bonn Constitution or the German Basic Law, provides another illustration. Although its drafters agreed to reject all semblance of 68 As Carl Schorske explains in a different context, “when men produce revolutionary changes, they screen themselves from their own frightening innovations by dressing themselves in the cultural clothing of a past to be restored.” CARL E. SCHORSKE, THINKING WITH HISTORY: EXPLORATIONS IN THE PASSAGE TO MODERNISM 88 (1998). 69 See Beyer, supra note 38, at 3. 70 Ernest J. Brown, Politics and the Constitution in the History of the United States, 67 HARV. L. REV. 1439, 1441 (1954) (reviewing WILLIAM WINSLOW CROSSKEY, POLITICS AND THE CONSTITUTION IN THE HISTORY OF THE UNITED STATES (1953)) (noting that the U.S. Constitution “borrowed and carried forward from the text of the Articles of Confederation not only ideas and concepts but even, in many instances, the very wording, verbatim or only slightly adapted, of phrases or whole clauses”). 71 Eric M. Freedman, Why Constitutional Lawyers and Historians Should Take a Fresh Look at the Emergence of the Constitution from the Confederation Period: The Case of the Drafting of the Articles of Confederation, 60 TENN. L. REV. 783, 793 n.35 (1993). 72 See Calvin H. Johnson, Homage to Clio: The Historical Continuity from the Articles of Confederation into the Constitution, 20 CONST. COMMENT. 463, 473 (2004). 73 Id. at 474. 914 University of California, Davis [Vol. 49:899 continuity with the Nazi past, they still looked to their previous constitution for guidance and found elements to incorporate into the new constitution.74 They abolished the provisions of the previous constitution that appeared most to blame for the rise of fascism, but retained other, more palatable portions.75 Another example appears in the successive iterations of the Egyptian Constitution. The 1971 Constitution required half of the members in the legislature to be “workers and farmers,” a remnant of the country’s socialist past.76 Although the provision had become anachronistic when the Arab Spring arrived, the 2012 Constitution retained it.77 The drafters of the 2014 Constitution also retained the provision, though they modified it by requiring the state to grant only “appropriate representation” to workers and peasants in the first House of Representatives to be elected after the Constitution was ratified.78 Third, constitutional stickiness might be completely overcome. Stickiness significantly resists fundamental alterations, but does not make them impossible.79 The drafters of the U.S. and Bonn Constitutions, to return to those examples, retained some features of their previous constitutions but also achieved significant transformations. As I describe in the next Part, switching constitutional paths is encumbered by a confluence of factors that may — but need not — lead to the retention of existing constitutional configurations.80 The analysis in the next Part also suggests that, all things being equal, some constitutional alterations will generate more stickiness than others. For example, the removal of existing provisions will tend to be more difficult than the addition of new ones. With some exceptions, alterations to structural provisions will also be more burdensome than changes to negative-rights provisions. I discuss why in the next Part. 74 Kim Lane Scheppele, A Constitution Between Past and Future, 49 WM. & MARY L. REV. 1377, 1402 (2008). 75 Id. at 1402-04. 76 Nathan J. Brown & Kristen Stilt, A Haphazard Constitutional Compromise, CARNEGIE ENDOWMENT FOR INT’L PEACE (Apr. 11, 2011), http://carnegieendowment. org/2011/04/11/haphazard-constitutional-compromise; see also CONSTITUTION OF THE ARAB REPUBLIC OF EGYPT art. 87, 11 Sept. 1971, as amended, 22 May 1980, 25 May 2005, 26 March 2007. 77 Brown & Stilt, supra note 76. 78 CONSTITUTION OF THE ARAB REPUBLIC OF EGYPT art. 243, 18 Jan. 2014. 79 See Beyer, supra note 38, at 4 (discussing how “path-dependent institutionalization” restricts but does not prevent change). 80 See id. at 3-4. 2016] Constitutional Stickiness II. 915 THE CAUSES OF CONSTITUTIONAL STICKINESS The previous Part provided a basic theory of constitutional stickiness. This Part turns to its causes. Because many constitutionmaking processes are opaque, it is prohibitively difficult to disaggregate the multiple factors that produce constitutional stickiness. In addition, constitutional design is highly variable, so different combinations of factors will be at work in different contexts. As a result, in this Part, I necessarily paint with a somewhat broad brush and discuss, drawing on rational-choice theory and behavioral law and economics, how a confluence of variables can generate constitutional stickiness. My goal here is not to create a behavioral model of constitutional design but rather to identify the mechanisms that contribute to constitutional stickiness. Section A first analyzes the causes of stickiness, applying a costbenefit analysis and traditional rational-choice assumptions. Section B then explains how behavioral biases and limitations may contribute to the stickiness of constitutional provisions. Although traditional law and economics often conflicts with behavioral law and economics, in the Sections that follow, I will show how rational-choice mechanisms and behavioral biases may complement and reinforce each other to cause constitutional stickiness. Before I proceed, two introductory points are in order. As I discussed above, constitutional stickiness is variable. Different types of constitutional alterations generate different costs and biases, which in turn produce varying degrees of resistance to change. I consider the salience of each type of cost and cognitive limitation with respect to different constitutional alterations in detail below. In addition, in discussing why constitutional provisions stick, I remain agnostic about whether certain constitutional configurations are more optimal than others. Optimality is a relative concept and a constitutional norm that is optimal in one context may be suboptimal in the other. In fact, as I discuss in Part III, constitution makers often confront more than one constitutional path that can lead to an optimal equilibrium. A. Rational Choice I begin this Section with a theoretical premise based on rationalchoice theory: Constitution-makers are self-interested actors who seek to minimize costs and risks and maximize personal payoff.81 At first 81 See, e.g., Robert J. Barro, The Control of Politicians: An Economic Model, 14 PUB. CHOICE 19, 19 (1973) (assuming that a public officeholder acts “to advance his own 916 University of California, Davis [Vol. 49:899 blush, constitutional stickiness might appear antithetical to rational choice. After all, if there is a superior alternative available to the constitutional status quo, one may assume that the rational course is to adopt it. Nevertheless, where the costs of adopting even demonstrably superior constitutional alternatives are high relative to their benefits, the constitutional status quo can endure. Put differently, if individuals are rational actors, they can retain constitutional provisions that are suboptimal from a societal perspective but that maximize their individual payoff. A methodological limitation should also be noted at the outset. Constitution-makers are not monolithic entities that follow a uniform code of “rational” conduct. Depending on context, constitutional design may bring together heterogeneous actors such as politicians, judges, academics, religious leaders, and individual citizens, among others, with divergent personal attributes and incentives. The assumptions associated with rational-choice theory may not apply equally to all of these individuals, some of whom, for example, may be motivated in their decision making by considerations of the greater public good even where it contradicts their self-interest.82 In addition, constitutions are sometimes imposed by outsiders — as in the case of Japan after World War II — who may have different sets of incentives than domestic constitution-makers.83 Further, the costs and benefits interests, and these interests do not coincide automatically with those of his constituents”); Donald J. Boudreaux & A.C. Pritchard, Rewriting the Constitution: An Economic Analysis of the Constitutional Amendment Process, 62 FORDHAM L. REV. 111, 113 (1993) (noting the assumption in public-choice analysis that “government actors and interest groups rationally pursue their own self-interests in the same way that private actors do in the marketplace”). This theoretical premise is also consistent with the strategic-realist theory of law and constitution-making. Under that theory, domestic institutions are the product of political bargains, and laws and constitutions reflect the self-interest of political elites. See Ran Hirschl, The Strategic Foundations of Constitutions, in SOCIAL AND POLITICAL FOUNDATIONS OF CONSTITUTIONS 157-58 (Denis J. Galligan & Mila Versteeg eds., 2013). But see BRUCE ACKERMAN, THE FUTURE OF LIBERAL REVOLUTION 48-54 (1992) (arguing that, during constitution-making periods, the relevant actors are motivated primarily by long-term principles, rather than shortterm partisan interests). 82 Note, however, any facially impartial arguments by constitution-makers in favor of the greater public good may be a strategic façade for a direct statement of their personal interest. See Jon Elster, Arguing and Bargaining in Two Constituent Assemblies, 2 U. PA. J. CONST. L. 345, 421 (2000) (noting “the simple fact of political life that selfinterest is often dressed up in impartial garb”). 83 See Zachary Elkins et al., Baghdad, Tokyo, Kabul: Constitution Making in Occupied States, 49 WM. & MARY L. REV. 1139, 1140-41 (2008) (discussing the inclusion of U.S. institutions in Japan’s Constitution drafted largely by American authorities in 1946); Benedikt Goderis & Mila Versteeg, The Transnational Origins of Constitutions: An 2016] Constitutional Stickiness 917 discussed below are distributed unequally across different actors who will make different cost-benefit calculations. Despite these limitations, rational choice can help identify some of the causes of constitutional stickiness and the likely responses by constitution-makers. I also recognize the objections levied against rational-choice theory by scholars of behavioral law and economics. Later in the Article, I relax these traditional rational-choice assumptions and apply behavioral research to analyze why constitutional stickiness may occur. This Section proceeds as follows. In Subsection 1, I first consider the incentives of constitution-makers. I examine how constitutional stickiness may result where constitution-makers derive few benefits from constitutional change. In Subsection 2, I analyze how the costs of constitutional change can preserve the status quo even where constitution-makers benefit from constitutional alterations. 1. Benefits of Constitutional Change Collective-action problems often cause stickiness in technological standards. People will be reluctant to switch to a different standard where they lack assurances that a sufficient number of others will also do the same.84 For example, an individual American may be reluctant to abandon the imperial system and adopt the metric system if she is uncertain whether the rest of the community will also make the switch. Change may be beneficial if made by the entire community, but without a community-wide adoption, the optimal choice might be to retain the imperial system. This problem results in part from the absence of a centralized authority that can mandate simultaneous change.85 For example, if legislation were adopted mandating or incentivizing the use of the metric system, the imperial system may be abandoned by a sufficient number of users. This, in turn, may alter the status quo and create a new equilibrium.86 Empirical Analysis 5-6 (Apr. 2011) (unpublished manuscript), available at http:// www.law.uchicago.edu/files/files/Versteeg_April2011[2].pdf (describing the imposition by outsiders of constitutions in over thirty former British colonies, as well as in Japan, Micronesia, Germany, Afghanistan, and Iraq, among others). 84 Gillette, supra note 17, at 819-20. 85 See id. at 820. 86 In 1975, Congress enacted the Metric Conversation Act, which established the U.S. Metric Board to “coordinate and plan the increasing use of the metric system in the United States.” The United States and the Metric System: A Capsule History, NAT. INST. OF STANDARDS & TECH., (May 2002), http://www.nist.gov/pml/wmd/metric/ upload/1136a.pdf. The American public largely ignored the Metric Board’s proposals, 918 University of California, Davis [Vol. 49:899 Unlike technological standards, constitutions are designed by a centralized authority, such as a constituent assembly, that can direct simultaneous change and move the polity to a new constitutional equilibrium (subject to a ratification process). The existence of this central authority decreases the uncertainty partially responsible for locking in suboptimal technological standards.87 After the constitution drafted by the central authority is ratified, the entire polity will move to a new constitutional equilibrium. The ability to change, however, does not necessarily yield change.88 As Tom Ginsburg and James Melton observe in a forthcoming paper,89 constitution-makers are unlike commercial businesses that derive significant benefits from innovation. Although path dependence has been documented in consumer products as well,90 competitive pressures may prompt the replacement of suboptimal products with more optimal alternatives.91 For example, a sleeker, more technologically capable cellular telephone, might attract more customers than an older, inferior design. In a competitive market, the iPhone 6 will outsell older cell phone models that resemble giant walkie-talkies, the iPod will outsell the Walkman, and the XBOX 360 will outsell the Gameboy. In contrast, even a casual survey of constitutions around the globe reveals many ancient relics that, in a competitive market, would have been replaced by superior alternatives. The Electoral College, which has outlived its usefulness, and the twenty-dollar threshold for the Seventh Amendment right to a jury trial are good examples from the U.S. Constitution. For several reasons, unlike the market for consumer products and services, the market for constitution-making is not particularly competitive.92 As an initial matter, unlike commercial businesses that profit from innovation, many constitution-makers lack a pecuniary interest in the constitutional provisions they draft.93 For example, the adoption of an establishment clause is unlikely to generate any economic benefits for an individual constitution-maker. The value derived from the which were not mandatory, and the Board was disbanded in 1982. Id. 87 See Gillette, supra note 17, at 824. 88 See id. 89 See generally Ginsburg & Melton, Innovation, supra note 7 (stating that “constitution-makers have relatively few incentives to innovate”). 90 See David, Clio, supra note 13, at 332-36. 91 Pierson, Increasing Returns, supra note 15, at 260. 92 See Ginsburg & Melton, Innovation, supra note 7, at 3-4. 93 Id. at 3. 2016] Constitutional Stickiness 919 separation of religion from government will not, in most cases, be pecuniary. To be sure, there are exceptions. McGuire and Ohsfeldt’s quantitative work, following Charles Beard’s famous thesis,94 finds support for the proposition that economic interests had some effect on voting behavior both at the Philadelphia Constitutional Convention of 1787 and the state ratifying conventions.95 But these pecuniary benefits, even where they exist, will often be less salient than the pecuniary benefits present in the economic marketplace. Unlike constitutional design, the economic marketplace provides individuals with clear entitlements to income from particular property rights and thus provides strong incentives to care about the long-term value of their economic assets.96 In addition to pecuniary benefits, reputational benefits, which can provide incentives to change in commercial products, are also likely to be slim in constitutional design.97 Constitution-makers are often anonymous bits of a large institution and, as a result, few obtain meaningful recognition for innovative constitutional policies.98 Most do not attain the ranks of James Madison or Alexander Hamilton99 and are lost to obscure history books or dusty records of constitutional conventions. In addition, many constitution-design processes are opaque, making it difficult to credit a single constitution-maker or even a group of constitution-makers for an innovative constitutional norm.100 Where reputational benefits are important for constitution- 94 See CHARLES A. BEARD, AN ECONOMIC INTERPRETATION OF THE THE UNITED STATES 73, 149-51 (1913). CONSTITUTION OF 95 See Robert A. McGuire & Robert L. Ohsfeldt, Economic Interests and the American Constitution: A Quantitative Rehabilitation of Charles A. Beard, 44 J. ECON. HIST. 509, 517 (1984). 96 Paul Pierson, The Limits of Design: Explaining Institutional Origins and Change, 13 GOVERNANCE 475, 481 (2000). 97 See Ginsburg & Melton, Innovation, supra note 7, at 3. 98 See id. 99 Id. 100 For example, it still remains open to serious debate which political group in Egypt — the Muslim Brotherhood or the non-Islamists — was responsible for adding a clause to the 2012 Egyptian Constitution that promised the consultation of the Islamic institution al-Azhar on questions of Islamic Law. See CONSTITUTION OF THE ARAB REPUBLIC OF EGYPT art. 4, 30 Nov. 2012; Clark Lombardi & Nathan J. Brown, Islam in Egypt’s New Constitution, FOREIGN POLICY (Dec. 13, 2012), http://mideast. foreignpolicy.com/posts/2012/12/13/islam_in_egypts_new_constitution (“Brotherhood members of the Constituent Assembly insist that they were not behind this language and indeed that it was non-Islamists who pushed al-Azhar into the document.”). 920 University of California, Davis [Vol. 49:899 makers, the opacity of constitutional design may hamper any individual motives for adopting innovative constitutional changes.101 Of course, constitution-makers will have the necessary incentives to adopt constitutional changes where change generates tangible benefits for them. For example, members of political parties with a popular candidate for president may advocate popular elections for the executive, regardless of the constitutional status quo.102 Conversely, the other parties may attempt to limit executive powers and support presidential elections by the parliament.103 During the 1919–21 constitution-making process in Poland, the conservative Parliament created a weak executive in the Constitution, in part because it was widely expected that the socialist Joseph Pilsudski would be elected president.104 Likewise, members of small political parties may favor proportional representation in elections, whereas large parties may insist on a first-past-the-post system,105 regardless of the existing constitutional configurations.106 Smaller states in a federal system may advocate equal representation in the federal government, whereas the larger states may oppose it, as was the case in the U.S. Constitutional Convention.107 In constitutional-design processes following a fundamental shift in political power, the newly empowered group may also derive benefits from changing the constitutional configurations that benefited their political rivals. For example, the post-apartheid South African Constitution introduced significant changes as political power shifted to the African National Congress during the country’s transition from apartheid to a more pluralistic democracy.108 Constitution-makers that cooperate with interest groups or constituents by constitutionalizing their preferences can also receive rewards in the form of votes, See Ginsburg & Melton, Innovation, supra note 7, at 4-5. Elster, Constitutionalism, supra note 5, at 474-75. 103 Id. 104 Id. at 475. 105 First past the post is a winner-take-all system, where political office is awarded to a single candidate that receives the most amount of votes. Mark Tushnet, 1937 Redux? Reflections on Constitutional Development and Political Structures, 14 U. PA. J. CONST. L. 1103, 1104 (2012). 106 See Jon Elster, Forces and Mechanisms in the Constitution-Making Process, 45 DUKE L.J. 364, 378 (1995) [hereinafter Forces and Mechanisms]. 107 Id. at 379. 108 See Richard J. Goldstone, The South African Bill of Rights, 32 TEX. INT’L L.J. 451, 451-52 (1997) (describing the major changes in the post-apartheid South African Constitution, including the addition of a justiciable bill of rights and the establishment of a new constitutional court to enforce those rights). 101 102 2016] Constitutional Stickiness 921 campaign contributions, and assistance with election efforts.109 All things being equal, this interest will be more salient for public officeholders who serve on the constitution-making body since they will have a vested interest in election or reelection to the government.110 Constitution-makers may also have incentives to adopt changes where it is too costly to retain the status quo.111 For example, various pathologies in state constitutions and the Articles of Confederation spurred significant alterations in the U.S. Constitution.112 The constitutional drafters for the French Fifth Republic moved to a semipresidential system to prevent the parliamentary chaos that led to the collapse of the Fourth.113 And, as noted above, the drafters of the postWorld War II German Basic Law abandoned many features of the Weimar Constitution blamed for permitting the rise of fascism.114 These past constitutional disasters can prompt constitution-makers to abandon or amend the culpable constitutional provisions. Importantly, however, constitution-makers can also receive rewards for preventing constitutional change, which can contribute to constitutional stickiness. Constitution-makers will have strong incentives to retain constitutional provisions and institutions that benefit them or their constituents. In the United States, recent attempts to adopt constitutional amendments to abolish the Electoral College or overturn the U.S. Supreme Court’s decision in Citizens United v. Federal Election Commission115 may have failed in part for this reason. The proposed Equal Rights Amendment — which would have protected gender equality in the U.S. Constitution — also failed in part because some state lawmakers opposed the amendment to undermine their opponents or to broker political deals that benefit them.116 See Gillette, supra note 17, at 830. Cf. Christine Jolls et al., A Behavioral Approach to Law and Economics, in BEHAVIORAL LAW AND ECONOMICS 13, 32 (Cass R. Sunstein ed., 2000) (contending that legislators always have an interest in obtaining their own reelection, and that interest will encourage them to respond to their constituents). 111 NEGRETTO, supra note 33, at 45 (“[U]nder certain conditions, the costs of replacing or amending constitutions may be lower than the costs of leaving these structures unreformed.”). 112 See Elster, Constitutionalism, supra note 5, at 477. 113 Id. 114 See id. 115 558 U.S. 310, 365 (2010). 116 See Donald T. Critchlow & Cynthia L. Stachecki, The Equal Rights Amendment Reconsidered: Politics, Policy, and Social Mobilization in a Democracy, 20 J. POLICY HIST. 157, 161-62 (2008). 109 110 922 University of California, Davis [Vol. 49:899 Likewise, in Poland, even though the upper house of the Parliament had lost its raison d’être, it was retained in its post-communist Constitution, primarily because the upper house had a veto on the Constitution and would not vote to abolish itself.117 In addition, all relevant groups order their conduct around existing constitutional configurations and can also have a vested interest in retaining them. That, in turn, can lead interest groups and constituents to pressure constitution-makers to refrain from adopting alternative constitutional provisions, even where they are more optimal than the status quo. Global pressures, in the form of diplomatic, military, or financial carrots or sticks, can also motivate constitution-makers to adopt changes. For example, David Law has argued that globalization can create an incentive for countries to compete in a race to the top in constitutional innovation.118 Specifically, Law argues that constitutionmakers will face growing incentives to adopt economic and human rights provisions to attract investors and elite workers.119 Likewise, the adoption of judicial review can provide legal assurances to foreign investors by protecting property rights and ensuring economic stability, especially in regimes with some level of government corruption.120 For that reason, the World Trade Organization (“WTO”) requires judicial supervision in trade-related areas, the establishment of which can convince a skeptical international community to invest in a state.121 Constitution-makers can also adopt standardized constitutional models in an effort to signal good intentions and become a part of the international community.122 Although these global pressures can serve as a potential source of constitutional changes, they may not necessarily result in the predictable evolution of legal norms.123 Countries may instead turn to 117 See Andrzej Rapaczynski, Constitutional Politics in Poland: A Report on the Constitutional Committee of the Polish Parliament, 58 U. CHI. L. REV. 595, 615 (1991). 118 David S. Law, Globalization and the Future of Constitutional Rights, 102 NW. U. L. REV. 1277, 1282 (2008). 119 Id. But see Ginsburg & Melton, Innovation, supra note 7, at 4 (expressing skepticism about the claim that globalization can prompt constitutional innovation since “[i]n a world of territorially defined nation states, populations are not able to ‘vote with their feet[,]’ emigrating freely to jurisdictions offering attractive rights packages”). 120 Ozan O. Varol, Stealth Authoritarianism, 100 IOWA L. REV. 1673, 1692 (2015). 121 Id. 122 Mila Versteeg, Unpopular Constitutionalism, 89 IND. L.J. 1133, 1183 (2014). 123 See Thelen, Historical Institutionalism, supra note 15, at 394. 2016] Constitutional Stickiness 923 existing domestic institutions to respond to global pressures,124 as opposed to altering the constitutional status quo.125 But even where constitution-makers derive benefits from change, these benefits may not be realized immediately and can take time to accrue.126 Although some constitutional choices generate immediate benefits, the consequences of many constitutional changes are often revealed in the long term as the new provisions are interpreted and enforced. Yet constitution-makers, especially politicians, will be more interested in short-term results and thus heavily discount long-term effects.127 Unlike a far-sighted commercial investor interested in the financial benefits of a long-term investment, elected officials tend to be more focused on the short term and the next election cycle.128 In contrast to the long-term (and therefore less certain) benefits from many instances of constitutional change, the costs associated with constitutional changes are typically incurred in the short run.129 For example, the establishment of a federalism system may require immediate, costly, and burdensome measures, but its consequences may not be revealed for a long period of time. Where the costs of constitutional change are incurred immediately but their benefits may be realized in the long term — and by someone else — constitutionmakers may have little incentive to invest in significant changes.130 In addition, in cases where constitution-makers participate in the design process only temporarily, it can be difficult to lengthen their time horizons by monitoring and sanctioning their behavior. Setting aside cases where the elected legislature drafts the constitution, members of a typical constituent assembly are elected or selected only for the purpose of writing a constitution. Unlike agents in an employment relationship — who might have strong incentives to Id. In addition, diffusion studies show that, despite the global informational flow generated by globalization, innovations tend to spread primarily to countries in close regional proximity. KURT WEYLAND, BOUNDED RATIONALITY AND POLICY DIFFUSION 19 (2007). 126 See Pierson, Increasing Returns, supra note 15, at 261 (contending that many complicated political decisions “only play out in the long run”); cf. Boudreaux & Pritchard, supra note 81, at 117 (contending that interest groups incur costs of constitutional amendments up front, but do not see benefits right away). 127 See Pierson, Increasing Returns, supra note 15, at 261. 128 See id. 129 See id. (discussing how politicians frequently discount the long-term effects of major political decisions because the nature of electoral politics encourages politicians to focus more on short-term effects). 130 See id. 124 125 924 University of California, Davis [Vol. 49:899 perform well over a long time horizon to retain employment or obtain a promotion — the employment of these constitution-makers automatically terminates at the end of the design process. The consequences of their constitutional choices, however, may not be revealed during the course of their employment or even within their lifetime. Even where consequences become apparent earlier, it may be difficult, as noted above, to assign individual responsibility to constitution-makers for specific constitutional choices where the drafting occurs in a complex, collective, and often opaque, bargaining process. These factors, in turn, can make it difficult to monitor and sanction constitution-makers and incentivize them to care about the long-term consequences of their choices.131 The lack of monitoring and sanctioning mechanisms might also suggest, however, that constitution-makers will be unencumbered to pursue constitutional changes. After all, even if their constitutional adventures produce undesirable consequences, they may escape accountability. Yet the empirical data on stickiness contradicts this conclusion.132 Although the absence of monitoring and sanctioning mechanisms may prompt some constitution-makers to abandon the handcuffs of the past, as a result of the costs and cognitive biases discussed below, existing constitutional provisions still serve as powerful anchors, producing varying degrees of stickiness. And even where constitutional change generates immediate benefits, alterations still may not occur. Where constitutional change benefits one group, it may harm another, which may stymie its adoption by holding out during the design process. Especially where the constitutional alteration concerns a particularly contentious issue, consensus may be prohibitively difficult to achieve because political groups benefit from maintaining a distinct position on such issues, decreasing the likelihood of defections by individual constitutionmakers.133 Even groups that are harmed by the constitutional status quo may prefer not to oppose the existing arrangements if the costs of attempting to renegotiate or altogether abandon the constitutional order are higher than the costs of retaining the status quo.134 In the United States, the Anti-Federalists, as Daryl Levinson explains, “rather quickly came to accept a constitution they had vehemently opposed, 131 See Versteeg, supra note 122, at 1180 (noting that “the experts that write the constitutions often do so in relative insulation from democratic pressures”). 132 See supra note 7 and accompanying text. 133 Cf. Rosalind Dixon & Tom Ginsburg, Deciding Not to Decide: Deferral in Constitutional Design, 9 INT’L J. CONST. L. 636, 656 (2011). 134 Levinson, supra note 1, at 712. 2016] Constitutional Stickiness 925 in large part because of the calculation that even a bad law was better than lawlessness.”135 2. Costs of Constitutional Change Simply put, changing existing constitutional provisions can be expensive. All things beings equal, continuing down the same constitutional path will cost less than changing it, at least in the short term.136 What is more, unlike benefits of constitutional change — which, as noted above, often accrue in the long term — constitutional change can impose significant transition costs immediately.137 These costs can be divided into three categories: costs of drafting, costs of negotiation and ratification, and costs of implementation. I consider each in turn below. a. Costs of Drafting Adhering to the constitutional status quo often reduces the workload of a constitution drafter. Especially where constitution drafters must complete the design process in a relatively short period of time, as they often do,138 retaining existing constitutional provisions saves them precious resources by allowing them to build on the foundations constructed by earlier designers. To be sure, some constitutional provisions are relatively easy to adopt through imitation of global or regional models, which lowers the drafting costs.139 For example, constitution-makers can copy, with little immediate costs, individual rights adopted in other influential constitutions or international human rights treaties. Britain’s former colonies in Africa, upon gaining independence, adopted the exact same boilerplate bills of rights.140 These bills of rights, in turn, were modeled on the European Convention on Human Rights and Fundamental Id. See Gerard Alexander, Institutions, Path Dependence, and Democratic Consolidation, 13 J. THEORETICAL POL. 249, 254 (2001); Hathaway, supra note 17, at 607. 137 See Alexander, supra note 136, at 254. 138 See infra text accompanying notes 256–64. 139 See Versteeg, supra note 122, at 1180 (suggesting that constitution-makers’ adoption of universal rights and ready-made constitutional model explains constitutions’ disconnect from popular values); see also VICKI C. JACKSON, CONSTITUTIONAL ENGAGEMENT IN A TRANSNATIONAL ERA 40 (2009) (“[M]any foreign constitutions drafted since World War II rely on international human rights instruments (or on other constitutions that relied on these instruments) as archetypes, leading to parallel rights-protecting provisions.”). 140 Versteeg, supra note 122, at 1188. 135 136 926 University of California, Davis [Vol. 49:899 Freedoms.141 Even with readily available constitutional models, however, constitution-makers may retain existing provisions to avoid formulation errors, ease drafting, and benefit from the available judicial interpretations of existing provisions.142 There are also information costs associated with changing the status quo at the drafting stage. Information costs refer to the costs of obtaining information about the consequences of a constitutional choice. Knowledge is often at a significant premium in constitutionmaking. Constitution-makers lack perfect, or even good, foresight, especially since there is often a lack of sustained expertise in constitution-drafting assemblies. Many constitution-makers have never written a constitution before and will not do so again.143 What is more, the consequences of a new and untested constitutional norm are usually revealed only through the passage of time.144 A new constitutional provision may prove to be unworkable in practice or produce undesirable substantive outcomes. Despite sustained efforts by scholars, the academic literature has produced little guidance on the consequences of constitutional choices.145 The parties at the bargaining table therefore often cannot predict which groups will be advantaged and disadvantaged by particular configurations. The drafters may also be uncertain about their own normative preferences and the acceptability of those preferences to the polity.146 Faced with such uncertainty, many constitution-makers, as Kim Lane Scheppele explains, “turn to history to find models to follow, ideas to plunder, and guides to steady themselves in their own troubled times.”147 If the existing provision has been in place for some appreciable duration, Id. Cf. Kahan & Klausner, supra note 19, at 350 (discussing how using a standard, commonly used contract term has many benefits, including “avoidance of formulation errors, ease in drafting, [and] availability of judicial rulings on the validity and interpretation of the term”). 143 Ginsburg & Melton, Innovation, supra note 7, at 4. 144 See Dixon & Ginsburg, supra note 133, at 644. 145 See Ginsburg & Melton, Innovation, supra note 7, at 5. 146 For example, Alexis de Tocqueville, who was elected to the Constituent Assembly following the 1848 revolution in France, openly admits his sense of doubt: “But what preyed most on my hopes and my nerves throughout the nine years spent in public affairs, and what still remains the most frightful memory of that time, was the constant doubt in which I was forced to live about what was best to do each day.” Scheppele, supra note 74, at 1399. 147 Id. at 1398; see also Elster, Constitutionalism, supra note 5, at 476 (“In most Eastern European countries today, there is a tendency to look to the precommunist constitutions as sources of inspiration.”). 141 142 2016] Constitutional Stickiness 927 constitution-makers should have much better information about it than a novel provision. Uncertain judicial construction of novel constitutional provisions can also increase information costs. In many constitutional systems, the provisions drafted by the constitution-making body will be interpreted by a separate institution, the judiciary.148 The judicial construction of new constitutional provisions might introduce a great deal of uncertainty into the constitution-design process and increase information costs, which might dissuade the constitution-makers from incurring the costs of deviating from the more certain constitutional status quo. Information costs might be reduced, at least to some extent, by examining the implications of any untested constitutional provisions adopted by other nations. The data provided by such comparative examination, however, may be highly context dependent. Constitutional norms appropriate for one context may be inappropriate for another for a multitude of historical, cultural, political, or legal reasons.149 If a constitutional norm has not been implemented and tested in that particular country, that norm might produce consequences difficult to forecast at the time of its adoption. In addition, similar constitutional norms can be interpreted drastically differently by the relevant judicial, political, and social actors in different contexts.150 Although high information costs may stymie constitutional changes, information costs will not always be high. Where the constitutional designers are adequately equipped with satisfactory levels of information and are reasonably confident about their normative preferences and the acceptability of those preferences to the polity, they can rationally adopt novel constitutional norms.151 148 See David S. Law & Mila Versteeg, The Evolution and Ideology of Global Constitutionalism, 99 CALIF. L. REV. 1163, 1199 (2011) [hereinafter Global Constitutionalism] (noting that by 2006, 87% of constitutions had either de jure or de facto judicial review). 149 See Michel Rosenfeld & András Sajó, Introduction, in THE OXFORD HANDBOOK OF COMPARATIVE CONSTITUTIONAL LAW 1, 14-15 (2012) (Michel Rosenfeld & András Sajó eds., 2012) (noting that the nexus between identity and norms is an important factor in the creation of constitutions). 150 See Vicki C. Jackson, Comparative Constitutional Law: Methodologies, in THE OXFORD HANDBOOK OF COMPARATIVE CONSTITUTIONAL LAW, supra note 149, at 71 (noting that “[c]onstitutions are made and then interpreted in complex and distinctive historical contexts”). 151 See ROBERT A. DAHL & CHARLES E. LINDBLOM, POLITICS, ECONOMICS, AND WELFARE 84 (1953) (arguing that a greater amount of knowledge and confidence in preferences 928 b. University of California, Davis [Vol. 49:899 Costs of Negotiation and Ratification Negotiating and ratifying a constitution are often delicate exercises in consensus building. The process ordinarily brings together representatives from major facets of the polity to agree to a document that will be acceptable to most citizens and respond to their needs. These groups will often have competing visions for the document and will disagree, and do so vehemently, over its content. The framers of the U.S. Constitution, for example, passionately disagreed, among other things, about the constitutional role of the federal government vis-à-vis the states, the necessity of a Bill of Rights, and the question of slavery. Recent efforts at constitution drafting across the Arab World also revealed deep conflicts about state-religion relations, judicial review, the constitutional rights of women and minorities, and the freedom of speech and assembly. As constitutional passions increase, so do decision costs, which refer to the costs associated with reaching a constitutional decision.152 Although some passion is necessary to jump start the constitutional-design process and provide the requisite motivation to take action, excessive passion can make consensus building, especially on controversial provisions, prohibitively difficult.153 As a general matter, decision costs are likely to be higher for altering entrenched provisions as opposed to retaining them.154 The phenomenon of legislative inertia is well-documented in the literature.155 Legislative inertia often freezes bad laws in place because, as Neal Katyal puts it, “it is so much harder to get legislatures to do something than it is to get them not to do something.”156 The legislative inertia phenomenon is even more salient in constitutional design. The mere labeling of a political issue as constitutional, as opposed to legislative, raises the stakes involved because constitutions are perceived as supreme and durable instruments. That, in turn, enable larger and rational amounts of change). 152 Samaha, supra note 6, at 612. 153 Dixon & Ginsburg, supra note 133, at 642. 154 Levinson, supra note 1, at 708 (“Maintaining coordination around the existing, and therefore focal, order will always be much easier than attempting to recoordinate around some alternative constitutional regime.”). 155 Cf. Russell Korobkin, Behavioral Economics, Contract Formation, and Contract Law, in BEHAVIORAL LAW AND ECONOMICS, supra note 110, at 123-25 (finding support for the inertia hypothesis in the context of contract drafting). 156 Neal Katyal, Sunsetting Judicial Opinions, 79 NOTRE DAME L. REV. 1237, 1240 (2004) (emphasis added). 2016] Constitutional Stickiness 929 makes it more difficult to adopt alterations.157 Even in nations where the constitutional amendment rule sets a much lower threshold than Article V of the U.S. Constitution, ratifying a constitutional amendment is still costlier than legislation. For example, in New Zealand, where the Constitution may be altered through an ordinary legislative act, documents that enjoy constitutional status tend to be more costly to amend or repeal.158 Likewise, in the United Kingdom, where the Parliament is not legally constrained by a codified constitutional text, a claim that legislation raises constitutional concerns often has the effect of “raising the temperature of the debate,”159 which can stymie its adoption. In addition, constitutional design usually takes the form of a bilateral monopoly.160 That is, the parties at the bargaining table have no available alternative negotiating partners yet they also have an incentive to hold out.161 In the world of constitutional design, which entrenches one group’s preferences into a durable document, finishing second does not count for much. When the stakes are high, the parties to the constitutional bargain may be more reluctant to yield and more likely to hold out for a better bargain.162 Holding out, in turn, prevents constitutional change and results in retention of the status quo. Two recent constitution design processes are illustrative. For example, sharp divisions over the replacement of Turkey’s parliamentary system with a presidential one was at least partially responsible for derailing the country’s recent constitution-making process.163 Although all major political parties agreed that the existing 157 See JANET L. HIEBERT, LIMITING RIGHTS: THE DILEMMA OF JUDICIAL REVIEW 3 (1996) (discussing the debate over adoption of the Canadian Charter of Rights and Freedoms, “described as the single most important innovation of the constitutional changes of 1982”); Stephen M. Griffin, The Nominee Is . . . Article V, in CONSTITUTIONAL STUPIDITIES, CONSTITUTIONAL TRAGEDIES, supra note 58, at 51-52. 158 See JON ELSTER, ULYSSES UNBOUND: STUDIES IN RATIONALITY, PRECOMMITMENT, AND CONSTRAINTS 101 & n.28 (2000); Julian N. Eule, Temporal Limits on the Legislative Mandate: Entrenchment and Retroactivity, 1987 AM. B. FOUND. RES. J. 379, 394 n.61 (noting that, even under legislation-like amendment procedures, “there remain moral and political restraints on the legislative alteration of constitutional doctrine”). 159 ERIC BARENDT, AN INTRODUCTION TO CONSTITUTIONAL LAW 30 (1998). 160 Ginsburg et al., supra note 45, at 337. 161 Id. at 337-38. 162 See Dixon & Ginsburg, supra note 133, at 639. 163 See Gulsen Solaker, Hopes Fade for a New Turkish Constitution, REUTERS (Nov. 18, 2013, 1:23 PM), http://www.reuters.com/article/2013/11/18/us-turkey-constitutionidUSBRE9AH0OV20131118 (reporting that the creation of a presidential system was “[o]ne of the most contentious issues” facing the drafters); see also Ozan Varol, Constitution-Making in Turkey: Towards a Presidential System?, I-CONNECT (Dec. 3, 2012), 930 University of California, Davis [Vol. 49:899 constitution should be amended significantly, the ruling party’s insistence on a presidential system — which many argued would spur authoritarian governance in Turkey — brought the drafting process to a halt.164 Likewise, disagreement during the 2010 constitution-making process in Kenya on whether and how to regulate abortion was almost sufficient to defeat the entire constitution-making process.165 Even where the constitution-making body is able to agree on the constitutional alterations, there are also decision costs associated with the ratification of a constitutional change.166 The often arduous process for ratification, and the preferences of the body that will ratify the constitutional changes, will constrain the constitution-making body’s proposals.167 Ratification often requires a popular referendum, a supermajority of the legislature, or both,168 which render constitutional alterations significantly more costly than legislation.169 Ratification also often involves costs associated with organizing elections, articulating the proposals to the public, and mobilizing political and public support.170 In Canada, for example, two sets of proposed constitutional amendments — the 1987 Meech Lake Accord and the 1992 Charlottetown Accord — failed ratification. The Meech Lake Accord failed to be ratified by its June 1990 deadline, despite initially high levels of public approval, overwhelming support in the House of Commons (242–16), and ratification within one year by eight of the ten provinces.171 The Charlottetown Accord was defeated in a public http://www.iconnectblog.com/2012/12/constitution-making-in-turkey-towards-apresidential-system (discussing how the proposal of replacing the parliamentary system in Turkey with a presidential system without checks and balances caused controversy). 164 See Solaker, supra note 163. 165 Dixon & Ginsburg, supra note 133, at 659. 166 Decision costs may vary with the type of alteration at issue. For example, all things being equal, a constitution that contains an expansive array of rights will tend to enjoy more popular support in a referendum than a constitution that adopts relatively few rights. See Versteeg, supra note 122, at 1168; see also Daniel Lansberg-Rodriguez, Wiki-Constitutionalism, THE NEW REPUBLIC (May 25, 2012), http://www. newrepublic.com/article/politics/75150/wiki-constitutionalism (“Latin American leaders have discovered that, by packaging ever-longer lists of promises and rights alongside greater executive functions, they can make a new constitution appealing enough to the masses that they will vote for it in a referendum.”). 167 See Elster, Forces and Mechanisms, supra note 106, at 374. 168 See id. at 366-67. 169 See Boudreaux & Pritchard, supra note 81, at 117. 170 NEGRETTO, supra note 33, at 45. 171 Christopher P. Manfredi, Institutional Design and the Politics of Constitutional Modification: Understanding Amendment Failure in the United States and Canada, 31 LAW & SOC’Y REV. 111, 123 (1997). 2016] Constitutional Stickiness 931 referendum even though it also initially enjoyed high levels of public support and obtained the assent of important political elites.172 Decision costs will be much lower where a unified group dominates the constitution-making process and has the ability to unilaterally impose outcomes.173 With sufficient institutional authority and popular support, a dominant group can adopt self-interested changes to the constitutional status quo without the consensus of opposition groups.174 This trend has been fairly common in Latin America.175 For example, during the 1949 constitution-making process in Argentina, the dominant party was able to alter the constitutional status quo to increase its institutional influence and weaken the competitiveness of the opposition.176 More recently, the Fidesz party in Hungary, armed with the requisite two-thirds parliamentary supermajority for constitutional amendments, implemented sweeping constitutional changes intended to stack the constitutional deck in its favor and undermine political opposition. c. Implementation Costs Some constitutional norms also require large set-up or implementation costs, which can stymie their adoption. Where set-up costs for a new constitutional path are high, constitution-makers, operating within budgetary strictures, will have a strong incentive to remain on the same trajectory.177 All things being equal, the addition of positive rights will cost more than the addition of negative rights, assuming that the positive right is not a “sham” provision that will go unenforced in practice.178 For example, the adoption of a positive right to government-sponsored health care would require the government to establish a comprehensive health-care system for all citizens, which, if faithfully implemented, can be quite costly. In contrast, negative rights tend to cost less than positive rights, providing more room for change in this area. Changes to structural provisions will often generate higher implementation costs than the addition of individual-rights provisions. Id. at 132. See NEGRETTO, supra note 33, at 109. 174 Id. at 113. 175 Id. 176 Id. at 113-14. 177 See Pierson, Increasing Returns, supra note 15, at 254. 178 See generally David S. Law & Mila Versteeg, Sham Constitutions, 101 CALIF. L. REV. 863, 916 (2013) (noting that positive rights can be costly to implement, harder to uphold, and easier to violate). 172 173 932 University of California, Davis [Vol. 49:899 To be sure, the line between structural provisions and individualrights provisions can be elusive since it is possible to interpret many rights provisions in structural terms.179 For example, the adoption of an individual right to a criminal defense attorney at the public’s expense may require the establishment of supporting institutional structures, such as a public defender’s office, which can also generate significant costs. As a general matter, changes to structural provisions — such as an electoral system or federalism — often require corresponding changes to complementary institutions far more extensive than rights provisions, making rights provisions less expensive to add.180 For example, any alterations to the Electoral College in the U.S. Constitution would require numerous changes to party platforms, election commissions, and federal and state electoral laws and regulations, which contributes to its stickiness. The empirically demonstrated global trend towards “rights creep” — which refers to the increasing number of constitutions that have adopted an increasing number of individual rights — supports the theory that the addition of individual-rights provisions will be less costly than changes to structural provisions.181 But where individual rights (such as a right to health care or criminal defense attorney) require supporting structures, their addition can impose similarly high implementation costs and generate stickiness. Self-reinforcement can also increase implementation costs for several reasons. In this context, self-reinforcement means that the adoption of the original constitutional norm foments a set of forces and complementary institutions that reinforce and strengthen the stickiness of the original norm.182 As Kathleen Thelen explains, “once 179 See Ozan O. Varol, Structural Rights (Jan. 10, 2016) (working paper) (on file with author). 180 As I explain below, however, the removal of existing individual-rights provisions can be quite difficult as a result of the endowment effect. 181 See Zachary Elkins et al., Getting to Rights: Treaty Ratification, Constitutional Convergence, and Human Rights Practice, 54 HARV. INT’L. L.J. 61, 63 (2013) (finding that both the number of rights included in national constitutions and the number of national constitutions with rights has increased steadily); Law & Versteeg, Global Constitutionalism, supra note 148, at 1195 (documenting that “[o]ver the last six decades . . . the number of rights in the average constitution has crept upward”); cf. Rosalind Dixon, Constitutional Non-Redundancy (Jan. 10, 2016) (working paper) (on file with author) (“The average length of national constitutions is now roughly 20,000 words; 50 years ago it was 16,000 words; and 100 years ago, it was 11,000 words.”). 182 See Page, supra note 15, at 88. Self-reinforcement follows a different logic from reactive sequence arguments. As James Mahoney explains, “[w]hereas self-reinforcing sequences are characterized by processes of reproduction that reinforce early events, reactive sequences are marked by backlash processes that transform and perhaps 2016] Constitutional Stickiness 933 a particular path is chosen, actors adapt to the existing institutions in ways that push them further along the trajectory.”183 Once initial steps are taken down a particular path, the costs of switching to another, previously plausible alternative increase and the path not taken becomes progressively remote.184 Some constitutional arrangements can create what Douglass North calls “the interdependent web of an institutional matrix” by requiring complementary organizations and support networks.185 As the concomitant matrix of laws, institutions, and expectations expands, the alteration of the provision grows costlier over time. Consider a polity that has created a constitutional guarantee of health care by the government. To effectuate that guarantee, the polity will need to construct government hospitals, hire doctors, nurses, and other hospital staff, and purchase medical equipment. Over time, these complementary bodies will develop specialized skills and expand relationships with other institutions, which in turn reinforces their own stability and generates increasing benefits and powerful inducements to remain on the same constitutional path.186 A constitutional right to government-sponsored health care may also lead employers to not offer health-care benefits, which also reinforces the initial choice of a government-sponsored health care system. As a result, altering this constitutional right would generate significant implementation costs by requiring the reconfiguration of settled expectations and complementary institutions. In addition, political parties, interest groups, and other political and social actors often shape their agenda to fit various aspects of the constitutional structure, such as federalism or the electoral system.187 The groups that have invested in and benefited from these structures will be deeply committed to preserving and expanding them.188 For example, the U.S. Constitution has prompted the development of a large array of institutions, such as a central bank, ministries, interest reverse early events.” Mahoney, supra note 16, at 526; see also Pierson, Not Just What, supra note 28, at 85 (noting that in reactive sequences “action and reaction shift the system in a new direction, but not one that reinforces the first move”). 183 Thelen, Historical Institutionalism, supra note 15, at 392 n.27. 184 Pierson, Increasing Returns, supra note 15, at 251; Thelen, Historical Institutionalism, supra note 15, at 392 n.27. 185 NORTH, supra note 49, at 95. 186 See Pierson, Increasing Returns, supra note 15, at 255, 259. 187 Levinson, supra note 1, at 713. 188 Id. at 713. 934 University of California, Davis [Vol. 49:899 groups, and legislative committees.189 These complementary institutions have also established their own constituencies who might resist any attempts to significantly modify the constitutional arrangements that prompted their creation.190 In addition to supporting the interests of their intended constituents, some constitutional institutions can also empower subordinate groups or the initial losers of the constitutional bargain.191 Where divergent segments of society begin to obtain leverage from existing constitutional configurations, these configurations will tend to collect greater political support over time.192 For example, although the Anti-Federalists vehemently resisted the adoption of the U.S. Constitution, they eventually came to accept it in part because they emerged victorious in the 1800 elections and began to benefit from the constitutional scheme of government.193 Where the initial losers of the constitutional bargain begin to benefit from it, however, selfreinforcement may not tell the whole story.194 Instead of wholly embracing these initial constitutional configurations, as selfreinforcement would suggest, the losers may turn these institutions into objects of ongoing political contestation, which, in turn, may transform them over time.195 Increasing the size of the existing constituency for a particular constitutional norm or configuration can also foment conflict over its meaning and goals, introducing new pressures for change.196 Although these modifications may not happen through formal alterations of the constitutional text, informal, subterranean changes may ensue through the re-interpretation of the existing provisions. Self-reinforcement can also result because of the learning that takes place after a constitutional system has adopted a particular norm. Brian Arthur refers to this phenomenon as “learning effects,” which occur when knowledge “gained in the operation of complex systems also leads to higher returns from continuing use.”197 The production of a body of constitutional law on a particular provision can generate a positive feedback loop, permit all relevant actors in the legal system to 189 190 191 192 193 194 195 196 197 ELKINS ET AL., supra note 7, at 20. Id. at 20. Cf. Thelen, How Institutions Evolve, supra note 10, at 216. See Levinson, supra note 1, at 715. Id. at 713. Cf. Thelen, How Institutions Evolve, supra note 10, at 231. See id. See id. ARTHUR, supra note 14, at 112-13. 2016] Constitutional Stickiness 935 carry out their tasks more effectively, and, up to a point, increase the pay-off for additional movement in the same constitutional path.198 For some constitutional provisions, the learning effects may be trivial, but for others, they can be quite substantial.199 Consider, for example, the Necessary and Proper Clause in the U.S. Constitution. As discussed above, although the Clause was thought to be an insignificant addition to the Constitution, its adoption has prompted thousands of judicial opinions interpreting it. Judges have accumulated experience over time after deciding cases that invoke the Clause and passed that accumulated experience onto future generations through reasoned judicial opinions.200 In addition, constitutional lawyers have developed expertise in the Clause. Law schools and law school textbooks have incorporated cases that educate students on the Clause. Where these learning effects are salient for constitution-makers — because, for example, they happen to be judges, lawyers, or law professors — they can contribute to the selfreinforcement of the constitutional status quo. In contrast, the introduction of novel constitutional provisions may impose costs on the legal system. The amendment or replacement of an existing constitutional provision would require the abandonment of some or all of the existing legal expertise on that provision and require the development of a new body of doctrine intended to effectuate it.201 That, in turn, would require judges, lawyers, and other relevant actors to adapt to the new provision. Novel provisions may also increase constitutional litigation since the applicable law is likely to be more uncertain after the adoption of a new constitutional provision.202 Therefore, even if an alternative constitutional norm might appear superior, the costs that would be imposed on the legal system from its replacement may result in its retention.203 The costs on the legal system might be lowered, to some extent, by constitutional borrowing. If another polity has adopted the same constitutional norm, the legal system can borrow some portions of the See Pierson, Not Just What, supra note 28, at 77. See Kahan & Klausner, supra note 19, at 351. 200 See Hathaway, supra note 17, at 627-28. 201 Cf. RICHARD A. POSNER, HOW JUDGES THINK 145 (2008) (“Adherence to precedent [limits the judicial workload] both directly, by reducing the amount of fresh analysis that the judges have to perform, and indirectly, by reducing the number of appeals, since the more certain the law, the lower the litigation rate.”). 202 Cf. id. at 144-45 (discussing how judicial precedent begets stability). 203 The costs imposed on the legal system will not be as salient for constitutionmakers who are not part of the relevant legal community affected by the constitutional alterations. 198 199 936 University of California, Davis [Vol. 49:899 already developed jurisprudence, reducing some of the costs associated with generating a new body of constitutional doctrine. This is a function of positive feedback, which means that a choice generates positive externalities204 if that choice is also made by other people.205 In the constitutional context, borrowing can permit the use of another legal system’s already developed jurisprudence on a constitutional provision. Although some constitutional borrowing may occur, the wholesale adoption of a foreign system’s jurisprudence would be highly unlikely. And even where borrowing occurs, the legal system will still incur some costs in adapting to the foreign jurisprudence and modifying it to fit the domestic context. As a result, even with the possibility of borrowing, transition costs will be much less for the retention of the status quo than the adoption of a constitutional norm that exists in another legal regime. Constitutional decisions also create negative externalities206 with respect to time and money.207 The more resources a constitutional provision requires, the greater its impact will be on future constitutional choices. For example, the creation of a constitutional right to education might require a substantial amount of resources, which can create negative externalities with respect to future constitutional rights. If the polity has spent its available resources on enforcing the constitutional right to education, it may lack adequate resources for a constitutional right to health care. *** In sum, where the costs of constitutional change exceed its expected benefits, the constitutional status quo will stick. Adherence to the constitutional past can therefore be rational even where superior alternatives are available. Where the benefits of constitutional change exceed the costs, however, one would expect alterations to occur under rational choice theory. Yet, constitutional provisions may stick even where the benefits of change outweigh its costs. I explain why in the next Section. 204 A positive externality is a benefit awarded to a third party who did not pay for it. See N. GREGORY MANKIW, PRINCIPLES OF ECONOMICS 199 (6th ed. 2012). 205 Page, supra note 15, at 88. 206 A negative externality is a cost imposed on a third party who did not choose to incur that cost. See MANKIW, supra note 204, at 196. 207 See Page, supra note 15, at 111-12 (explaining that time and money create negative externalities). 2016] Constitutional Stickiness 937 B. Behavioral Law and Economics In the previous Section, I applied rational-choice theory to explain how constitutional stickiness may occur despite rational behavior by constitution-makers. Rational choice admittedly does not fully capture the entire complexity of the incentives and motivations of constitution-makers.208 It also neglects the suboptimal choices that constitutional drafters make as a result of their cognitive limitations and biases, which can result in inaccurate perceptions of the relevant costs and benefits and deviations from comprehensive rationality.209 Simply put, human judgment is not perfect.210 Important research in behavioral law and economics has challenged the “rational actor” model of traditional economics in favor of a more nuanced view of decision-making termed bounded rationality.211 This line of research shows that cognitive biases and heuristics cause individuals to “form confident opinions based on inadequate or badly biased information and then hold to these opinions in the face of substantial disconfirming data.”212 This does not mean that human behavior randomly fluctuates around rational judgments.213 Rather, behavioral research shows that human beings exhibit certain systematic biases, which permits predictive analysis. In this Section, I examine several systematic biases and cognitive limitations that may affect constitution makers during the design process to produce constitutional stickiness: status quo bias, anchoring bias, availability heuristic, hedonic adaption, and excessive veneration of the constitution. At this juncture, two methodological caveats are in order. First, the biases and limitations I discuss here are neither inevitable nor invariable. Nor will they uniformly affect the heterogeneous actors 208 See Richard H. Pildes & Elizabeth Anderson, Slinging Arrows at Democracy: Social Choice Theory, Value Pluralism, and Democratic Politics, 90 COLUM. L. REV. 2121, 2143 (1990). 209 See WEYLAND, supra note 125, at 11; Russell B. Korobkin & Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 CALIF. L. REV. 1051, 1069 (2000); Cass Sunstein, Introduction, in BEHAVIORAL LAW AND ECONOMICS, supra note 110, at 3 [hereinafter Introduction]. 210 Norbert L. Kerr et al., Bias in Judgment: Comparing Individuals and Groups, 103 PSYCHOL. REV. 687, 687 (1996). 211 Christopher Buccafusco & Cristopher Jon Sprigman, The Creativity Effect, 78 U. CHI. L. REV. 31, 33 (2011) (describing the erosion of the “rational actor” model “in favor of a more nuanced and empirically robust view of human decision-making as ‘boundedly rational’”). 212 ROBERT BOYD & PETER J. RICHERSON, CULTURE AND THE EVOLUTIONARY PROCESS 168 (1985). 213 Kerr et al., supra note 210, at 687. 938 University of California, Davis [Vol. 49:899 involved in constitutional design. In some settings and for some of the relevant actors, some of these biases and limitations may have low or zero salience. Where appropriate, I discuss when these biases and limitations are likely to be of low significance and how they can affect various public and private actors differently. Second, there are also undeniable behavioral differences in individual decision-making and group decision-making, such as a constitutionaldesign process. Despite sustained efforts, behavioral research has not produced a simple, coherent answer to the question of whether group judgments are more or less biased than individual judgments.214 As a result, for the purposes of this Section, I generally assume that the biases discussed here are equally effective in the context of group decision-making and, where available, I discuss behavioral research that documents these biases in group decision-making. 1. Status Quo Bias The status quo bias refers to a behavioral preference for the current state of affairs.215 The bias results from a preference for inaction when decision-makers are presented with multiple choices.216 In addition to preferring the status quo, people tend to be biased against novel ideas that generate uncertainty about their consequences.217 The bias can affect both private actors and public decision-makers charged with overhauling the constitution. As a result of the status quo bias, they might be reluctant to alter pre-existing constitutional provisions and 214 Id. at 692-93 (surveying the empirical literature); id. at 714 (“At best, our analyses offer an existence proof that collective rationality can sometimes be superior to individual rationality, but they also suggest that over a large and plausible region of relevant parameter space, group decision making actually exacerbates the biases observed in individual decisions.”). 215 See William Samuelson & Richard Zeckhauser, Status Quo Bias in Decision Making, 1 J. RISK & UNCERTAINTY 7, 8 (1988). 216 See Ilana Ritov & Jonathan Baron, Status Quo and Omission Biases, 5 J. RISK & UNCERTAINTY 49, 49 (1992). 217 Jennifer S. Mueller et al., The Bias Against Creativity: Why People Desire but Reject Creative Ideas, 23 PSYCHOL. SCI. 13, 13 (2012) (“Uncertainty is an aversive state that people feel a strong motivation to diminish and avoid.” (citations omitted)); see also Eric F. Rietzschel et al., The Selection of Creative Ideas After Individual Idea Generation: Choosing Between Creativity and Impact, 101 BRIT. J. PSYCHOL. 47, 65 (2010) (“[P]eople appear to have a strong preference for ideas they believe can and should be adopted, and . . . seem to believe that this is incompatible with the selection of original ideas.”). 2016] Constitutional Stickiness 939 demand a great deal to justify departures even where the optimal outcome is their amendment or replacement.218 From a behavioral perspective, existing constitutional provisions enjoy a first-mover advantage. The prevailing orthodoxies in the existing constitutional order occupy an “almost monopolistic position.”219 The repeated application of existing constitutional provisions elevates them to a higher position.220 Competing constitutional norms may thus be perceived as presumptively undesirable.221 Take, for example, the choice between a presidential system and a parliamentary system. The U.S. Constitution establishes a presidential system, which has remained in place since its adoption. As a result, the presidential system has come to occupy a unique position in the minds of both private actors and public decisionmakers, similar to first-movers in consumer products such as CocaCola, Kleenex, and Xerox.222 The market for regime types in the United States is thus skewed towards the dominant product (presidentialism), which, in turn, reinforces the position of presidentialism, making it difficult to dislodge and replace with an alternative form of government.223 Even where a search for a novel constitutional idea is prompted by inadequacies in the existing constitutional order, the status quo bias will skew the search in favor of provisions compatible with existing constitutional norms.224 218 See Samuelson & Zeckhauser, supra note 215, at 8 (“Faced with new options, decision makers often stick with the status quo alternative, for example, to follow customary company policy, to elect an incumbent to still another term in office, to purchase the same product brands, or to stay in the same job.”); Sunstein, Introduction, supra note 209, at 4; Russell Korobkin, The Status Quo Bias and Contract Default Rules, 83 CORNELL L. REV. 608, 675 (1998) (“Because individuals tend to prefer the status quo to alternative states, they are likely to prefer the default [contract] term, whatever it may be, to other options, all other things being equal.”). 219 Eoin Carolan, Diffusion of Ideas and Comparative Law 11 (2012) (unpublished manuscript), available at http://law.mc.edu/files/7113/3397/7796/Carolan.pdf; see also Levinson, supra note 1, at 708 (“[J]ust by virtue of its status quo position, the existing constitutional order will enjoy a special salience that conceivably preferable alternatives will lack.”). 220 See Carolan, supra note 219. 221 See id. 222 Marvin B. Lieberman & David B. Montgomery, First-Mover Advantages, 9 STRATEGIC MGMT. J. 41, 46 (1988). 223 See Carolan, supra note 219, at 12. 224 See id. at 15. Alison LaCroix invokes a similar rationale in describing why the Confederate Constitution showed strong continuity with the U.S. Constitution: “The words of the [U.S.] Constitution, its ways of framing questions, and indeed its very structure dominated the American consciousness to such a degree that even secessionists could not escape it.” LaCroix, supra note 12, at 1. 940 University of California, Davis [Vol. 49:899 But where the dominant product suffers from a negative association, the status quo bias may be less salient since public and private actors are less likely to ascribe normative desirability to existing constitutional provisions. The drafters of the post-World War II German Basic Law were able to discard those constitutional provisions that appeared most to blame for the rise of fascism in large part because of the evils associated with the dominant constitutional products of the Nazi past. For example, the vote of no confidence in the pre-World War II Weimar Constitution was constrained significantly in the post-World War II Basic Law since the instability caused by the frequent use of the vote was blamed, at least in part, for the rise of the Nazi Party.225 Likewise, the new Egyptian Constitution limits states of emergency to three months and requires the consent of the Parliament,226 which was most likely motivated by the abuse of the executive emergency power under the Mubarak regime. In other cases, however, the strength of the status quo bias may dominate any negative associations with the constitutional status quo. Although the drafters of the Egyptian Constitution limited executive declaration of emergency, they nevertheless retained the presidential system, which many commentators at least partially blamed for the country’s authoritarian past. One can hypothesize seven factors that affect the salience of the status quo bias in constitutional design. First, all things being equal, the salience of the bias may increase with the lifetime of the constitutional provision at issue. For example, a constitutional provision that has been in place for twenty-five years may be more likely to stick than a provision adopted two years prior to the constitution-drafting moment. Under this hypothesis, the status quo bias will tend to be particularly strong with respect to the U.S. Constitution, which has not been replaced since it went into force in 1791. The status quo bias may also be more salient for provisions that have been retained in successive constitutional-design processes. These provisions may be perceived as having withstood the test of time despite repeated opportunities to amend or discard them and therefore may be more difficult to displace. Second, the perceived degree of departure of the proposed constitutional norm from the status quo is also relevant to the salience of the status quo bias. If the novel constitutional provision is 225 See GRUNDGESETZ FÜR DIE BUNDESREPUBLIK DEUTSCHLAND [GRUNDGESETZ] [GG] [BASIC LAW], May 23, 1949, BGBl. arts. 67–68 (Ger.). 226 CONSTITUTION OF THE ARAB REPUBLIC OF EGYPT art. 154, 18 Jan. 2014. 2016] Constitutional Stickiness 941 compatible with the existing constitutional order, that can bolster the adaptability of the new provision and decrease the salience of the status quo bias. Conversely, if the new provision represents a marked departure from the constitutional status quo, that can increase the salience of the status quo bias and hamper attempts to adopt the new, markedly different provision.227 Third, behavioral studies show that the strength of the status quo bias increases with the number of available alternatives.228 Put differently, the more alternatives that people confront, the more likely they are to retain the status quo. Constitution-makers often have numerous choices available to them in designing constitutions. Even seemingly limited alternatives such as the tripartite choice between presidentialism, parliamentarism, and semi-presidentialism conceal many complexities. In a recent paper, Cheibub, Elkins, and Ginsburg demonstrate that these traditional categories show great internal heterogeneity across time and space, presenting constitution-makers with numerous options in allocating powers between the legislature and the executive.229 Confronted with copious alternatives, constitution-makers might opt for the status quo. Fourth, the endowment effect and loss aversion may also increase the salience of the status quo bias.230 The endowment effect refers to the empirical finding that people tend to overvalue things (including rights and privileges) that they already own.231 Loss aversion, a corollary to the endowment effect, refers to the human tendency to fear losses more than gains.232 Under the endowment effect and loss aversion, those who benefit from the constitutional status quo will value those benefits more highly than those who would benefit from 227 See EVERETT M. ROGERS, DIFFUSION OF INNOVATIONS 243 (5th ed. 2003); Carolan, supra note 219, at 15 (“[A]n idea which involves little or no innovation is more likely to be adopted than one that genuinely involves a departure from previous practice.”). 228 See Daniel Kahneman et al., Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias, 5 J. ECON. PERSP. 193, 198 (1991). 229 Jose Antonio Cheibub et al., Beyond Presidentialism and Parliamentarism, 44 BRIT. J. POL. SCI. 515, 516-17 (2014). 230 See Kahneman et al., supra note 228, at 194 (illustrating the relationship between the endowment effect, loss aversion, and status quo bias). 231 See Buccafusco & Sprigman, supra note 211, at 31; Steffen Huck et al., Learning to Like What You Have — Explaining the Endowment Effect, 115 ECON. J. 689, 689 (2005); Kahneman et al., supra note 228, at 194. 232 Ginsburg et al., supra note 45, at 321; Kahneman et al., supra note 228, at 19798 (“One implication of loss aversion is that individuals have a strong tendency to remain at the status quo, because the disadvantages of leaving it loom larger than advantages.”). 942 University of California, Davis [Vol. 49:899 constitutional change.233 As a result, even where a constitutional alteration would produce a net social benefit, one might expect those who benefit from the constitutional status quo to invest more in retaining it than those who would profit from constitutional change.234 This theory is supported by the empirical evidence, which shows that individual rights, once introduced, tend to maintain their popularity over time.235 All things being equal, under the endowment effect and loss aversion, the removal of existing provisions will be more difficult than the addition of new provisions. That is because where new provisions do not result in a loss of existing constitutional rights or privileges, their addition does not implicate the endowment effect or loss aversion. The result changes, however, where the design process resembles a zero-sum game. In other words, if the gain of one group from the addition of a constitutional provision results in a loss to another group, then the endowment effect and loss aversion can support the entrenchment of the status quo. For example, several commentators have argued that the provision of constitutional rights to crime victims would undermine the constitutional rights of criminal defendants.236 Likewise, in several states in the United States, the recent proposed addition of a state constitutional right to engage in farming and ranching practices generated significant resistance from animal rights groups.237 See Gillette, supra note 17, at 827. See id. at 827-28. 235 Elkins et al., supra note 181, at 72 (finding only four rights with a negative or flat trajectory over time: “the right to bear arms, the right to citizenship of those born in the state’s jurisdiction (jus soli citizenship), intellectual property rights, and the right to a jury trial”). 236 See, e.g., David E. Aaronson, New Rights and Remedies: The Federal Crime Victims’ Rights Act of 2004, 28 PACE L. REV. 623, 672 (2008) (“From the defendant’s perspective, providing crime victims a significant participatory role in criminal proceedings echoes back to injustices of the colonial period when alleged crime victims played a dominant role in criminal prosecutions through a system of private prosecution.”). 237 Julie Bosman, Missouri Weighs Unusual Addition to Its Constitution: Right to Farm, N.Y. TIMES (Aug. 2, 2014), http://www.nytimes.com/2014/08/03/us/missouri-considersadding-right-to-farm-to-state-constitution.html; Ryan Sabalow, Indiana Senate Kills ‘Right to Farm’ Amendment, INDY STAR (Feb. 24, 2015), http://www.indystar.com/story/news/ 2015/02/24/indiana-senate-kills-right-farm-amendment/23944627. Several other design features have also been described as zero-sum games in the literature. See, e.g., A.C. Pritchard & Todd J. Zywicki, Constitutions and Spontaneous Orders: A Response to Professor McGinnis, 77 N.C. L. REV. 537, 538-39 (1999) (stating that the structural constraints of the U.S. Constitution were intended to “pit government actors in a zero-sum game”); Barbara Bennett Woodhouse, The Constitutionalization of Children’s Rights: Incorporating Emerging 233 234 2016] Constitutional Stickiness 943 Although the endowment effect and loss aversion bolster the salience of the status quo bias, over the lifetime of a constitution, the initial coalition that lobbied for the adoption of a constitutional norm may dissipate.238 Depending on the time that has elapsed, the coalition for whose benefit the provision was adopted might no longer exist and efforts to recreate it might fail.239 In such cases, the endowment effect and loss aversion may not result in constitutional stickiness. The Prohibitionists are a good example of a group that coalesced only temporarily.240 Although they successfully lobbied for the adoption of the Eighteenth Amendment, which established Prohibition, they were unable to stop its repeal thirteen years later due in part to the temporary nature of their alliance.241 Fifth, the composition of the constitution-drafting body can also affect the salience of the status quo bias. Constitution-makers are often selected because they played a historical role in the event that gave rise to the new constitution-making process.242 In the United States, for example, many framers had served in the Continental Army or as officials for the Confederation or the Continental Congress.243 These constitution-makers, selected for their historical roles, can be more prone to turn to history than to the future when making constitutional choices.244 Likewise, judges, who have spent their careers interpreting and applying the status quo, may be more affected by the status quo bias during constitutional design than law professors who are rarely awarded tenure for publishing articles arguing that the status quo is “just fine as it currently exists.”245 Demographic factors can also be relevant.246 For example, ordinary citizens serving on a constitution-making body may be less loyal to the status quo than career politicians.247 Likewise, age can also play a role. Several behavioral studies have found a significant positive age effect Human Rights into Constitutional Doctrine, 2 U. PA. J. CONST. L. 1, 8 (1999) (arguing that the constitutionalization of parental rights erected barriers to constitutionalization of children’s rights). 238 See Gillette, supra note 17, at 828. 239 See id. 240 Boudreaux & Pritchard, supra note 81, at 119. 241 See id. 242 Scheppele, supra note 74, at 1379. 243 Id. at 1407 n.9. 244 Id. at 1379. 245 Gillette, supra note 17, at 824. 246 A comprehensive behavioral assessment of these factors is beyond the scope of this Article. 247 See Ginsburg & Melton, Innovation, supra note 7, at 12. 944 University of California, Davis [Vol. 49:899 on loss aversion — older participants were more loss averse and therefore potentially more susceptible to the status quo bias.248 Younger constitution-makers may be less wedded to the status quo since their innovative powers are at their peak and they have much less to cite from their own history to address the challenges of the present.249 Behavioral research reaches mixed conclusions on the effects of education and income on the status quo bias. Although some studies found that loss aversion decreases with higher levels of education,250 others did not find a significant effect.251 The results are also mixed with respect to the effects of high income.252 Sixth, time pressure can also increase the salience of the status quo bias. Although no study has examined the effects of time pressure on constitution drafting, behavioral research generally shows that severe time pressure negatively affects individuals’ ability to generate novel solutions to problems and reduces the quality of decisions. It also encourages “closing of the mind”: individuals discount available alternatives, fail to thoroughly process relevant information, and refrain from critical probing.253 Further, individuals are more likely to rely on heuristics when making decisions under severe time pressure, which also inhibits change.254 The creativity-inhibiting effects of time pressure have also been documented in group-negotiation settings.255 248 Katrine Hjorth & Mogens Fosgerau, Loss Aversion and Individual Characteristics, 49 ENVIRON. & RESOURCE ECON. 573, 586-87 (2011); Simon Gächter et al.,, IndividualLevel Loss Aversion in Riskless and Risky Choices 17 (CeDEx Discussion Paper Series, Working Paper No. 2010-20, 2010), available at http://hdl.handle.net/10419/49656; Eric J. Johnson et al.,, Exploring the Nature of Loss Aversion 19 (IZA Discussion Papers, Working Paper No. 2015, 2006), available at http://hdl.handle.net/10419/33324. But see Ye Li et al., Complementary Cognitive Capabilities, Economic Decision Making, and Aging, 28 PSYCHOL. & AGING 595, 607 (2013) (finding that older participants were slightly less loss averse than younger participants, although this difference was not statistically significant). 249 See Posner, supra note 20, at 592. 250 Hjorth & Fosgerau, supra note 248, at 588; Gächter et al., supra note 248, at 17-18. 251 Johnson et al., supra note 248, at 19. 252 Id. at 17-19 (finding that individuals with higher incomes were more loss averse). But see Hjorth & Fosgerau, supra note 248, at 586-87 (finding no significant effect of income on loss aversion). 253 Martin G. Kocher & Matthias Sutter, Time Is Money — Time Pressure, Incentives, and the Quality of Decision-Making, 61 J. ECON. BEHAV. & ORG. 375, 378 (2006); see also Anne Edland & Ola Svenson, Judgment and Decision Making Under Time Pressure: Studies and Findings, in TIME PRESSURE AND STRESS IN HUMAN JUDGMENT AND DECISION MAKING 27, 30 (Ola Svenson & A. John Maule eds., 1993). 254 Carsten K.W. De Dreu, Time Pressure and Closing of the Mind in Negotiation, 91 ORGANIZATIONAL BEHAV. & HUM. DECISION PROCESSES 280, 282 (2003). See generally Amos Tversky & Daniel Kahneman, Judgment Under Uncertainty: Heuristics and Biases, 2016] Constitutional Stickiness 945 Constitution-making is a frustratingly long process that often must be condensed, due to internal and external pressures, to a short time frame.256 Ginsburg, Elkins, and Melton observe that, of the eighty-six constitutions for which data is available, 50% were drafted in six months or less and over 25% were drafted in less than four months.257 The polity’s perceived need to reach a swift constitutional resolution of immediate concerns may lead to the imposition of internal time restraints on constitutional drafters. Political agreements may stipulate a time limit for the constitutional-design process in order to establish a basic framework for governance, as was the case in Kenya, Nepal,258 and most recently, Egypt.259 In addition to domestic pressures, foreign occupiers, anxious to terminate their involvement in a constitutional reconstruction, may also impose external time restraints on the design process.260 For example, international actors, including the United States and the United Nations, required the constitution-design process in Afghanistan to be completed within two years — a formidable challenge in a society emerging from twenty-five years of civil war.261 Similarly, United Nations officials required the constitutional drafting 185 SCI. 1124 (1974) (discussing the systematic errors that result from reliance on heuristics). 255 See De Dreu, supra note 254, at 291; Janice R. Kelly & Joseph E. McGrath, Effects of Time Limits and Task Types on Task Performance and Interaction of FourPerson Groups, 49 J. PERSONALITY & SOC. PSYCHOL. 395, 404-05 (1985) (finding that time constraints reduce the quality of performance, including originality and creativity, in a group essay writing task); Kocher & Sutter, supra note 253, at 388 (“Our results suggest that time pressure has . . . a negative effect on the quality of decision-making in an interactive context as well.”); Rick van der Kleij et al., Effects of Time Pressure and Communication Environment on Team Processes and Outcomes in Dyadic Planning, 67 INT’L J. HUM.-COMPUTER STUD. 411, 420-21 (2009) (finding that time pressure reduced the quality of teams’ written planning); Michael A. West, Sparkling Fountains or Stagnant Ponds: An Integrative Model of Creativity and Innovation Implementation in Work Groups, 51 J. APPLIED PSYCHOL. 355, 365, 379 (2002) (reviewing the literature that finds time restraints to impede creative problem solving in group settings). 256 Dixon & Ginsburg, supra note 133, at 642-43. 257 Ginsburg & Melton, Innovation, supra note 7, at 5. 258 See Dixon & Ginsburg, supra note 133, at 643. 259 See sources cited infra note 264 and accompanying text. 260 See Dixon & Ginsburg, supra note 133, at 642-43 (discussing how the United States pushed for a swiftly completed constitution in Iraq). 261 J. Alexander Thier, Big Tent, Small Tent: The Making of a Constitution in Afghanistan, in FRAMING THE STATE IN TIMES OF TRANSITION: CASE STUDIES IN CONSTITUTION MAKING 535, 558 (Laurel Miller ed., 2010). 946 University of California, Davis [Vol. 49:899 in East Timor to take place within ninety days.262 The United States occupation in Iraq likewise forced a rushed constitution-design process, which was completed in less than six months through a process that excluded Sunni factions.263 These temporal restraints on the constitutional-design process often do not permit sufficient deliberation over the adoption of novel constitutional norms and impede constitutional change. Several commentators have argued, for example, that the domestic time constraints imposed on the 2012 constitution-drafting process in Egypt contributed to the retention of many suboptimal provisions from the previous constitution.264 In addition, temporal restraints — which, as noted above, impede information gathering and processing — can increase even further the high information costs associated with new and untested constitutional provisions.265 Seventh, many constitution-drafting moments occur following a revolution or transition from one regime type to another, where the status quo bias is likely to be particularly salient. Revolutions tend to produce outbreaks of nostalgia.266 In the political and social turmoil that a regime transition produces, many wistfully harken back to the socially and economically stable days of the former regime.267 Change may be costly, difficult to comprehend, and questionable.268 As 262 Louis Aucoin & Michele Brandt, East Timor’s Constitutional Passage to Independence, in FRAMING THE STATE IN TIMES OF TRANSITION: CASE STUDIES IN CONSTITUTION MAKING, supra note 261, at 254. 263 See Vicki C. Jackson, What’s in a Name? Reflections on Timing, Naming, and Constitution-Making, 49 WM. & MARY L. REV. 1249, 1273 & n.61, 1275 (2008). 264 See, e.g., Saïd Amir Arjomand, The Islam and Democracy Debate After 2011, 20 CONSTELLATIONS 297, 308 (2013) (arguing that the abrupt closure of the inconclusive constitutional debate resulted in the retention of many provisions from the old constitution); Yasmine El Rashidi, Egypt: The Rule of the Brotherhood, N.Y. REV. BOOKS (Feb. 7, 2013), http://www.nybooks.com/articles/archives/2013/feb/07/egypt-rule-brotherhood (discussing the drafting process of the Egyptian Constitution and noting that the head of the Constitutional Assembly actively pushed for the swift completion of the document). 265 See supra Part II.A.2.a (discussing information costs). 266 SVETLANA BOYM, THE FUTURE OF NOSTALGIA xvi (2002). 267 GUILLERMO O’DONNELL & PHILIPPE C. SCHMITTER, TRANSITIONS FROM AUTHORITARIAN RULE: TENTATIVE CONCLUSIONS ABOUT UNCERTAIN DEMOCRACIES 4 (1986) (“Compared to periods of ‘order’ which characterize the high point of authoritarian rule, the uncertainty and indirection implied in movements away from such a state create the impression of ‘disorder.’ This impression some compare nostalgically with the past, while overlooking or regretting the transition’s revival of precisely those qualities which the previous regime has suppressed: creativity, hope, self-expression, solidarity, and freedom.”). 268 Levinson, supra note 1, at 691, 708; Adam Przeworski, Democracy as a Contingent Outcome of Conflicts, in CONSTITUTIONALISM AND DEMOCRACY 59, 75 (Jon 2016] Constitutional Stickiness 947 “revolution fatigue” sweeps over the nation,269 inherited institutional structures may appear normatively superior to theoretical alternatives.270 For example, according to an August 2013 nationwide survey of Egyptians by the Pew Research Center’s Global Attitudes Project, 80% of Egyptians believed that the country was worse off following Mubarak’s ouster.271 A constitution drafted in this environment may be prone to entrenching existing provisions as opposed to adopting novel alternatives. The relative insignificance of the changes to the Egyptian and Tunisian Constitutions following their revolutions in 2011 largely support that theory. Thus, even in the aftermath of a revolution — when one might expect tectonic constitutional shifts — the status quo bias, along with the other biases to which I now turn, can stymie constitutional changes. 2. Anchoring Bias The anchoring bias, which is related to the status quo bias, refers to a tendency to insufficiently adjust one’s judgments up or down from an initial starting value.272 Under the anchoring bias, initial reference points influence, or “anchor,” judgments.273 After these anchors are established, final judgments are influenced in the direction of the anchor.274 For example, when asked to estimate various numerical values — such as the percentage of African countries in the United Nations or the likelihood of nuclear war — people give significantly higher or lower estimates if they are first asked whether the value is greater or less than some arbitrary high or low value, compared to people who are asked to estimate the value without a starting point.275 Elster & Rune Slagstad eds., 1988) (“[T]he authoritarian power apparatus may resist the transition to democracy even when the forces within the civil society upon which the regime rests are willing to try their chances under democratic conditions.”). 269 See Michael Slackman, When a Punch Line Is No Longer a Lifeline for Egyptians, N.Y. TIMES (Apr. 5, 2011), http://www.nytimes.com/2011/04/06/world/middleeast/ 06cairo.html. 270 See Levinson, supra note 1, at 691. 271 Mohamed Younis, Egyptians See Life Worse Now than Before Mubarak’s Fall, GALLUP (Aug. 16, 2013), http://www.gallup.com/poll/164015/egyptians-life-worsemubarak-fall.aspx. 272 S. Plous, Thinking the Unthinkable: The Effects of Anchoring on Likelihood Estimates of Nuclear War, 19 J. APPLIED SOC. PSYCHOL. 67, 67 (1989); Tversky & Kahneman, supra note 254, at 1128. 273 Tversky & Kahneman, supra note 254, at 1128-30. 274 See Plous, supra note 272, at 67-68; Tversky & Kahneman, supra note 254, at 1128. 275 Tversky & Kahneman, supra note 254, at 1128. 948 University of California, Davis [Vol. 49:899 The anchoring bias is robust even when people are aware of the bias, they are paid to be accurate in their estimations, they are familiar with the subject matter, the anchors are outrageously high or low, or an expert claims that the given anchor point is not an accurate estimate.276 The anchoring bias has also been documented in the legal context. For example, several scholars have observed that sentencing guidelines and prosecutorial sentencing demands serve as influential anchors in judicial decision-making.277 Studies have also found that the anchoring bias affects federal magistrate judges278 and bankruptcy judges.279 The anchoring bias can also affect constitution-makers. In the context of constitution-making, the existing constitutional provisions serve as the “anchors” or the initial reference points.280 As a result of the anchoring bias, constitution-makers may be reluctant to alter existing constitutional provisions. And even where changes occur, the anchoring bias can skew them in the direction of the anchor — i.e., the constitutional status quo. 3. Availability Heuristic As Jon Elster aptly observes, “the task of constitution-making generally emerges in conditions that are likely to work against good constitution-making.”281 With some exceptions,282 constitutions are ordinarily written in the aftermath of a war, revolution, economic or social crisis, or other exceptional circumstances.283 The constitutional reconstructions following the 2011 Arab Spring exemplify the turbulent conditions under which constitution-makers often must See Plous, supra note 272, at 68, 83-85. Birte Englich & Thomas Mussweiler, Sentencing Under Uncertainty: Anchoring Effects in the Courtroom, 31 J. APPLIED SOC. PSYCHOL. 1535, 1538 (2001); Nancy Gertner, What Yogi Berra Teaches About Post-Booker Sentencing, 115 YALE L.J. POCKET PART 137, 138 (2006). 278 See Chris Guthrie et al., Inside the Judicial Mind, 86 CORNELL L. REV. 777, 778 (2001). 279 Jeffrey J. Rachlinski et al., Inside the Bankruptcy Judge’s Mind, 86 B.U. L. REV. 1227, 1236 (2006). 280 Cf. Kahan & Klasuner, supra note 19, at 363 (arguing that standard contract terms may anchor contracting parties). 281 Elster, Forces and Mechanisms, supra note 106, at 394. 282 Id. at 370 (noting the exceptional case of Sweden’s 1974 constitution drafting, which occurred in unexceptional circumstances). 283 Id. at 370-71. 276 277 2016] Constitutional Stickiness 949 operate. These turbulent circumstances create ripe conditions for the availability heuristic to stymie constitutional change. Under the availability heuristic, people tend to overestimate the seriousness of a risk if the risk is readily available or easy to recall.284 The observed frequency and the salience of the risk affect the strength of the heuristic.285 First, if a particular hazard has occurred recently, people will tend to believe that it has a higher probability of reoccurring in the future.286 For example, the number of people who purchase earthquake insurance increases in the immediate aftermath of a major earthquake.287 Conversely, residents of flood plains are less likely to purchase flood insurance if floods have not occurred in recent memory.288 Second, if the hazard at issue is particularly salient, the availability bias will be more pronounced.289 The salience of the hazard, in turn, increases if it is particularly dramatic or is widely publicized.290 The availability heuristic can contribute to constitutional stickiness in post-conflict constitutional design. Affected by the availability heuristic, constitution-makers and the public alike may overestimate and overreact to newly recognized threats generated by the chaotic post-conflict moment, however unlikely their recurrence may be.291 A constitution drafted in these tumultuous moments (or in their immediate aftermath) may therefore focus on the short-term societal needs of achieving economic and social stability at the expense of other constitutional goals.292 The drafters may shun experimentation with novel constitutional norms and retain preexisting provisions believing that established institutions and norms will better promote stability. In contrast, the adoption of untested constitutional provisions may destabilize the polity because they have the potential to generate unknown and perhaps undesirable consequences. 284 Sunstein, Introduction, supra note 209, at 5; Cass R. Sunstein, What’s Available? Social Influences and Behavioral Economics, 97 NW. U. L. REV. 1295, 1300-01 (2003) [hereinafter What’s Available]; see also Tversky & Kahneman, supra note 254, at 1127. 285 Jolls et al., supra note 110, at 37. 286 Id. 287 Sunstein, What’s Available, supra note 284, at 1301; Jolls et al., supra note 110, at 37 (“The same phenomenon may occur in other areas of regulatory law; an example here is the move toward heavy regulation of school bus safety in the wake of media coverage of school bus accidents in which children were killed.”). 288 Sunstein, What’s Available, supra note 284, at 1301. 289 Jolls et al., supra note 110, at 37. 290 See id. 291 Ozan O. Varol, Temporary Constitutions, 102 CALIF. L. REV. 409, 428-29 (2014). 292 Id. 950 University of California, Davis [Vol. 49:899 Exasperated by social and economic turmoil, the public also may lobby primarily for constitutional provisions that promote stability, which, in turn, may lead constitution-makers to misperceive the benefits of altering the status quo. Consider, for example, the preventive-detention provisions in the Indian Constitution. The Constitution was drafted after the end of British colonialism and during a conflict-laden period in Indian history marked by an armed rebellion in Telangana and violence with Pakistan over contested territories.293 Although fully aware of the use of preventive detention as a tool for tyranny by the British, the Constituent Assembly rejected due process protections for detainees and allowed the use of preventive detention.294 The Constitution permitted extrajudicial detention without charges to prevent future crime as a legitimate law enforcement tool.295 Preventive detention was perceived as a necessary evil to safeguard the new state at its conflict-laden transitional moment, but its preservation in a durable constitution allowed later government officials to use the provision to suppress opposition, long after the security threats that necessitated preventive detention dissipated.296 All things being equal, establishing consensus on existing provisions will be less costly than securing agreement on novel provisions. Significant changes may unnecessarily prolong the design process, which may detract attention from other pressing societal concerns. At times, conflict over novel constitutional provisions can derail the entire constitution-design enterprise,297 which may further destabilize the polity. To avoid these consequences and achieve a swift resolution of contentious constitutional questions, the drafters may retain the constitutional status quo. In some cases, however, the availability heuristic can prompt the adoption of novel provisions. First, achieving stability may require constitutional alterations. For example, even where a preexisting constitution does not contain any emergency powers, the post-conflict constitution may authorize government officials to declare emergencies in times of national crisis to ensure stability.298 Second, as 293 Hallie Ludsin, Peacemaking and Constitution-Drafting: A Dysfunctional Marriage, 33 U. PA. J. INT’L L. 239, 259 (2011). 294 See INDIA CONST. art. 22, § 3(b); Ludsin, supra note 293, at 259. 295 See INDIA CONST. art. 22, § 3(b); Ludsin, supra note 293, at 259. 296 Ludsin, supra note 293, at 260. 297 See supra notes 166–72 and accompanying text. 298 Cf. Gabriel L. Negretto & Jose Antonio Aguilar Rivera, Liberalism and Emergency Powers in Latin America: Reflections on Carl Schmitt and the Theory of 2016] Constitutional Stickiness 951 a result of the availability heuristic, a constitution-maker may selectively recall only those events that are particularly vivid for her. For example, if a constitution-maker has experienced discrimination in the past, she might lobby for novel constitutional policies to eradicate discrimination. Although the availability heuristic can result in individual motivations to lobby for novel constitutional policies, it is less likely to prompt the entire constitution-making body to adopt changes, where the most recent and salient societal concerns are the achievement of social and economic stability. Finally, the availability heuristic can also prompt constitutional change where constitutional designers emulate “available” global constitutional models or constitutional provisions adopted in temporal and geographical proximity.299 4. Hedonic Adaptation Hedonic adaptation, which refers to the human capacity to adapt to unpleasant circumstances,300 may also contribute to constitutional stickiness. Behavioral research shows that, despite dramatic events such as a disability, human beings have a tendency to recover fairly quickly and recapture their pre-event level of happiness.301 Although hedonic adaptation is neither inevitable nor invariable across different individuals,302 where it occurs, it may contribute to constitutional stickiness. Due to hedonic adaptation, relevant public and private actors may adapt to the constitutional status quo, leading to a belief, correct or incorrect, that the existing constitutional Constitutional Dictatorship, 21 CARDOZO L. REV. 1797, 1803-04 (2000) (noting that emergency provisions were adopted in many Latin American constitutions, including in Chile, Argentina, Mexico, and Colombia, in the mid-nineteenth century after a lack of legal mechanisms to address emergencies led to problems of constitutional legitimacy and gave authoritarian leaders opportunities to seize power). 299 See WEYLAND, supra note 125, at 6. 300 See John Bronsteen et al., Hedonic Adaptation and the Settlement of Civil Lawsuits, 108 COLUM. L. REV. 1516, 1527-28 (2008) [hereinafter Hedonic Adaptation]; Shane Frederick & George Loewenstein, Hedonic Adaptation, in WELL-BEING: THE FOUNDATIONS OF HEDONIC PSYCHOLOGY 302, 311-20 (Daniel Kahneman et al. eds., 1999). 301 Bronsteen et al., Hedonic Adaptation, supra note 300, at 1517; Daniel Kahneman & Richard H. Thaler, Anomalies: Utility Maximization and Experienced Utility, 20 J. ECON. PERSP. 221, 230 (2006) (“[A]s the new state loses its novelty it ceases to be the exclusive focus of attention, and other aspects of life again evoke their varying hedonic responses.”). 302 Bronsteen et al., Hedonic Adaptation, supra note 300, at 1530-31; see Ed Diener et al., Beyond the Hedonic Treadmill: Revising the Adaptation Theory of Well-Being, 61 AM. PSYCHOLOGIST 305, 311-13 (2006) (finding that individuals vary significantly in their rate of adaptation). 952 University of California, Davis [Vol. 49:899 arrangements are functioning reasonably well. As a result, they may retain the status quo even where the adoption of alternatives would lead to a more optimal outcome.303 This is not to suggest, however, that an equilibrium that results from hedonic adaptation is necessarily suboptimal. Hedonic adaptation, even to suboptimal constitutional configurations, can increase societal happiness and welfare.304 5. Excessive Veneration of the Constitution Another behavioral factor associated with constitutional stickiness is the societal veneration of the constitution. Constitutional veneration varies across both time and space,305 producing dramatically divergent prevailing cultural attitudes about resistance to constitutional change. In some nations, such as the United States, large portions of the public treat their constitutions as a sacred text306 that should remain untouched except in matters of profound significance.307 In other nations, constitutions are of little normative significance and are frequently amended and discarded.308 Although stickiness can be correlated with the societal veneration of the constitution, I do not suggest a one-directional causal relationship where constitutional veneration causes stickiness. The causation arrow may point in the opposite direction as well because the society may be more likely to revere a constitution that has remained relatively unchanged. In other words, it is equally plausible that causation is bidirectional, in that the stickiness of the constitution and the societal reverence for it mutually bolster each other. Although a detailed examination of the cultural factors that produce societal veneration of the constitution are beyond the scope of this Article, I will include some preliminary thoughts here. The veneration of the constitution may stem from a veneration of the generation that drafted it. As Thomas Jefferson wrote in 1816, future generations “ascribe to the men of the preceding age a wisdom more than human, and See Ginsburg et al., supra note 45, at 324. Id. at 324 n.69. 305 See Ginsburg & Melton, Amendment Cultures, supra note 7, at 13-14. 306 See Thomas C. Grey, The Constitution as Scripture, 37 STAN. L. REV. 1, 17 (1984) (noting the “well-attested phenomenon of American constitution-worship”); Sanford Levinson, “The Constitution” in American Civil Religion, 1979 SUP. CT. REV. 123, 12324 (1979). 307 Ginsburg & Melton, Amendment Cultures, supra note 7, at 13. 308 Id. 303 304 2016] Constitutional Stickiness 953 suppose what they did to be beyond amendment.”309 Max Weber refers to this quality as charismatic authority, which he defines as follows: [Charisma is] a certain quality of an individual personality by virtue of which he is considered extraordinary and treated as endowed with supernatural, superhuman, or at least specifically exceptional powers or qualities. These are not accessible to the ordinary person, but are regarded as of divine origin or as exemplary, and on the basis of them the individual concerned is treated as a “leader.”310 Out of enthusiasm, despair, or hope, later generations may recognize the charismatic authority of earlier political leaders,311 and the posterity of these leaders may continue to look to their principles and ideals for guidance. That, in turn, may produce a reluctance in future generations to alter the superior constitutional ideals of the past. This phenomenon is in part a function of monumentalistic historywriting that glorifies the past at the expense of the present.312 In his famous essay, On the Uses and Disadvantages of History for Life, Friedrich Nietzsche argues that the study of history can undermine our ability to meet the challenges presented by the modern world. Of the several criticisms he lodges against historical sense, one is particularly relevant to constitutional stickiness. Nietzsche contends that the study of history can belittle the present and make us feel like “latecomers.”313 He ascribes to old age “an appropriate senile occupation, that of looking back, of reckoning up, of closing accounts, of seeking consolation through remembering what has been,” rather than what can be.314 Selection bias also plays a role: It leads us to compare “the best of the past with the average of the present” since the best of the present has yet to be sorted from the average.315 As a result, we might feel inferior to the quasi-divine leaders of the past. If the principles and ideals of these foregone leaders are crystallized in a written constitution, those principles may tend to stick and trump the 309 Letter from Thomas Jefferson to Samuel Kercheval (July 12, 1816), in 10 THE WRITINGS OF THOMAS JEFFERSON 1816–1826, at 37, 42 (Paul Leicester Ford ed., 1899). 310 MAX WEBER, 1 ECONOMY AND SOCIETY: AN OUTLINE OF INTERPRETIVE SOCIOLOGY 241 (Guenther Roth & Claus Wittich eds., 1968). 311 See id. at 242. 312 Posner, supra note 20, at 590. 313 FRIEDRICH NIETZSCHE, UNTIMELY MEDITATIONS 83 (Daniel Breazeale, ed., R.J. Hollingdale, trans., 1997); Posner, supra note 20, at 576. 314 NIETZSCHE, supra note 313, at 109. 315 Posner, supra note 20, at 590. 954 University of California, Davis [Vol. 49:899 ideas produced by what the society views as the inferior, short-sighted political leaders of later generations. In modern United States, for example, large swaths of the public hold the founding generation in especially high regard. American children are taught to admire their forefathers from the very early stages of their education and literary works about the founders consistently top national bestseller lists.316 Over the years, values espoused by the founders have, according to Jamal Greene, “acquired a presumption of rightness within our political culture.”317 The reverence of the founding generation in the United States has contributed, at least in part, to the stickiness of most provisions in the U.S. Constitution, which remain formally unaltered. The veneration of the founding generation, and the Constitution they drafted, also suggests that even absent the high amendment threshold in Article V, many provisions in the U.S. Constitution would prove to be rather sticky. *** In sum, constitutional stickiness can also result from imperfect human judgment. As a result of the cognitive biases and heuristics discussed in this Section, constitution-makers can retain the constitutional status quo even where the optimal outcome would be its amendment or replacement. III. NORMATIVE IMPLICATIONS OF CONSTITUTIONAL STICKINESS Having described the causes of constitutional stickiness, this Part turns to the normative question: Is constitutional stickiness undesirable? As an initial matter, constitutional provisions may not stick. Where the costs and the salience of the biases I described above 316 See Tom Donnelly, Our Forgotten Founders: Reconstruction, Public Education, and Constitutional Heroism, 58 CLEV. ST. L. REV. 115, 117 (2010) (listing works about the founding generation that have been on national bestsellers lists); Richard S. Kay, “Originalist” Values and Constitutional Interpretation, 19 HARV. J.L. & PUB. POL’Y 335, 337 (1996) (“The constitutional Founders still seem to enjoy a regard, if not reverence, that has not significantly diminished over time, an attitude evidenced in popular culture, as well as in Supreme Court opinions.”); see also Michael Kirby, Constitutional Interpretation and Original Intent: A Form of Ancestor Worship?, 24 MELB. U. L. REV. 1, 2 (2000) (characterizing the American fascination with “original intent” of the Constitution as a form of legal “ancestor worship”). 317 Jamal Greene, Selling Originalism, 97 GEO. L.J. 657, 713 (2009); see also LaCroix, supra note 12, at 16 (“Examining the Confederate Constitution demonstrates the degree to which nineteenth-century Americans were constitution worshippers — not only the U.S. Constitution, but constitutions in general.”). 2016] Constitutional Stickiness 955 are sufficiently low, constitution-makers can break from the constitutional past and forge new paths. As discussed above, although prior decisions can still influence future choices, it may not block them. Where stickiness occurs, its normative consequences depend on the nature of the sticky provision. As relevant here, three types of provisions may stick: (1) an optimal provision; (2) a suboptimal provision that would be inefficient to change today; and (3) a suboptimal provision that can be corrected efficiently. I analyze each in turn below. First, even where a constitutional provision sticks, that provision might happen to be optimal (though, perhaps, not exclusively optimal). Some constitutional choices are as good as — if not better than — other alternatives. For example, a decision to adopt a constitutional right to freedom of religion might cause stickiness with respect to that right. But the costs and benefits of that right might have been fully appreciated and foreseen through a rational decisionmaking process by the initial constitution-makers. As a result, the initial adoption of that right may lead to an optimal outcome, even if it produces stickiness. In addition, it is possible for a suboptimal provision to become optimal over time as a result of societal evolution and adaptation. Importantly, constitutions do not always — or even frequently — change for the better. At first blush, one may conclude that constitutional stickiness will inevitably impede the development of superior constitutional standards since unrestricted markets are assumed to generate more optimal outcomes.318 Yet markets can also be dominated by powerful interest groups whose interests deviate from the furtherance of the public good.319 Constitutional design is no exception. For example, in constitution-making processes that occur without effective political pluralism and opposition, the opportunity is ripe for powerful groups to capture the constitution-design process and stack the constitutional deck in their favor. In Venezuela, President Hugo Chavez was able to seize unilateral control over the constitution-design process, producing a constitution that marginalizes opposition groups and creates a competitive authoritarian regime.320 The constitution-making processes in Russia, 318 Cf. Gillette, supra note 17, at 816 (noting the “assumption that unfettered markets inherently generate more efficient equilibria”). 319 See id. 320 See David Landau, Constitution-Making Gone Wrong, 64 ALA. L. REV. 923, 940 (2013). 956 University of California, Davis [Vol. 49:899 Belarus, and Kazakhstan also produced similar results.321 In these circumstances, stickiness can resist constitutional changes towards less optimal configurations. Japan provides another illustration. Adopted after Japan’s defeat in World War II, the 1947 Japanese Constitution includes a provision, Article 9, which renounces war and prohibits the maintenance of a military and other “war potential.”322 Most recently, Prime Minister Abe reinterpreted Article 9 to allow collective self-defense (i.e., aiding allies under attack when Japan itself is not threatened). Large swaths of the public vigorously oppose the relaxation of the pacifist ideals of Article 9, fearing a return to pre-World War II militarism and a renewal of the traumas of war.323 I reserve any normative judgments about the validity of these competing arguments, but simply note that the stickiness of Article 9 appears to be impeding change towards what wide segments of the public view as a proposed move to a less optimal constitutional framework. Second, constitutional stickiness can generate a suboptimal outcome that would be inefficient to change today. Initial constitutional choices may turn out to be inferior to other alternatives.324 If constitution-makers had known what we know now, they would not have made that choice. This problem can occur in many contexts outside of constitutional design. For example, a company’s decision to hire Employee X today may foreclose the hiring of Employee Y tomorrow, even where Employee Y would have been preferable to Employee X.325 Similarly, in the constitutional-design context, a federal system of government may have been perceived to be optimal, but the passage of time can reveal that choice to be suboptimal.326 In addition, external societal conditions might change, rendering suboptimal a norm that was optimal at the time of its adoption. The Electoral College is a good example. The result of bitter debate and uneasy compromise, the Electoral College has outlived its usefulness, a consequence the founders did not foresee. As Akhil Amar 321 See William Partlett, The Dangers of Popular Constitution-Making, 38 BROOK. J. INT’L L. 193, 209-23 (2012). 322 NIHONKOKU KENPŌ [KENPŌ] [CONSTITUTION], art. 9, para. 2 (Japan). 323 See Matthew J. Gilley, Japan’s Developing Military Potential Within the Context of Its Constitutional Renunciation of War, 14 EMORY INT’L L. REV. 1681, 1694-95 (2000). 324 See Liebowitz & Margolis, Fable of Keys, supra note 41, at 3; Gillette, supra note 17, at 823. 325 Gillette, supra note 17, at 813-14. 326 See Liebowitz & Margolis, Fable of Keys, supra note 41, at 3. 2016] Constitutional Stickiness 957 puts it, the Electoral College “was a brilliant eighteenth-century invention that makes no sense today.”327 In this second category, although a more optimal alternative exists, it would be inefficient to change the status quo. Put differently, the costs of adopting the superior alternative are higher than its benefits. Although the initial constitution-makers should, in hindsight, have chosen a different constitutional path, it would be too costly to alter it now. The suboptimal arrangement may therefore rationally persist. Even though change is inefficient in this second category, it still has significant normative implications. Although the status quo rationally persists, it would have been more optimal to adopt an alternative initially. In other words, the retention of the constitutional status quo should not necessarily imply a determination that the status quo is optimal. The acknowledgement that an existing provision is suboptimal, yet inefficient to change, can generate important signaling effects for both the domestic polity, as well as foreign polities searching other constitutions for provisions to borrow. In addition, the costs and benefits of constitutional change can fluctuate. Although the costs of changing the status quo may exceed the benefits today, that cost-benefit calculus may change in the future, permitting efficient constitutional alterations. Finally, recognizing that initial design choices can lock-in suboptimal configurations that would be inefficient to subsequently alter can force constitution-makers to pay closer attention to preliminary design choices. Although most nations already have written constitutions and therefore will not engage in the design process for the very first time, some nations still forge new constitutional paths by adopting new provisions with no roots in the old document. In those circumstances, the second category of stickiness carries significant potential normative weight. Third, constitutional stickiness may generate a suboptimal result that can be corrected efficiently, yet remain uncorrected. Unlike in the second category, here, the benefits of adopting a superior constitutional alternative exceed its costs and therefore change is efficient from a rational-choice perspective. But a suboptimal provision may persist because of a combination of cognitive limitations and biases that impede rational decision-making by constitution-makers. In this category, constitutional stickiness may prove to be a straitjacket that prevents modern generations from adequately confronting novel challenges or correcting serious 327 Akhil Reed Amar, A Constitutional Accident Waiting to Happen, 12 CONST. COMMENT. 143, 143 (1995). 958 University of California, Davis [Vol. 49:899 constitutional errors, even where it would be efficient to do so.328 For example, Egypt’s powerful presidency — which many commentators argued was at least partially to blame for the country’s authoritarian past — proved to be rather sticky and was retained in all constitutional makeovers that Egypt experienced following Hosni Mubarak’s overthrow.329 Even where constitutional stickiness prevents optimal changes, however, it can still produce some tangible benefits. Stickiness can reduce decision costs and promote consensus building by taking contentious questions off the bargaining table. As discussed above, all things being equal, the decision costs of retaining existing constitutional norms are lower than bargaining over novel provisions. In other words, constitution-makers are more likely to agree to retain existing constitutional norms than adopt novel provisions. The reduction of decision costs, in turn, can prevent endless, destructive conflict over constitutional choices that has the ability to derail the entire constitutional design enterprise, which, in some cases, may generate disastrous consequences. Stickiness can also promote constitutional stability and continuity. Constitutions often need to be grounded in historical institutions to succeed. Madison, for example, famously resisted Jefferson’s suggestion that “the earth belongs always to the living generation” and that the constitution should thus be rewritten by each successive generation.330 Madison argued that negotiation and lobbying in the drafting of each successive constitution may lead to factionalism and undermine continuity and stability.331 Constitutional stability, in turn, facilitates the settlement of disputes when they arise and encourages 328 In a separate paper, I analyze several strategies that constitution-makers can employ to diminish stickiness where it produces undesirable results, including vagueness, layering, logrolling, temporary constitutions, and non-constitutional means. See Ozan O. Varol, Remedying Constitutional Stickiness, 37 NAT’L J. CON. L. __ (forthcoming 2016). 329 See Bruce Ackerman, To Save Egypt, Drop the Presidency, N.Y. TIMES (July 10, 2013), http://www.nytimes.com/2013/07/11/opinion/to-save-egypt-drop-the-presidency.html. 330 Letter from Thomas Jefferson to James Madison (Sept. 6, 1789), available at http://press-pubs.uchicago.edu/founders/documents/v1ch2s23.html; Letter from James Madison to Thomas Jefferson (Feb. 4, 1790), available at http://press-pubs.uchicago. edu/founders/documents/v1ch2s24.html. 331 See Jacob E. Gersen, Temporary Legislation, 74 U. CHI. L. REV. 247, 254-55 (2007); David A.J. Richards, A Theory of Free Speech, 34 UCLA L. REV. 1837, 1840 (1987) (“Madison argued that the reflective values institutionalized in a properly designed republic are least compromised by oppressive democratic factions if the written constitution is understood on all sides as an enduring charter of just government subject to amendment only by extraordinary procedures.”). 2016] Constitutional Stickiness 959 beneficial reliance and investment, allowing relevant public and private actors to organize their conduct around existing constitutional configurations.332 Constitutional change has the potential to disrupt these expectations and cause socioeconomic turmoil.333 In some cases, it is better for a constitutional rule to be settled than to be settled correctly.334 As Aristotle argued, instability in law can undermine the rule of law since the citizens’ habits of obedience to a stable set of rules give law its power.335 Stability may also avoid frequent attempts to change the existing constitutional configurations, which may lead to deadweight losses.336 *** In sum, constitutional stickiness is not always undesirable. Although stickiness can prevent the adoption of superior alternatives, it also has the potential to generate tangible benefits in terms of consensus building, continuity, and stability. Whether these benefits are outweighed by the benefits of changing the constitutional status quo is a highly context-dependent question to be evaluated on a caseby-case basis. 332 See David Hume, Of the Original Contract, in DAVID HUME, ESSAYS: MORAL POLITICAL, AND LITERARY 465, 476 (Eugene F. Miller ed., Liberty Fund rev. ed. 1987) (“[A]s human society is in perpetual flux, one man every hour going out of the world, another coming into it, it is necessary, in order to preserve stability in government, that the new brood should conform themselves to the established constitution . . . .”); Elster, Constitutionalism, supra note 5, at 471 (“The constitution will lose many of its desirable properties — notably that of inspiring confidence and creating a climate in which investors are willing to make long-term investments — if everyone expects that it will be continually revised.”); David L. Shapiro, The Role of Precedent in Constitutional Adjudication: An Introspection, 86 TEX. L. REV. 929, 942 (2008). 333 See Robert N. Clinton, Original Understanding, Legal Realism, and the Interpretation of “This Constitution,” 72 IOWA L. REV. 1177, 1262-64 (1987) (noting that constitutional stability has “political, social, and economic importance” and is necessary to support societal and economic growth and the legitimacy of the government); Levinson, supra note 1, at 712 (“The absence of constitutional stability — leaving nothing but chaos, economic stagnation, civil war, and vulnerability to external conquest — will be enormously costly to most if not all.”). 334 See Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 406 (1932) (Brandeis, J., dissenting) (“[I]n most matters it is more important that the applicable rule of law be settled than that it be settled right.”). 335 ELKINS ET AL., supra note 7, at 17. 336 Boudreaux & Pritchard, supra note 81, at 126. 960 University of California, Davis [Vol. 49:899 CONCLUSION In the famous and oft-misunderstood poem, The Road Not Taken, Robert Frost describes a traveler’s choice between two roads that diverge in the woods. The poem is often misinterpreted as a testament to individualism and self-determination because, in the final lines of the poem, the traveler asserts that he took the road “less traveled by” and “that has made all the difference.” A close inspection of the poem, however, reveals important nuances that are often overlooked. The two roads, Frost writes, had been worn “really about the same” and both paths “equally lay in leaves no steps had trodden black.” In other words, neither path was more or less traveled, and the traveler’s choices were just about equal, his evaluation in hindsight notwithstanding.337 In addition, had the traveler chosen the path that was more traveled on, that also could have “made all the difference.”338 Like the traveler in Frost’s poem, constitution-makers often face divergent paths as they design constitutions. In this Article, I discussed how the taking of one path can reinforce further movement along the same path and foreclose others. A combination of rational-choice and behavioral mechanisms can lock-in the constitutional status quo, creating constitutional stickiness. Where the costs of constitutional change exceed the benefits, previous constitutional choices may persist despite rational behavior by constitution-makers. Constitutional designers can also fall prey to their own cognitive biases and, similar to the traveler in Frost’s poem, continue to follow the path they took, believing, in hindsight, that the chosen path was superior to all others. This, in turn, challenges the prevailing orthodoxy in the United States that Article V’s high threshold for constitutional amendment is the primary culprit for lack of formal constitutional change and that significant changes might follow if the threshold were lowered or a new constitutional convention were called. The Article also discussed the normative implications of constitutional stickiness. There is often no single optimal path in constitutional design. The constitutional roads ahead of the designers, though divergent, may be “really about the same.” Even where two constitutional roads differ, they may both lead to optimal outcomes for the polity. But where a suboptimal path locks in its travelers at the expense of more optimal 337 Matthew Hollis, Edward Thomas, Robert Frost and the Road to War, GUARDIAN (July 29, 2011), http://www.theguardian.com/books/2011/jul/29/robert-frost-edwardthomas-poetry. 338 Id. 2016] Constitutional Stickiness 961 alternative paths, constitutional stickiness can be a strong and unwelcome force to contend with in constitutional design. The study of constitutional stickiness carries major implications for contemporary constitutional theory. It suggests that constitutional scholarship has misplaced its focus on synchronic normative explorations of constitutional substance. In drawing attention to the significance of oft-neglected temporal and sequential elements in constitutional design, the study of constitutional stickiness promises to open new frontiers in constitutional theory.