Using Social Norms as a Substitute for Law

Bryan H. Druzin

This paper follows the law and norms literature in arguing that policymakers can use social norms to support or even replace regulation. Key to the approach offered here is the idea — borrowed from the folk theorem in game theory — that cooperative order can arise in circumstances where parties repeatedly interact. This paper proposes that repeated interaction between the same agents, specifically the intensity of it, may be used as a yardstick with which to gauge the potential to scale back regulation and use social norms as a substitute for law. Where there are very high levels of repeated interaction between people, policymakers can reduce regulation and then evaluate the emergent social order on a case by case basis. The contribution of the paper to the law and norms literature is that it proposes a practical technique to pinpoint the precise areas of social discourse where the possibility of using social norms as a substitute for law is most feasible and perhaps even more crucially — it highlights precisely where it is not.

SOCIAL NORMS AS A SUBSTITUTE FOR LAW Bryan H. Druzin* This paper follows the law and norms literature in arguing that policymakers can use social norms to support or even replace regulation. Key to the approach offered here is the idea—borrowed from the folk theorem in game theory—that cooperative order can arise in circumstances where parties repeatedly interact. This paper proposes that repeated interaction between the same agents, specifically the intensity of these interactions, may be used as a yardstick with which to gauge the potential to scale back regulation and use social norms as a substitute for law. Where there are very high levels of repeated interaction between people, policymakers can reduce regulation and then evaluate the emergent social order on a case-by-case basis. The contribution of the paper to the law and norms literature is that it proposes a practical technique to pinpoint the precise areas of social discourse where the possibility of using social norms as a substitute for law is most feasible—and perhaps even more crucially, it highlights precisely where it is not. INTRODUCTION There is a war of ideology between those who support an expansive role for government and those who wish to shrink it. Advocates of state minimalism—those who wish to shrink it—often speak about the ability of market forces to sustain social order and the normative * Assistant Professor and Deputy Director of LL.M. Programs, Faculty of Law, The Chinese University of Hong Kong. Ph.D. in Law, King’s College London; LL.M., LL.B., B.A., University of British Columbia. I am deeply indebted to Eric Posner at the University of Chicago; Richard Epstein at NYU; Bryan Mercuro and Eric Ip at the Chinese University of Hong Kong. My deepest appreciation also to Andrew Simester at the University of Cambridge for reviewing a much earlier draft of this paper. As well, the writing of this paper was helped greatly by Direct Grant CUHK funding. 67 68 Albany Law Review [Vol. 79.1 benefits in doing so.1 In this sense, the market is put forward as an alternative to government. This can be thought of as market-based minimalism. There is a vast literature (much of it heterodox) arguing that the functions of government may be provided by “the private sector” and the “market.”2 These theorists often adopt a fiercely ideological, strident tone in their condemnation of the state—rhetoric that can be quite off-putting. While these voices often downplay or simply discount the necessity of regulation,3 this does not, however, imply that there are no benefits to be had from reducing the intensity of legislation where possible. A crucial fact that must be understood is that the law is already minimalist: it does not seek to regulate every facet of human activity, nor could it. There exists a vast ocean of informal social ordering that goes unregulated by the state.4 This paper argues that we may go further in the direction of legal minimalism,5 and that, crucially, market-based minimalism is not the only game in town we can use to achieve this. It is possible to embrace an entirely different notion of minimalism, one that involves other kinds of “invisible hand” self-ordering. This broader vision is captured by the law and norms literature, which examines law’s relationship with social norms.6 Social ordering born from social norms is all around us, from the spontaneously self-assigned seating arrangements of students in a classroom, to the social rules of lining up, or the complex customary law of prepolitical societies.7 1 Bryan Druzin, Restraining the Hand of Law: A Conceptual Framework to Shrink the Size of Law, 117 W. VA. L. REV. 59, 66–67 (2014). 2 CHRISTOPHER W. MORRIS, AN ESSAY ON THE MODERN STATE 56 (1998). 3 See id. 4 This ocean of informal social order is the subject of the literature on legal pluralism. It may be better conceptualized as “normative pluralism”—semidiscrete normative orders that exist in the shadow of “official” state law. Indeed, “[n]ormative pluralism is an everyday experience for all of us (the rules of the road, of grammar, of our workplace, etc [sic]).” M.D.A. FREEMAN, LLYOD’S INTRODUCTION TO JURISPRUDENCE 926 (9th ed. 2014). 5 The term “legal minimalism” is found in the Postmodern Jurisprudence literature. See, e.g., Boaventura de Sousa Santos, The Postmodern Transition: Law and Politics, in THE FATE OF LAW 105, 112, 113 (Austin Sarat & Thomas R. Kearns eds., 1991) (discussing “the transition from maximal law to minimal law”). It is used here, however, in a different, narrower sense. 6 For a good summary of the early law and norms literature, see Robert C. Ellickson, Law and Economics Discovers Social Norms, 27 J. LEGAL STUD. 537, 537–38 (1998). While the law and norms literature is mostly entrenched in a homo economicus model of behaviour, it may be distinguished from purely market-based approaches to self-ordering. See id. at 539, 541–42 (explaining how too many economic theories can overlook the importance of self-interested individuals). 7 Social ordering of this kind has been defined as “normative order observed by a population, having been formed by regular social behavior and the development of an accompanying sense of obligation.” Gordon R. Woodman, A Survey of Customary Laws in Africa in Search of Lessons for the Future, in THE FUTURE OF AFRICAN CUSTOMARY LAW 9, 10 (Jeanmarie Fenrich et al. eds., 2011) (employing the term “customary law”). See also H. L. A. HART, THE CONCEPT OF 2015/2016] Social Norms as a Substitute for Law 69 Prominent legal theorists such as Robert Ellickson, Robert Cooter, Dan Kahan, Lawrence Lessig, and Cass Sunstein advocate using social norms as efficient alternatives to legal rules.8 As Richard Posner argues, social norms may be “both a source of law and often a cheap and effective substitute for law . . . .”9 Following the law and norms literature, this paper argues that policymakers can harness the energy of social norms in creating and sustaining social order.10 In contrast to market-based minimalism, we may think of this approach as norm-based minimalism. Norm-based minimalism as envisioned here comprises both strong and weak versions. These versions permit degrees of state intercession: from a total absence of regulation, to the codification of existing social patterns, to minor regulatory adjustments aimed at correcting inefficiencies, to traditional top-down law.11 In adopting such an approach, policymakers can take advantage of the natural patterning of social norms. There is abundant evidence that social norms can generate complex systems of cooperative order without the need for a centralized coercive authority. Ellickson’s pioneering work on cattle ranchers in Shasta County showed that agents who frequently interact will tend to create cooperative systems that in fact LAW 91–92 (3d ed. 2012) (discussing “custom” in relation to his primary rules). For a fantastic treatment of custom and law, see DAVID J. BEDERMAN, CUSTOM AS A SOURCE OF LAW 181 (2010). 8 See, e.g., ROBERT C. ELLICKSON, ORDER WITHOUT LAW: HOW NEIGHBORS SETTLE DISPUTES 1 (1991); Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. LEGAL STUD. 115, 115 (1992) (noting that the diamond industry has rejected traditional laws, and instead, use their own internal set of rules when handling disputes); Robert D. Cooter, Structural Adjudication and the New Law Merchant: A Model of Decentralized Law, 14 INT’L REV. L. & ECON. 215, 215–16 (1994) (noting that courts can find law through the form of community customs and norms); Dan M. Kahan, Social Influence, Social Meaning, and Deterrence, 83 VA. L. REV. 349, 350, 351 (1997) (suggesting that the relationship between the law and social influence could be utilized to aid in crime deterrence); Lawrence Lessig, The New Chicago School, 27 J. LEGAL STUD. 661, 666 (1998) (suggesting that norms have the power to offer a wider range of regulatory power than just laws alone); Cass R. Sunstein, Social Norms and Social Roles, 96 COLUM. L. REV. 903, 907 (1996) (suggesting that norms could help further the objectives of laws). See also Richard H. McAdams, The Origin, Development, and Regulation of Norms, 96 MICH. L. REV. 338, 340 (1997) (noting the various aspects of the relationship between the law and norms). But see ERIC A. POSNER, LAW AND SOCIAL NORMS 7–8 (2000) (cautioning against such approaches). However, it should be noted that many of the law and norm scholars are better understood as advocating using social norms merely as support for law rather than as endorsing minimalism. Nevertheless, many of their arguments can be marshaled to support a minimalist approach to law. 9 Richard A. Posner, Social Norms and the Law: An Economic Approach, 87 AM. ECON. R. 365, 365 (1997). 10 Throughout, I use the terms “custom,” “customary order,” “customary law,” “bottom-up order,” and “social norms” interchangeably. 11 Throughout, top-down law is contrasted with bottom-up law or bottom-up order—that is, normative order not produced formally under the auspices of the state. 70 Albany Law Review [Vol. 79.1 maximize the aggregate welfare of the group.12 Agents, he explains, who “repeatedly interact can generate [legal] institutions through communication, monitoring, and sanctioning.”13 Informal social norms are perfectly capable of producing all three of the core functions of law: dispute resolution, rule formation, and enforcement.14 Correctly harnessed, bottom-up social order can be tremendously useful in that it can lighten the legislative and enforcement burden on the state.15 Social norms do not need to be legislated or enforced because they are self-producing and often highly robust, internalized,16 and self-enforcing.17 Norm-based minimalism capitalizes on this. Social norms are like an untapped resource. Policymakers can exploit this resource, letting the natural emergence of social order do much of, or in some cases, even all of the heavy lifting. Yet this is not possible in many areas of law. Social norms are not always able to produce stable ordering. Moreover, even where social norms can generate bottom-up order, this social patterning may be massively inefficient. As such, we need to know exactly where there is at least the potential to utilize social norms and where there is not. While the law and norms literature is rich, it has yet to articulate a method by which to clearly identify the precise areas of law predisposed to such an approach. This paper proposes such a method. Borrowing from game theory, this paper argues that repeated interaction between the same agents, specifically the intensity of it, 12 See ELLICKSON, supra note 8, at 3–4 (“The end reached is exactly the one . . . predicted: coordination to mutual advantage without supervision by the state.”). 13 Robert C. Ellickson, Property in Land, 102 YALE L.J. 1315, 1366 (1993); see also Elinor Ostrom et al., Covenants With and Without a Sword: Self-Governance is Possible, 86 AM. POL. SCI. REV. 404, 405 (1992) (outlining empirical support). 14 As Robert Cooter opines, the “utilitarianism of small groups has been demonstrated for cattle ranchers, Chinese traders, medieval merchants, and modern merchant associations. Research on property rights has revealed variety and detail in the political arrangements by which small groups manage their assets.” Robert Cooter, Normative Failure Theory of Law, 82 CORNELL L. REV. 947, 950 (1997). 15 Indeed: [I]t is widely held that strong social norms reduce the burden on law enforcement; that laws supported by social norms are likely to be significantly more enforceable; and that laws that are formulated in ways that are congruent with social norms are much more likely to be enacted than laws that offend such norms. Amitai Etzioni, Social Norms: Internalization, Persuasion, and History, 34 L. & SOC’Y REV. 157, 159 (2000). 16 That is, participants feel emotionally obliged to observe these norms. Internationalization is arguably the most socially powerful component of normative order. For a deeper discussion of social norms and internalization, see id. at 159–60. 17 See id.; Francesco Parisi, Spontaneous Emergence of Law: Customary Law, in 5 ENCYCLOPEDIA OF LAW AND ECONOMICS: THE ECONOMICS OF CRIME AND LITIGATION 603, 611 (Boudewijn Bouckaert & Gerrit De Geest eds., 2000). 2015/2016] Social Norms as a Substitute for Law 71 may be used as a yardstick with which to gauge the potential to scale back regulation, and allow social norms to shoulder more of the burden of creating and sustaining social order. To this end, policymakers may look to whether an area of law involves preexisting relationships of repeated interaction between the same actors, using it as a heuristic to pinpoint where norm-based minimalism is most viable and where it is not. The idea that repeated interaction can generate cooperation is known in game theory as the folk theorem.18 Cooperation, however, is only a possible result. The presence of repeated play is hardly a guarantee that self-sustaining cooperative equilibria will emerge or, perhaps even more importantly, that such equilibria, where they do emerge, will be normatively just and not simply entrench a social imbalance in power.19 Where the emergent order is unjust or grossly sub-optimal, the state has a vital role to play in pushing parties toward one equilibrium over another.20 The characteristic of repeated interaction between the same individuals is useful in that it indicates the possibility of stable and efficient ordering.21 The less an area of law possesses this quality, the weaker its ability to self-order.22 Where there are high levels of repeated play, policymakers can scale back regulation and then evaluate the consequences on a case-bycase basis. Where results prove sub-optimal, regulation can simply be reintroduced to whatever degree necessary. What distinguishes the present thesis from the prior law and norms literature is the concrete technique and taxonomy of law it provides. The paper proposes a simple, yet reliable, technique to pinpoint the precise areas of social discourse where the possibility of decreasing regulation is most feasible. Law regulates a wide spectrum of human activity: some of this activity entails repeated interaction and some of it does not. The paper identifies the areas of law where cooperative order might arise without, or with minimal state involvement. However, equally as important, the paper identifies precisely where such an approach is not possible. While the discussion deals with the utilization of social norms to ease the legislative and enforcement burden on the state, the 18 For a good summary of the Folk theorem, see KEN BINMORE, GAME THEORY: A VERY SHORT INTRODUCTION 75–79 (2007). 19 See supra Part V.B. 20 See supra text accompanying note 142. 21 See supra text accompanying notes 57–58. 22 See supra text accompanying note 108. 72 Albany Law Review [Vol. 79.1 ideological argument for minimalism is not litigated here.23 This is not because the matter is closed to debate—far from it. Rather, I do this because the ideological case for minimalism is not the paper’s focus. The aim of the discussion is merely to articulate a technique policymakers may use to gauge the possibility of scaling back regulation. Whether doing so aligns with our broader social values is another matter. For advocates of regulatory minimalism, the discussion provides a practical tool to reduce regulation. For those unsympathetic to the idea, the discussion will still be of interest in that it clarifies where minimalism is simply not possible. A good way to think of this paper is as a kind of conceptual roadmap. Like the pathways on a map, the potential of norm-based minimalism is explored and its limits are carefully charted. The paper proceeds as follows. Part I lays out a basic model for norm-based minimalism. Part II discusses the importance of repeated interaction in the emergence of bottom-up social order. Part III applies this, differentiating areas of law with respect to the degree of repeated interaction between private actors implied by those areas (if at all). Part IV then details how the presence of repeated interaction may serve as a guide to policymakers in assessing in what situations norm-based minimalism may be most feasible. Part V then explores some likely objections to the model. I. TOWARDS A MODEL OF NORM-BASED MINIMALISM In the literature, theorists do not usually distinguish clearly between market-based and norm-based minimalism.24 To some extent the distinction is present in the two antithetical wings of the anarchist tradition: what can be thought of as libertarian anarchism, which has gained recent intellectual ascendency, particularly in the United States, and socialist anarchism, which has been more predominant in Europe.25 While libertarian anarchism looks to the market to sustain stateless order, the latter emphasizes the importance of community and social forces in the formation of social order.26 While socialist anarchism has not attracted much interest in many decades, libertarian anarchism, on the other hand, has enjoyed 23 The ideological argument relates to issues of individual liberty and the implicit injustice of state encroachment on this liberty. See Cass R. Sunstein, Beyond Judicial Minimalism, 43 TULSA L. REV. 825, 825 (2008). 24 Henceforth, where the term “minimalism” is used in isolation, it is meant to connote normbased minimalism as opposed to market-based minimalism. 25 MORRIS, supra note 2, at 61. 26 Id. at 61–62. 2015/2016] Social Norms as a Substitute for Law 73 a renaissance of late.27 In this view, as is the proclivity of economists, society is conceptualized as a vast market and all forms of human interaction are simply tossed into the bin of economic discourse.28 But this is not entirely accurate. While extremely useful, as with most economic models, this is an overly-simplified paradigm. The problem is much of human society is not accurately approximated by microeconomic models.29 The market is just one example of a selfordering system (albeit one with properties that give it significant heft). The idea of self-ordering (known as self-organization in the hard sciences) is far more expansive.30 Self-ordering systems exist everywhere, from the formation of migrating birds to the highlyorganized crystalline structure of snow. For our purposes, differentiating between market-based and normbased minimalism is important because most areas of social discourse in fact lack many of the technical properties of a “market” as defined in mainstream economics (i.e. enforceable contracts, clearly-articulated property rights, a market-based pricing mechanism, or trade in the formal sense) yet we see the emergence of highly complex, robust systems of social order bubbling up with stunning regularity. The core idea of market-based minimalism is that society will self-order around market principles.31 There may be a case for this and many right-of-center libertarians make it; however, this formal market dynamic is not always present and yet there exists a massive amount of bottom-up ordering in our social arrangements. Thus, if we can hone in on what fosters this social Id. at 61, 74. Id. at 63. 29 It should be noted that the advocates of market-based minimalism come mostly from heterodox economic schools, such as the Austrian school of economics. See Druzin, supra note 1, at 71. 30 There is a lot of scientific work regarding self-organization, for example Heinz von Foerster’s second generation cybernetic model, HEINZ VON FOERSTER, Ethics and Second-Order Cybernetics, in UNDERSTANDING UNDERSTANDING: ESSAYS ON CYBERNETICS AND COGNITION 287, 303 (2003); Varela and Maturana’s theory of “autopoiesis,” HUMBERTO R. MATURANA & FRANCISCO J. VARELA, AUTOPOIESIS AND COGNITION, at xviii–xix (D. Reidel Publishing Company 1980) (1972); and Ilya Prigogine’s thermodynamics of open systems and dissipative structures, G. NICOLIS & I. PRIGOGINE, SELF-ORGANIZATION IN NON-EQUILIBRIUM SYSTEMS: FROM DISSIPATIVE STRUCTURES TO ORDER THROUGH FLUCTUATIONS 19, 24, 160 (1977). Swarm intelligence is a fascinating concept related to self-organization in the biological realm. Inspired by the collective “intelligence” of self-ordering systems such as ant and bee colonies, the concept is being applied in artificial intelligence research. ERIC BONABEAU ET AL., SWARM INTELLIGENCE: FROM NATURAL TO ARTIFICIAL SYSTEMS, at xi–xii, 1 (1999). Very much related to this is the concept of emergence in philosophy, systems theory, complexity theory, and science. For a good, thorough examination of the concept, see JOHN H. HOLLAND, EMERGENCE: FROM CHAOS TO ORDER 11–12 (1998). 31 See Druzin, supra note 1, at 70–71. 27 28 74 Albany Law Review [Vol. 79.1 ordering, we can build a model for minimalism around an entirely different principle. The principle in short is this: it has been wellestablished in game theory that cooperative social ordering can arise in situations where there exists sufficient repeated interaction between people.32 This is the well-known folk theorem in game theory. Drawing on this basic insight, the paper sets forth a framework for norm-based minimalism based on the intensity of repeated interaction between people. Law regulates a wide spectrum of human activity: some of this activity entails repeated interaction between individuals and some of it does not. Where there is repeated interaction (and the ability to monitor the behavior of other people), the folk theorem suggests it is possible to scale back formal regulation and allow social norms to take up some or all of the slack of creating and sustaining social order.33 Conversely, in areas of law that do not involve repeated interaction between people, minimalism is not possible—robust regulation is required to sustain social order.34 Thus, using this principle as a yardstick with which to measure the built-in potential for self-ordering, the model identifies the areas of law where norm-based minimalism may succeed and where it cannot. Such an understanding may be extremely useful for those who wish to minimize regulation—that is, those who advocate state minimalism. Michael Taylor has posited a similar thesis regarding the potential of repeated interaction to fashion stateless social order, what he calls “community.”35 Yet Taylor concludes that a stateless society is only possible in “small and stable communities.”36 Unfortunately, this is not the world in which we live.37 The present thesis adopts a more 32 In game theory, this has been extensively studied in the context of iterated games which solve the prisoner’s dilemma. The economics literature alone is extremely large (not to say anything of its application more broadly in the social sciences), thus I refer the reader to the foundational work regarding this idea. See generally Robert Axelrod & William D. Hamilton, The Evolution of Cooperation, 211 SCI. 1390, 1390, 1392 (1981) (outlining various theories of competition and cooperation, including those which emerge from repeated iterations of the prisoner’s dilemma) [hereinafter Axelrod & Hamilton]; Robert Axelrod, The Emergence of Cooperation Among Egoists, 75 AM. POL. SCI. REV. 306, 307 (1981) (explaining the cooperation will result if the players believe an indefinite number of interactions will follow). 33 BINMORE, supra note 18, at 79 (explaining that folk theorem indicates that an external enforcement agency is not necessarily needed to foster cooperation in repeated interactions). 34 Id. (explaining that folk theorem requires repeated interaction and perfect observation of the players involved in order to render external enforcement unnecessary). 35 See generally MICHAEL TAYLOR, COMMUNITY, ANARCHY AND LIBERTY 2–3 (1982) (arguing essentially that spontaneous cooperation emerges in repeated interactions). 36 Id. at 94. 37 Indeed, Taylor himself seems to have little optimism regarding the implementation of his 2015/2016] Social Norms as a Substitute for Law 75 practical turn, arguing that this insight may still be of great use even in a large fragmented society. It applies this idea to identifying pockets of human discourse presently subjected to regulation that involve people repeatedly interacting with one another. Armed with this insight, regulation in these areas may be selectively scaled back in a strategic fashion, allowing bottom-up ordering to take the reins. Because not all social behavior flawlessly tracks a market model, we cannot count on a market-based model as a universal foundation for minimalism. However, provided certain structural dynamics are present, what we can count on is the force of social norms. The paper’s thesis contributes to a much broader conversation regarding the perceived over-intrusiveness of government regulation—a subject of intense debate both in Europe and the United States38—by offering a pragmatic solution to the “problem” of the state that may be realistically implemented. A deeper discussion on repeated interaction is provided in the following section. Here, it is sufficient to simply note its central role in our thesis. The purpose of the remainder of this section, rather, is to unpack the concept of social order driven by social norms. The idea of this kind of social ordering is extremely counterintuitive. Conventional wisdom suggests that order needs to be designed and imposed, if not through the vehicle of formal law, then at least by some other kind of authority. Yet theorists from a broad range of disciplines reject this Hobbesian hypothesis, asserting that social order may evolve in the absence of the State.39 The legal centrist40 belief that social order is to be conceived of in terms of external coercion “has become so ingrained in the popular imagination that the position has become almost axiomatic—almost own theory. Id. at 139. 38 See, e.g., Over-regulated America, THE ECONOMIST, Feb. 18, 2012, at 9 (providing examples of overregulation in the United States); Mary Ellen Synon, Face the Facts Europe is Going Bust, DAILY MAIL, Feb. 20, 2006, at 14 (noting that too much regulation is affecting the European economy). 39 While nowhere near a comprehensive list, in the field of law, see ELLICKSON, supra note 8, at 4–5; Cooter, supra note 8, at 215–16. In economics, see 1 F. A. HAYEK, LAW, LEGISLATION AND LIBERTY: RULES AND ORDER 2–3 (1973). See also MURRAY N. ROTHBARD, MAN, ECONOMY, AND STATE: A TREATISE ON ECONOMIC PRINCIPLES 1024 (2d ed. 2009) (concluding that while it may seem as though intervention by the government imposes order within the community, it is not necessarily always the case). 40 For an overview of the concept of legal centrism including evidence that refutes this belief, see generally ELLICKSON, supra note 8, at 138–47 (highlighting the tradition of legal centrism as well as refutable evidence of the belief). E.g., OLIVER E. WILLIAMSON, THE ECONOMIC INSTITUTIONS OF CAPITALISM: FIRMS, MARKETS, RELATIONAL CONTRACTING 20–21 (1985) (offering a brief discussion on the tradition of legal centrism). 76 Albany Law Review [Vol. 79.1 unassailable in its legitimacy.”41 But it is simply not true. Most social order is not created through an overarching authority—it arises spontaneously. Indeed, some of the most sophisticated forms of social order on earth are found among the social insects: the wasps, the ants, the bees.42 And so it is with much human social order. The idea that complex systems of social order may be self-generating is captured by the concept of spontaneous order.43 The economist Friedrich Hayek wrote extensively on the idea. He argued that there are two ways in which order may originate: “made” and “grown” order.44 By “made” order, Hayek meant order that was imposed from above by some hierarchical overlord promulgating rules: a state, a monarch, a tribal leader, etc.45 By “grown” order, he meant order that arises spontaneously through individuals’ actions, yet not as the result of any centrally-planned coordination.46 It is not difficult to find earlier traces of the idea of spontaneous order in the theories of Adam Smith, David Hume, Adam Ferguson, and Edmund Burke.47 Of these, perhaps Smith is best known for advancing this position. Smith famously posits a theory of spontaneous order—an “invisible hand” that guides the market place.48 Top-down regulation of social order can be problematic. In its effort to order society, the state often over-regulates, arguably creating more harm than good in the process. An ocean of law has flooded into the social relationships and institutions within which people live their lives.49 We arguably live in a more legalistic society than at any point in human history. This is perhaps most obvious in the case of 41 Bryan H. Druzin, Opening the Machinery of Private Order: Public International Law as a Form of Private Ordering, 58 ST. LOUIS U. L.J. 423, 430 (2014). 42 THEORIES OF SOCIAL ORDER 3 (Michael Hechter & Christine Horne eds., 2d ed. 2009). 43 The term “spontaneous order” is usually used to describe social and economic selfordering, while “self-organization” (mentioned previously) is typically reserved for the emergence of order in systems of a physical or biological nature. 44 HAYEK, supra note 39, at 37 (“The grown order, on the other hand, which we have referred to as a self-generating or endogenous order, is in English most conveniently described as a spontaneous order.”). Similarly, Lon Fuller distinguishes between what he calls “horizontal forms of order” and “vertical dimension” imposed by the State. LON L. FULLER, THE MORALITY OF LAW 233 (16th prtg. 1979). 45 See HAYEK, supra note 39, at 37. 46 Id. 47 Norman Barry, The Tradition of Spontaneous Order, 5 LITERATURE LIBERTY 7, 8, 9, 21, 23, 25 (1982), http://lf-oll.s3.amazonaws.com/titles/1303/0353-18_1982v2_Bk.pdf; Timothy Sandefur, Some Problems with Spontaneous Order, 14 INDEP. REV. 5, 7 (2009). 48 1 ADAM SMITH, THE WEALTH OF NATIONS 400 (J.M. Dent & Sons Ltd. 1981) (1910) (“[H]e is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention.”). 49 But see Mila Sohoni, The Idea of “Too Much Law”, 80 FORDHAM L. REV. 1585, 1586–87, 1591 (2012) (disputing the notion that regulations have grown too numerous). 2015/2016] Social Norms as a Substitute for Law 77 overcriminalization in the criminal law.50 However, a strong case could be made that this problem extends to all corners of law and across many jurisdictions: a general spirit of regulatory intervention prevails. It is only that with the case of overcriminalization that the problem has grown so extreme that it has drawn attention.51 From a public policy perspective, value could be gleaned by scaling back the degree of legal intervention and allowing bottom-up ordering to unfold unhindered. Yet we must be careful to not oversimplify the matter. It would be naïve to assume that relying upon natural ordering to sustain social order is always realistic. The trick is in knowing where and to what extent we can safely defer to the ordering force of social norms, and where we need to impose legal order through the instruments of formal regulation. Knowing the structural limitations of minimalism is thus vital for our project. As one scholar so insightfully points out: “Law is not the foundation of social order but a remedy for the deficiencies of custom.”52 We need to learn how to more deftly use this remedy. To the uninitiated, law may seem like a relatively straightforward affair: simply write down the rules and then apply them. Yet to those of us who actually create, apply, practice, or study the law, reality is far messier. The law is a highly complex, continually evolving system of inter-related principles and rules. Because the state assumes a monopoly on coercive authority, these rules typically proceed “top down,” however, this need not always be the case. The question we need to ask is what stimulates and sustains bottom-up order? For this we need to understand the important role that repeated interaction plays in this process. II. THE IMPORTANCE OF REPEATED INTERACTION IN THE EMERGENCE OF BOTTOM-UP ORDER The key to norm-based minimalism is not the machinery of the market—rather it is the principle of repeated interaction. Agents that repeatedly interact naturally generate rules to regulate their 50 See SANFORD H. KADISH, BLAME AND PUNISHMENT: ESSAYS IN THE CRIMINAL LAW 21 (1987). There is a growing body of very interesting scholarship within criminal law in relation to the problem of overcriminalization, for a good introduction to this literature, see id. at 21– 61. 51 See id. at 21. 52 James Bernard Murphy, Habit and Convention at the Foundation of Custom, in THE NATURE OF CUSTOMARY LAW 53, 76 (Amanda Perreau-Saussine & James Bernard Murphy eds., 2007). 78 Albany Law Review [Vol. 79.1 interactions.53 This may spawn a market dynamic, but this is not necessary. As already discussed, game theorists have widely noted that repeated interaction can induce cooperative order (the folk theorem).54 This holds true across a broad range of social dynamics captured in game theory.55 The paradigmatic illustration of the potential for stable cooperation to emerge even in the face of chaotic uncertainty and a paucity of trust is the well-studied prisoner’s dilemma. An absence of repeated interaction in the prisoner’s dilemma invariably drives the participants towards equilibriums of noncooperation and disorder, yet, with the mere introduction of repeated play, the same game is capable of producing stable cooperative order.56 Repeated interaction allows for the possibility of sophisticated forms of coordination because the shadow of future encounters can foster a cooperative equilibrium.57 In real-life settings, even those that resemble the prisoner’s dilemma, social relationships are typically open-ended and so the mere possibility of future interaction helps nurture bottom-up order.58 Even where future interaction is certain to not occur, the reflexive behaviour of individuals can go far in inducing the emergence of order. Human behaviour is not always perfectly calculated: Because we are so habituated to repeatedly interacting with others, cooperative behaviour is largely ingrained, and thus, often arises reflexively even where there is no possibility of future interaction, that is, even in oneshot interactions. Indeed it is very difficult to imagine a life where one encountered all other persons just once, never experiencing a second interaction. Repeated interaction is the lynchpin to stable bottom-up social order. Indeed, many contend that repeated games provide the foundational constituent of social order.59 Various mechanisms permitted by repeated interaction help sustain informal order. As a stick, individuals can rely on the threat of retaliation and 53 See Robert D. Cooter, Decentralized Law for a Complex Economy: The Structural Approach to Adjudicating the New Law Merchant, 144 U. PA. L. REV. 1643, 1646–47 (1996). The dynamic is fleshed out more cleanly in terms of game theory infra. 54 See supra note 18 and accompanying text. 55 See generally GEORGE J. MAILATH & LARRY SAMUELSON, REPEATED GAMES AND REPUTATIONS: LONG-RUN REPUTATIONS 1–8 (2006) (discussing the application of repeated interactions in regards to prisoner’s dilemma, oligopolies, and the product-choice game). 56 See Robert Axelrod, The Evolution of Cooperation, in THEORIES OF SOCIAL ORDER, supra note 42, at 175, 178, 179. 57 See ROGER B. MYERSON, GAME THEORY: ANALYSIS OF CONFLICT 308, 309 (1991). 58 See id. 59 INTERNATIONAL ENCYCLOPEDIA OF ECONOMIC SOCIOLOGY 299 (Jens Beckert & Milan Zafirovski eds., 2006). 2015/2016] Social Norms as a Substitute for Law 79 reputational costs as ex post enforcement mechanisms to promote rule compliance.60 The carrot of mutual gain can animate cooperative relationships, strengthening the social rules that emerge.61 The crucial constituent of monitoring the behaviour of others is possible where there is repeated interaction. Overtime, these patterns of social order usually become internalized, strengthening their dominion.62 All of this has been extensively studied by game theorists. Game theorists focus specifically on the emergence of cooperation.63 For our purposes, we may conceptualize this as rulesystems for where there is stable cooperation; by definition this means there are rules of a formal or informal nature. That repeated interaction disciplines social behavior is actually not that extraordinary. It happens all the time, and there is nothing particularly magical about it. It is common sense: the more people regularly interact, the more fixed standards of behavior emerge. In the case of small groups, we see the folk theorem at work quite clearly. In small, homogenous groups where there is a high level of repeated interaction, natural enforcement mechanisms can often be counted on to foster and sustain social order.64 Indeed, it is important to note that these relationships need not be binary relationships between just two individuals. They may include multiple actors 60 See Thomas C. Scott-Phillips, Evolutionary Stable Communication and Pragmatics, in LANGUAGE, GAMES, AND EVOLUTION: TRENDS IN CURRENT RESEARCH ON LANGUAGE AND GAME THEORY 117, 125 (Anton Benz et al. eds., 2011). 61 See id. 62 Such social order is internalized where people “feel an emotional or psychological compulsion to obey the norms . . . .” Eric A. Posner, Law, Economics, and Inefficient Norms, 144 U. PA. L. REV. 1697, 1708, 1709 (1996). 63 MARTIN J. OSBORNE & ARIEL RUBINSTEIN, A COURSE IN GAME THEORY 133 (1994). Although there is no guarantee that the equlibria that emerges from repeated interaction will be optimal in terms of social welfare, our basic notions of justice, and so forth, the idea that repeated interaction can produce stable equilibria in games of various kinds is extensively written upon. For a concise and not overly-technical overview, see id.at 133–62. 64 This has been written on extensively. For this idea, see, for example, ELLICKSON, supra note 8, at 123 (discussing the cooperation in close-knit communities); Bernstein, supra note 8, at 115 (discussing the success of an internal set of rules in regulating the diamond industry); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms and Institutions, 99 MICH. L. REV. 1724, 1724, 1725 (2001) (noting the fluid functioning of the cotton industry under a mostly private system of laws since the mid-1800’s); Karen Clay, Trade Without Law: Private-Order Institutions in Mexican California, 13 J. L. ECON. & ORG. 202, 202, 203 (1997) (finding that trade coalitions in California flourished in the 1800s under private regulations); David Friedman, Private Creation and Enforcement of Law: A Historical Case, 8 J. LEGAL STUD. 399, 400 (1979) (noting the survival of medieval Icelandic institutions where law enforcement was mostly done privately); Avner Greif, Reputation and Coalitions in Medieval Trade: Evidence on the Maghribi Traders, 49 J. ECON. HIST. 857, 859 (1989) (discussing how traders benefitted from being part of an organization that was selfregulated). 80 Albany Law Review [Vol. 79.1 loosely interacting with each other. While bottom-up order is produced most robustly by two parties repeatedly interacting, repeated interaction within small groups is often sufficient to produce strong social norms.65 In such cases, formal law plays a less important role. However, in larger groups where participants do not repeatedly interact, the legislative and enforcement mechanisms of formal law are critical to maintaining social order.66 Frequently repeated interaction of a binary nature will often yield the greatest capacity to produce stable bottom-up order built on strong social norms, because the intensity of repetition can be very high.67 All the necessary ingredients—monitoring, reciprocity, retaliation—are robustly present.68 As numbers grow, this becomes less the case. Small groups also demonstrate this capacity; however, because repeated interaction is less intense, this is to a less robust extent.69 Very large groups that experience virtually no repeated interaction between the same agents, however, fare the worse.70 They lack this capacity—only custom of an extremely weak kind (generated by scattered pockets of repeated interactors) will emerge.71 We must be careful to distinguish custom from more robust and comprehensive forms of normative ordering. Custom is best understood as the inchoate traces of order, like wisps of smoke floating at a distance from a raging fire.72 Extending this logic, this paper simply points out that some areas of social activity regulated by law capture this crucial dynamic in that they possess a high degree of repeated interaction between the same agents, while other areas of law do not exhibit any degree of it, or markedly less so. Repeated interaction is a powerful engine of bottom-up social ordering. If repeated interaction is not present, See supra note 64 and accompanying text. Bryan H. Druzin, Planting Seeds of Order: How the State Can Create, Shape, and Use Customary Law, 28 BYU J. PUB. L. 373, 384 (2014). Social dilemma research clearly shows that cooperation decreases in large groups. See, e.g., David De Cremer & Geoffrey J. Leonardelli, Cooperation in Social Dilemmas and the Need to Belong: The Moderating Effect of Group Size, 7 GROUP DYNAMICS THEORY RES. & PRAC. 168, 169 (2003) (noting that past research has found that cooperation goes down in larger groups); Norbert L. Kerr, Illusions of Efficacy: The Effect of Group Size on Perceived Efficacy in Social Dilemmas, 25 J. EXPERIMENTAL SOC. PSYCHOL. 287, 309 (1989) (stating that cooperation decreases in large groups). 67 Druzin, supra note 66, at 377, 380–81. 68 See Axelrod & Hamilton, supra note 32, at 1395; Druzin, supra note 66, at 381–82; Ellickson, supra note 13, at 1366. 69 See Druzin, supra note 66, at 382–83. 70 Id. at 383–84. 71 For a more a more detailed discussion of the ability of small groups to produce stable social ordering vis-à-vis larger groups, see id. at 382–84. 72 See infra Part V.B for a discussion about custom. 65 66 2015/2016] Social Norms as a Substitute for Law 81 robust bottom-up order simply cannot arise. This simple insight is useful because we can use it to construct a model for minimalism that may guide legislators. III. LAW MAY BE DISTINGUISHED WITH REFERENCE TO THE EXTENT THAT IT INVOLVES REPEATED INTERACTION BETWEEN PARTIES Having discussed the importance of repeated interaction in terms of fostering informal rule-systems, we are now in a position to consider the conceptual implication that flow from this. The folk theorem asserts that cooperative social order can arise where there exists repeated interaction between the same agents.73 The fact that repeated interaction maximizes the potential for self-ordering—as the folk theorem establishes—introduces an important conceptual distinction. Law may be classified with reference to the extent that it involves repeated interaction between parties. Some forms of law relate to areas of behaviour that concern existing relationships of repeated interaction between the same agents.74 Other areas of law only tangentially so, or do not concern interaction between individuals at all.75 Where there is repeated interaction between agents, the folk theorem can be applied; where there is no interaction, it cannot.76 This implies that certain areas of law are categorically more predisposed to bottom-up ordering than others. This distinction is broadly captured by the civil law division between private and public law. Private law involves relationships between individuals; public law concerns individuals’ relationship with government.77 In the case of private law, where these relationships are repeated interactions between the same agents, it is possible to apply the folk theorem and all that it entails.78 Areas of public law, however, do not allow for the application of the folk theorem in that such areas do not involve interactions between private parties, let alone repeated interactions.79 This allows us to See, BINMORE, supra note 18, at 79. See HILAIRE BARNETT, UNDERSTANDING PUBLIC LAW 1 (2010). 75 See id. 76 See BINMORE, supra note 18, at 76–79 (providing an example of how the folk theorem is applicable to individuals). 77 BARNETT, supra note 74, at 1. 78 See BINMORE, supra note 18, at 79 (explaining how the folk theorem is successful when applied repeatedly to individuals). 79 See id. at 77–79; see also BARNETT, supra note 74, at 1 (explaining that public law does not cover the regulation of private individuals). 73 74 82 Albany Law Review [Vol. 79.1 hone in on which areas of law that are most hospitable to norm-based minimalism. We can make the general claim that public law precludes the application of the folk theorem because it does not involve interaction between individuals, and private law allows for it because it does involve interaction between individuals. In this respect, the distinction between private and public law is tremendously useful. Yet while this classification applies generally, it is not perfect. For example, some areas of public law actually relate indirectly to interactions between private parties. This is the case with, for example, the criminal law and aspects of constitutional law.80 Making this more confusing, while private law concerns interactions between individuals, these interactions are often not repeated, rendering such areas completely unreceptive to the folk theorem. Interaction must be repeated—the more frequently the better.81 Thus, the question we need to ask is really this: to what degree does a specific area of law involve relationships of frequent and repeated interaction between the same actors? The more intensely this is the case, the more this area of law will be receptive to the folk theorem. This is really a matter of degree and so is best conceptualized as a continuum rather than a sharp divide. Understanding this as a continuum rather than in stark binary terms is elaborated upon below. For now, however, it is sufficient to merely note that areas of law can be distinguished generally in terms of the degree of repeated interaction between the same agents they involve. It may be useful at this point to import new terms into the discussion. I will refer to the extent that they may be characterized by frequent and repeated interaction between the same agents as how interactive the area of law is.82 Highly interactive law relates to law that regulates existing relationships of interaction between individuals that are frequently repeated.83 Examples would include many commercial relationships, familial relationships, long-term contractual relationships, work relationships, and other such social arrangements that relate to frequent repeated interaction between the same parties. These represent pockets of social discourse that involve the same people repeatedly interacting with one another. This can be clearly See infra notes 112–15 and accompanying text. Interaction is a necessary, not a sufficient, condition. 82 I use the terms interactive because it relates to the degree that agents are interacting with each other. 83 See BARNETT, supra note 74, at 1 (noting areas of law governing relationships between individuals). 80 81 2015/2016] Social Norms as a Substitute for Law 83 distinguished from areas of law that do not concern relationships between people and so do not directly relate to interaction (let alone repeated interaction)—for example, paying income tax and the procedures to obtain a patent. Because these areas of law do not directly relate to repeated interactions between the same parties, the folk theorem and all it entails simply cannot track this dynamic. Areas of law such as immigration or tax law are not interactive. Indeed, such areas of law have no (or only oblique) connection to interactions between private parties.84 Because in very large groups there is no repeated interaction (or very little of it), law that does not possess this interactive character needs to be centrally created.85 Thus, the state (or some centralized authority) is required to create and sustain a great deal of social order in large societies. With giant masses of disconnected parties that do not repeatedly interact, selfordering is simply not possible.86 The state is necessary to step in and create legal order.87 Without the state (or some version of centralized power), such systems languish perpetually within a condition of social disorder unable to self-organize.88 For law that is not interactive, the state is required to formulate these rules; such rules cannot otherwise ever get off the ground.89 However, this does not mean that the state is necessary to create all forms of social order. That areas of law can be distinguished with reference to how interactive they are is theoretically significant because it leads to the conclusion that certain areas of law are more predisposed to the emergence of bottom-up ordering than others. This, of course, has obvious implications for minimalism. It provides us with an entirely 84 There are exceptions of course, for example, sales tax in commercial transactions between individuals, etc. But, it is not controversial to say that tax law, for the most part, concerns the individual’s relationship with the State rather than with other individuals. People’s relationships with each other are merely dragged into this state-individual regulation. 85 See Bryan Druzin, Law Without the State: The Theory of High Engagement and the Emergence of Spontaneous Legal Order Within Commercial Systems, 41 GEO. J. INT’L L. 559, 587 (2010) (noting that some laws infrequently have repeat involvement, thus necessitating state involvement). 86 See S. R. EPSTEIN, FREEDOM AND GROWTH: THE RISE OF STATES AND MARKETS IN EUROPE, 1300-1750, at 8 (2000) (noting that a joint monopolist is far better than decentralised monopolists because the later cannot coordinate the group as a whole). 87 See id. at 8–9 (examining the use of political regimes as a positive force for facilitating cooperation and mutual advantage). The state is also vital in sustaining social rules; preventing defection through the threat of coercive enforcement. See HART, supra note 7, at 91 (noting that even in societies governed by customs instead of law there still must be restrictions on unwanted behavior, such as violence and theft). 88 See EPSTEIN, supra note 86, at 8 (noting that without a centralized monopoly, the public suffers from various coordination failures). 89 See Druzin, supra note 85, at 560–61 (noting that many laws require the backing of the state in order to be effective). 84 Albany Law Review [Vol. 79.1 fresh set of criteria with which to assess the potential and limits of minimalism. Because interactive law involves individuals repeatedly interacting with each other it is more susceptible to degrees of selfordering. The greater the degree of repeated interaction, the greater the area of law’s potential to self-order. Let us now look at how we can use this on a practical level. IV. USING THE CHARACTERISTIC OF HOW “INTERACTIVE” THE LAW IS AS A GUIDE FOR POLICYMAKERS In this section, I will discuss how the presence of repeated interaction may serve as a guide to policymakers in assessing in what situations minimalism may be most viable. What is offered here is a heuristic to “cash in” on the energy of bottom-up order. The characteristic of repeated interaction may be marshaled as a kind of marker to identify where the state can and cannot realistically adopt policies in the direction of minimalism. Law that relates to behaviour characterized by a high level of repeated interaction is fertile soil for bottom-up order. In contrast, where there is no repeated interaction, norm-based minimalism really has no hope. Thus, policymakers can look to the presence of repeated interaction, specifically the intensity of it, as an indication that there exists the potential for a minimalist approach—the greater the repetition, the greater this potential. A. Strong and Weak Minimalism With highly interactive areas of law, the state can defer to the force of social norms and contemplate a less interventionist tack. Repeated interaction helps build coordinating structures (i.e. rules) and lift men from the blind morass of social disorder.90 Thus contract law, for example, does not need the state for rule-formation to the same degree as other forms of law. It is very good at producing rule-based order precisely because the parties engage in some degree of repeated interaction.91 This is particularly true in the case of long-term contracts. Law is needed to set forth and enforce basic property rights; however, the element of repeated interaction implicit in such dealings profoundly reduces the need for top-down law—it does not need the hand of the state to press down quite as hard. Indeed, this is true for a great deal of commercially-oriented law where Druzin, supra note 85, at 586–87. In that the parties can draw up their own terms that the State will then enforce, it is crafted to achieve precisely this outcome. 90 91 2015/2016] Social Norms as a Substitute for Law 85 individuals repeatedly deal with each other.92 All of this has practical implications for the State’s role as a producer of social order. The question before us is to what extent can policymakers make use of social norms and rely less upon top-down law? Our answer is that the state can favor a more minimalist tack in areas of law that are more interactive, taking advantage of the selfordering tendencies of such systems. Policymakers can achieve this by adopting whatever degree of minimalism is viable given how interactive the area of law is. As this is really a continuum, different areas of law will allow for different degrees of minimalism. Minimalism may come in both weak and strong versions. Strong minimalism is (1) complete noninterference: if the existing pattern of rules is already optimal (or simply sufficiently functional) there is no need for legislators to regulate it. It can be just left alone. Of course the state already does this to a vast and acknowledged extent—the proposal here is simply that the state may do this more. To take an extreme example, there is no need to regulate walking patterns. Simple yet ubiquitous social rules are more than sufficient to efficiently order massive flows of people through our cities. Indeed, foot traffic is a simple but excellent illustration of relatively efficient bottom-up order. There is no such thing as the law of foot traffic. It does not require regulation.93 While the efficiency of foot traffic in large congested cities may arguably be improved upon thorough regulation, the legislative and enforcement burden on the state would be tremendous, rendering such a proposal ridiculous. A less strong version of norm-based minimalism is (2) codification. Here the state simply grants formal recognition to an informal pattern of social order. Existing customary rules are merely given enforcement teeth to preempt occasional shifts in incentive structures that may undermine compliance.94 These sanctions often need not be very extreme, or in cases where actors are merely trying to coordinate, but 92 This has obvious implications for the evolution of the new law merchant, private law theory, and the rise of transnational law, more generally where the role of the State is increasingly minimized. This is expanded upon below. See infra Part IV.C; see also Druzin, supra note 85, at 586 (positing a theory of “high engagement” that attempts to account for the ability of commercial law to grow in a transnational context without resorting to a central legislative authority). 93 This is not the case, however, with vehicular or flight traffic, which is heavily regulated. 94 It should also be noted that the mere act of codifying an existent system of rules is in itself socially useful in that it clarifies the rules for participants already willing to comply but unable to perfectly coordinate (a coordination game as it is called in game theory). This is a largely unappreciated aspect to codification—one completely unrelated to the coercive power of law. For a fascinating treatment of this idea, see Richard H. McAdams, The Legal Construction of Norms: A Focal Point Theory of Expressive Law, 86 VA. L. REV. 1649, 1712 (2000). 86 Albany Law Review [Vol. 79.1 are unable due to a lack of clarity as to what the social rules are, codification alone may be sufficient. A far weaker form of minimalism is (3) modification. Here legislators tweak the social patterns on a structural level in order to correct minor inefficiencies while allowing the bulk of the system to function mostly untouched by regulation. Finally, a fourth option, always available, is (4) to simply engage in full and expansive top-down regulation. A good example of a situation where a small modification may reap huge benefits is where there is imperfect monitoring. Suppose, for example, parties engaging in repeated interactions are sufficiently patient but lack the ability to detect perfectly whether the counterparty shirked. In such situations, some classic results suggest that there will be inefficient punishment along the equilibrium path.95 This invariably undermines stable social ordering. Such systems, although they possess repeated interaction, may give rise to free riding, the existence of which will undermine bottom-up social ordering.96 With very small groups, the free rider problem is overcome through informal enforcement mechanisms implicit in social cooperation (i.e. being kicked out of the group or ostracized, etc.).97 However, as the size of the group grows larger and monitoring more difficult, free riding may become harder to prevent.98 Thus, we may have a fully-functional system of social order that, because the number of participants is growing, the system is threatening to collapse as more and more free riders leech its benefits without contributing. In such situations, policymakers can nudge equilibria onto a path of stable ordering by simply enhancing the players’ ability to monitor. Imagine, for example, a successful community watch program in a small town that emerged on an informal and voluntary basis being undermined by free riding as there is an influx of new 95 See generally Edward J. Green & Robert H. Porter, Noncooperative Collusion Under Imperfect Price Information, 52 ECONOMETRICA 87, 88 (1984) (noting inefficient punishment in collusion). 96 See Talia Fisher, Separation of Law and State, 43 U. MICH. J.L. REFORM 435, 457–58 (2010). 97 See Lars Udéhn, Twenty-five Years with The Logic of Collective Action, 36 ACTA SOCIOLOGICA 239, 244 (1993). 98 Id. at 240. For the foundational work on this problem, see MANCUR OLSON, THE LOGIC OF COLLECTION ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS 3 (20th prtg. 2002) (arguing that as group size increases, the problem of free riding will grow in relation to nonexcludable public goods). The core problem with informal social sanctions “is that they are, themselves, collective goods subject to a collective action problem.” Udéhn, supra note 97, at 246. See Douglas D. Heckathorn, Collective Action and the Second-Order Free-Rider Problem, 1 RATIONALITY & SOC’Y 78, 98 (1989); Pamela Oliver, Rewards and Punishments as Selective Incentives for Collective Action: Theoretical Investigations, 85 AM. J. SOC. 1356, 1368 (1980). 2015/2016] Social Norms as a Substitute for Law 87 residents. Previously, the ordering was able to emerge because of the repeated interaction of these neighbors and the ability to monitor compliance. However, participants may be tempted to cease contributing if they believe that other members of the group may be free riding, something they can no longer be sure of due to the inability to monitor such large numbers. Codification without sanctions will achieve very little in this situation, as this will not discourage free riding. Codification with sanctions against free riders would resolve the problem in a straightforward fashion. However, a remedy where participation is forced through the threat of formal sanctions may prove unacceptably paternalistic. As such, modification in the form of improving monitoring may be more appropriate. For example, a registry system that publicly records participation could be established by local government. By simply strengthening monitoring, social norms could still be efficiently harnessed while top-down regulation is kept to a minimum. Choosing between weak and strong minimalism is comparable to medical treatment.99 An experienced physician knows how to work with the human body. Mostly, this is achieved by simply not hindering the body’s ability to grow, function, and heal itself (noninterference). However, sometimes for the body to heal it requires some gentle intervention—a well-placed splint or cast around a broken arm (codifying). Sometimes, however, a medical emergency may require a more drastic measure—invasive surgery to remove a cancerous tumor (modification). However, in every case, the experienced physician adopts a minimalist approach unless absolutely necessary, relying on the body’s natural functioning wherever possible. Much like a physician, the state can exploit the natural emergence of bottom-up order—learn how to work with it. If done skillfully, the state can deftly shift much of the burden and complexity of legal creation to the participants themselves. The minimalist approach is quite apparent with the classical theory of contract, which discourages courts and legislatures from encroaching upon the autonomy of the contracting parties,100 preferring instead to 99 For a similar analogy regarding gardening, see Druzin, supra note 1, at 60, 62 (articulating a detailed taxonomy of legislative strategies to achieve what the author refers to as “legislative minimalism”). 100 See Harry N. Schreiber, Introduction, in THE STATE AND FREEDOM OF CONTRACT 1–2 (Harry N. Schreiber ed., 1998). The presence of repeated interaction in contractual dealings demonstrates clearly an area of law where the State can adopt a less interventionist approach. In many respects, this supports the laissez-faire liberal vision of contract. This does not, however, necessarily entail a complete lack of regulation, merely the opportunity to scale back the degree and vigor in which the State presently regulates. Clearly, private parties will still 88 Albany Law Review [Vol. 79.1 allow the parties involved, governed by rational self-interest, to regulate their own affairs (subject to some basic legal constraints).101 Similarly, in other highly interactive areas of law, the state could potentially adopt the same kind of hands-off approach, allowing the self-ordering machinery of interactive law to shoulder as much of the burden of legal creation and enforcement as possible. However, regardless of which approach is adopted—noninterference, codification, or modification—the first step is to know where minimalism, as a general policy, is even viable. A reliable way to gauge this is by considering how interactive the particular area of law is. B. Identifying the Areas of Law Predisposed to Minimalism An important point is that the present model is useful for both the advocates of minimalism and those not persuaded that there is value in such an approach. To the extent that the model identifies the limits to minimalism’s scope of application, it is of value in that it clarifies where top-down law is indispensable. As the degree of repeated interaction decreases, the State’s role in legal creation becomes increasingly more vital.102 As already discussed, this is best understood as a continuum. On one end of this continuum (the highly interactive end) are areas of private law such as family law with its well-established relationships of repeated interaction.103 Indeed, areas of law that traditionally fall under private law tend to be interactive, some extremely so. These relationships would exist in the absence of the state. Indeed, they are antecedent to the state and existed in prepolitical societies. prefer the extra security of detailed long-term agreements, enforceable in court or through arbitration. 101 RICHARD STONE, THE MODERN LAW OF CONTRACT 6 (Routledge 10th ed. 2013). Under the banner of self-regulation and freedom of contract, the courts are generally very reluctant to interfere with the parties’ agreement. RYAN MURRAY, CONTRACT LAW—THE FUNDAMENTALS 90 (2008). 102 This can be read as a structural argument for government beyond the standard justifications for the State: for example, as a solution to collective action problems, such as free riding and the tragedy of the commons, etc. 103 See infra Figure 1. 2015/2016] Social Norms as a Substitute for Law 89 _______________________________________________________________ _______________________________________________________________ Figure 1. A selection of areas of private law with respect to the level of frequent and repeated interaction between the same agents (i.e. how interactive it is). The classification, however, breaks down with respect to tort law in that it discourages repeated interaction. Here we see areas of law such as family law, contract law, employment law, many aspects of commercial law,104 and to a far lesser extent, aspects of real property.105 In that they involve preestablished relationships of frequent and repeated interaction between the same agents, these areas of law are more inclined to minimalism. On the other end of our continuum, we see areas of law where the dynamic of repeated interaction between the same individuals is not present in the same way, if at all—for example, constitutional law, immigration law, and most other areas of administrative law.106 For such areas of law, minimalism is not viable. They do not by definition concern relationships of frequent and repeated interaction between people. Rather, they concern the individual’s relationship with the state. As such, the state is required to prop up and sustain such areas of law. Indeed, these areas of law would not exist but for the state. They are born with the state.107 104 An umbrella term used here for certain forms of commercially-oriented law where there is the possibility of repeated individual to individual interaction, such as law related to trade, corporate contracts, and partnerships. 105 See supra Figure 1. 106 See infra Figure 2. 107 Criminal law is the exception here. However, as discussed below, it is classed here as noninteractive because it does not involve repeated interaction. See infra note 111 and accompanying text. In that it obliges inaction, it in fact inhibits interaction. See infra notes 112–14 and accompanying text. 90 Albany Law Review [Vol. 79.1 _______________________________________________________________ _______________________________________________________________ Figure 2. A selection of areas of public law with respect to the level of frequent and repeated interaction between the same agents (i.e. how interactive it is). This is a nonexhaustive list. Note that IP law is traditionally thought of as a form of private law; however, it is placed here in that it does not directly relate to relationships of repeated interaction (although it may involve them).108 Criminal law, while regularly concerning relationships between individuals, actually inhibits repeated interaction and so is included here. These two figures, however, are far from perfect. Private and public law is a convenient separation that helps, I think, to clarify the distinction between interactive and noninteractive areas of law. Yet while it is a useful (albeit somewhat simplified) dichotomy, the broad labels of private and public law are at times imprecise. For example, depending upon which aspect of the law we wish to focus, some areas of law may be characterized as highly or not at all interactive. This ambiguity is very much a function of the blunt, catch-all categories we use to define areas of law. Indeed, some areas of law defy perfect classification. For instance, financial services law involves vast numbers of often repeated interactions between the same individuals, whether in the form of contracts (such as leases) or dispositions of trust.109 If financial services were conceived as an area of contract law, then it should be understood as interactive and thus listed in Figure 1.110 If, however, it is conceived as a form of state regulation, then it is not at all interactive and should be listed in Figure 2.111 Another point, already touched upon, is that some areas of law may concern individual to individual interaction yet are not properly interactive law because this interaction is not repeated (between the 108 IP law arguably possesses features of both private and public law. See GRAHAM DUTFIELD & UMA SUTHERSANEN, GLOBAL INTELLECTUAL PROPERTY LAW 47, 48 (2008). The focus here is upon the administrative aspect to intellectual property. 109 See Anita K. Krug, Escaping Entity-Centrism in Financial Services Regulation, 113 COLUM. L. REV. 2039, 2041, 2045 (2013). 110 See supra Figure 1. 111 See supra Figure 2. 2015/2016] Social Norms as a Substitute for Law 91 same individuals). This is the case with both tort and criminal law. While interaction between private individuals may occur within these areas of law, it is very limited in nature. In the language of game theory, they are one-shot encounters. While the vast majority of criminal law relates to interaction between private individuals,112 it has almost nothing to do with repeated interaction and so falls in the noninteractive camp. Criminal law is framed mostly in the negative as injunctions against certain acts.113 It is what one should not do. In that it demands inaction, criminal law actually discourages interaction. Most aspects of tort law also fall into this category. However, where it bears upon pre-existing relationships of frequent and repeated interaction—for example, in the case of nuisance between two neighbors who regularly interact—it may show glimmers of interactive law.114 No doubt the basic components of criminal law and tort law have their roots in the primal soup of social norms within small groups where informal monitoring and retaliation are possible, yet in large modern societies they are no longer the interactive areas of law they once were in times past.115 Some readers may take issue with the taxonomy in Figure 2, arguing that several of these areas of law at times touch on relationships between agents. For example, sales tax informs the relationship between a buyer and a clerk, and intellectual property will relate to the relationship between a musician and a purchaser of music. However, while they may at times have relevance to individual to individual interaction, such areas of law remain distinct from highly interactive areas of law: tax law and IP law are not wholly constructed around and related to pre-established relationships of repeated interaction between the same agents in the same way as, for example, family law. Family law is an already complex system of order that antecedes the state. This is not the case with tax law, IP law, and immigration law. Moreover, and more importantly, these relationships are not necessarily ones of repeated and frequent interaction between the same people of the kind that 112 Some areas of criminal law, however, involve the individual’s relationship directly with the State. For example, tax evasion, treason, obstruction of justice, and perjury. 113 There are, of course, some exceptions; for example, where the criminal law imposes positive duties. See JOHN FORGE, THE RESPONSIBLE SCIENTIST: A PHILOSOPHICAL INQUIRY 236 (2008). 114 The reader should note that the taxonomy is not perfect: many sub-areas within these generalized areas of law may be interactive or noninteractive as the case may be. 115 See generally BINMORE, supra note 18, at 79 (highlighting the different folk theory limitations which arise in modern, urban life as opposed to smaller bands of historical huntergatherers). 92 Albany Law Review [Vol. 79.1 would allow for a clean application of the folk theorem. Another way to gauge how predisposed an area of law may be to minimalism is to pose the question differently. Instead of asking whether an area of law can produce bottom-up law, we instead ask how much an area of law needs top-down law. Both methods drive at the same ends (the capacity for self-ordering); they just do so from different starting positions.116 Thus we may ask: What would happen if the state got out of the business of legal creation and enforcement in that specific area of law? That is, without the state, to what extent could this area of law continue to function? The need for state enforcement is a good indication of an area of law’s inability to generate bottom-up order as robustly as areas of law that more directly concern relationships of repeated interaction between the same parties. Tax law, IP law, and immigration law, as indeed all areas of public law, would all fare rather badly without the state to create and enforce such law—in fact, they would likely cease to exist entirely. This is decidedly not the case with, for example, family law. Order of this kind would evolve with or without government. Granted, family law, contract law, or commercial law would look substantially different than their present form; however, unlike tax law, for example, they do not need the state’s continued patronage for their very existence because such areas are in fact largely the product of bottom-up order. This is a crucial difference, and it speaks directly to their respective capability to generate bottom-up order. Noninteractive law simply does not possess the mechanics to generate bottom-up order as robustly as law that entails the frequent, repeated interaction of private parties.117 Many areas of law militate against any application of the folk theorem in that they simply do not involve individuals interacting with each other, or where they do, this is only obliquely. As such, they are intrinsically inhospitable to minimalism. Yet as we move along this continuum and encounter ever higher levels of repeated interaction between agents, minimalism becomes increasingly more viable. Ultimately, the ability of the state to scale back its legal involvement will depend on just how interactive the area of law is. 116 Richard A. Posner, Legal Reasoning from the Top Down and from the Bottom Up: The Question of Unenumerated Constitutional Rights, 59 U. CHI. L. REV. 433, 433 (1992). 117 See Jeffrey J. Rachlinski, Bottom-Up Versus Top-Down Lawmaking, 73 U. CHI. L. REV. 933, 933–35 (2006). 2015/2016] Social Norms as a Substitute for Law 93 C. Limits in Scope of Application Many of the above areas of law are not highly interactive. This suggests that minimalism’s scope of application is somewhat restricted. On the other hand, areas of highly interactive law are very good candidates for norm-based minimalism. Compared to other areas of law, legislation in, for example, contract and commercial law tends to be fewer, and where it does exist, it often codifies customary practices among merchants with case law declaring and ascertaining these practices instead of introducing anything fundamentally at odds with the status quo. Indeed, the structural stability of international trade, which has emerged primarily within a vacuum of State-imposed legal order, is testament to the potential of commerce to self-order.118 International commerce now constitutes between 20 and 25 percent of the world’s entire GDP—an impressive system of private ordering that exists largely within a state of technical anarchy.119 The ability of commercial communities to generate customary law in the absence of the state is well documented. History is replete with examples. The medieval law merchant, which saw the emergence of commercial customary law across Europe in the tenth, eleventh and twelfth centuries, is perhaps the most frequently referenced example of this.120 The Law Merchant sprang from the business customs prevalent at the time and served as a tool of unified commercial discourse that transcended the hotchpotch of differing local systems of law that traders in that period would encounter, such as ecclesiastical, manorial, or civil.121 Arguably this continues today unabated with the new law merchant and the rise of international arbitration.122 The ability of large commercial communities to produce customary law, I posit, is primarily due to the repeated interaction present in such communities—the natural consequence of trade. Ultimately, it is 118 See Bryan H. Druzin, Anarchy, Order, and Trade: A Structuralist Account of Why a Global Commercial Legal Order is Emerging, 47 VAND. J. TRANSNAT’L L. 1049, 1050–52, 1056 (2014) (arguing that the emergence of a global commercial legal order may be largely attributed to the structural properties of commerce). 119 Peter T. Leeson, Anarchy Unbound: How much Order can Spontaneous Order Create?, in HANDBOOK ON CONTEMPORARY AUSTRIAN ECONOMICS 136, 141 (Peter J. Boettke ed., 2010). 120 Barry Macleod-Cullinane, Lon L. Fuller and the Enterprise of Law, 22 LEGAL NOTES 1, 5 (1995). 121 Id. 122 Leon E. Trakman, From the Medieval Law Merchant to E-Merchant Law, 53 U. TORONTO L.J. 265, 282 (2003). See also Gillian K. Hadfield, Privatizing Commercial Law, 24 REGULATION 40, 41 (2001) (“From the Middle Ages to the infant digital age, there are examples of law developed and administered by private entities with varying degrees of state involvement.”). 94 Albany Law Review [Vol. 79.1 possible because commerce is a highly interactive area of law.123 The degree of minimalism the state can adopt is directly commensurate with how interactive a particular area of law is: the more interactive, the more the state can consider allowing selfordering to drive the formation of legal structures (such as with commercial customs), stepping in only where necessary to codify or tweak informal rule-systems by employing strong or weak forms of minimalism. Repeated interaction can in this way serve as a yardstick with which to gauge where the state may adopt a less interventionist tack.124 Yet we must recognize that, while the potential for minimalism exists beyond the borders of contract, the ability of the state to employ such a legislative tack to any significant extent decreases as we step further beyond the boundaries of highly interactive law. Contract, at least in the case of long-term contracts, is a highly interactive form of law and therefore fertile soil for minimalism. Contract is, in a sense, the definitive example of minimalism. It is an area of law where the custom-based approach is already largely in use.125 In less interactive areas of law, however, the state may find itself seriously hamstrung in its ability to rely on social norms to help create and sustain social order. While minimalism need not be restricted to the realm of contract, the state will eventually run up against a wall where minimalism becomes increasingly untenable as repeated interaction between the same agents becomes scarcer. For this reason, some of the most problematic areas of modern regulation such as banking law, environmental law, and most areas of administrative law may remain fundamentally inhospitable to norm-based minimalism. It is difficult to assess to what extent minimalism can be employed in less interactive areas of law. This is a question that would have to be fleshed out through actual implementation. It remains a bit of an open question. However, in the case of highly interactive areas of law, minimalism holds great promise. While the guiding hand of regulation is a necessity in the case of noninteractive law, in the case of more interactive law, a less interventionist tack is certainly achievable. The key question is how interactive is the area of law. Wherever repeated interaction is sufficiently present, even if this is relatively limited in scope, minimalism to some degree becomes feasible. The door is swung open for the state to capitalise on See Druzin, supra note 85, at 561–62, 586. Another issue that would need to be considered is the frequency with which the interaction is repeated. 125 Druzin, supra note 66, at 391–92. 123 124 2015/2016] Social Norms as a Substitute for Law 95 informal social ordering. V. LIKLEY OBJECTIONS TO THE MODEL Let us now turn to some likely objections to the model. Some may object that the discussion thus far has given short thrift to the important role of custom across all areas of law—both private and public. Another likely (and valid) objection is that norm-based minimalism may lead to socially unjust outcomes. Let us address these potential objections in this order.126 A. Most Areas of Law Already Incorporate Custom to Some Extent None of the discussion up to this point should be read as suggesting that noninteractive areas of law are devoid of custom. Social norms underpin a great deal of noninteractive law. “Custom” is respected by courts and lawmakers. The intensity of repeated interaction will determine the capacity for self-ordering and the generation of custom. As previously discussed, however, frequent and repeated interactions of a binary nature will tend to yield the greatest capacity to produce stable bottom-up order because the intensity of repetition is high.127 Such relationships thus have the greatest potential to generate reciprocity, monitoring, retaliation, and self-enforcement.128 Small groups can also achieve this although to a less robust degree (because the intensity of repeated interaction tends to be less). 129 However, when we jump to far larger groups where repeated interactions appear very infrequently if at all, the ability to generate stable cooperative order on any substantial level is virtually nonexistent.130 Even still, normative order in a relatively weak form 126 Another potential objection may be as follows: from the standpoint of methodological individualism, one could argue that there is in fact the potential for repeated interaction between individuals and the state. The state is not a homogenous or lifeless thing; it is made up of individuals with divergent preferences. Private individuals coordinate with each other and form groups that interact with state officials repeatedly if not regularly and thus some areas of administrative law may be understood as interactive law to a limited degree. Indeed, a la public choice theory, rent-seeking may emerge from repeated interaction between agencies and interest groups. See Bruce G. Linster, Cooperative Rent-Seeking, 81 PUB. CHOICE 23, 24 (1994). With respect to minimalism, however, this objection is rather odd. The goal of minimalism is to reduce state involvement in social order, yielding instead to natural ordering. The partner in these interactions is the very entity we wish to minimize. 127 See supra note 67 and accompanying text. 128 See supra notes 67–68 and accompanying text. 129 See supra note 69 and accompanying text. 130 See MYERSON, supra note 57, at 350. 96 Albany Law Review [Vol. 79.1 will still arise.131 This is what the law recognizes as “custom.” We still see the emergence and use of custom in a wide variety of law that is not very interactive. Indeed, much of (effective) law is based (ultimately) on social conventions.132 If not, their prescriptive power would be much diminished. Constitutional law is a good example of this. While not dealing directly with pre-existing repeated dyadic relationships, constitutional law nevertheless draws heavily from general custom and deeply-entrenched social norms.133 While ostensibly a purely state affair, constitutional law builds upon customary rules that are generated from (loosely) interacting agents.134 These are omnipresent social norms that spring forth from the greater cultural discourse of a society. Indeed, constitutional law is underpinned by constitutional conventions,135 and constitutional law language tends to be vague (separation of powers, equal protection, etc.).136 In the absence of exhaustive codes, the courts rely a great deal on “custom” to make decisions.137 This is particularly true in the case of “unwritten” constitutions, such as the Constitution of the United Kingdom.138 However, we must exercise conceptual caution. The presence of custom should not be confused with far more robust forms of selfordering, as is the potential with highly interactive law. While all areas of law may generate custom—areas of law that are not highly interactive cannot produce the complex cooperative structures capable of monitoring and self-enforcement—the ingredients See id. at 351. William Twining, A Post-Westphalian Conception of Law, 37 LAW & SOC’Y REV. 199, 230 n.73 (2003) (“[T]he rule of recognition is based on social conventions that identify the sources of law. These social conventions represent the community’s acceptance of a scheme grounding the criteria of valid law.”) (citation omitted). 133 See Jeremy Waldron, Are Constitutional Norms Legal Norms?, 75 FORDHAM L. REV. 1697, 1712 (2006). 134 ZACHARY ELKINS ET AL., THE ENDURANCE OF NATIONAL CONSTITUTIONS 45 (2009). There is a well-established literature within the fields of law and economics, which speaks to selfenforcing constitutions as coordination devices for iterated interactions within society-at-large, and/or amongst state officials and/or between the state and individuals. See ELKINS ET AL., supra, at 90; MIKHAIL FILIPPOV ET AL., DESIGNING FEDERALISM: A THEORY OF SELFSUSTAINABLE FEDERAL INSTITUTIONS 142 (2004); RUSSELL HARDIN, LIBERALISM, CONSTITUTIONALISM, AND DEMOCRACY 14 (1999); Yadira González de Lara et al., The Administrative Foundations of Self-Enforcing Constitutions, 98 AM. ECON. REV. 105, 105, 109 (2008); Sonia Mittal & Barry R. Weingast, Self-Enforcing Constitutions: With an Application to Democratic Stability in America’s First Century, 29 J.L. ECON. & ORG. 278, 297–98 (2013). 135 For example, it is impossible to understand the United States Constitution by merely reading its text and ignoring political customs tied to the party system, etc. 136 Lawrence B. Solum, Originalism and the Unwritten Constitution, 2013 U. ILL. L. REV. 1935, 1950 (2013). 137 ELKINS ET AL., supra note 134, at 84. 138 See id. at 49. 131 132 2015/2016] Social Norms as a Substitute for Law 97 required for robust bottom-up order.139 The normative order (if it can even be called that) that arises is nowhere nearly as robust as with highly interactive areas of law.140 This can be clearly discerned if we again consider the degree of state support required by such areas of law. Indeed, it is very difficult to see how constitutional law or the various areas of administrative law could survive without a tremendous amount of state involvement. Notwithstanding the appearance of custom in areas of noninteractive law, areas of law such as tax law, immigration law, environmental law, capital markets regulation, and IP law require the hand of top-down law to fashion, structure, and sustain it to a far greater degree than, for example, that of contract, which in its most minimal form, requires no more than state enforcement of property rights. Thus, noninteractive law may at times embrace customary norms, unlike areas of interactive law that sees vast noninteractive areas of law are not powerful engines for the generation of normative order. Custom is but a faint tracing of the powerful bottom-up social ordering possible where relationships of repeated interaction between the same agents are involved. We should not confuse the mere presence of custom with stable complex bottom-up cooperative systems. It would be like confusing a ripple of water for a surging torrent. The presence of custom obscures the decisive point that these areas of law lack the crucial element of discrete relationships of frequent and repeated interaction that may generate truly robust systems of bottom-up order. While custom may inform top-down law, something for it to draw upon when needed, custom can never replace top-down law. B. Unjust Outcomes While highly active areas of law may produce complex cooperative systems, we must tread very carefully. Clearly, top-down law has a vital role to play in regulating social behavior in many areas of law. Even where self-ordering can flourish, the appropriate degree of minimalism remains an issue. Excessive minimalism may prove disastrous. The correct dosage of minimalism would have to be determined in relation to the specific system of order that emerges. One important consideration is the potential for grossly unfair bottom-up ordering resulting from an inequality of bargaining power. 139 See id. at 78–80 (arguing that constitutions created with the cooperation of citizens can be self-enforced). 140 See id. at 78. 98 Albany Law Review [Vol. 79.1 This may be a good reason for the state to step in even where selfordering may otherwise emerge. Indeed, this is precisely what the state does in the case of contract: strong deference is given to bottomup ordering but the state nevertheless legislatively intervenes where needed.141 A good illustration of the thorny problem of unequal bargaining power and how the state has chosen to respond is minimum wage laws. Regulation can be viewed as an intercession on the part of the state to correct an imbalance in bargaining positions. The state imposes duties that are owed to the state, but which can be claimed by individuals. Employment law, consumer protection law, and antidiscrimination law all involve repeated interactions, but policymakers intervene to correct equilibria that dictate inequality.142 For example, it is obviously possible to develop stable ordering through repeated interaction between a monopoly supplier and a consumer, or a monopsony employer and an employee. But policymakers will want to know more about the substantive justice of the pattern or ordering. Moreover, the motivation for this may not be purely normative. It may be driven by hard-nosed concerns regarding market efficiency. Antitrust or competition law is a good example of an area of law that often exhibits a high degree of repeated interaction between agents and yet cannot be left to selforder purely for functional reasons related to competition and market efficiency.143 There is a question of whether bottom-up order will be able to incorporate broader societal goals rather than merely the limited self-informed goals of the private actors engaging in areas of interactive law. As such, many will doubt that minimalism is necessarily to the interest of society. Put differently, there may well be self-regulation in certain areas of social interaction, but that does not mean that the resulting self-regulation necessarily represents a social optimum, as it may well only regulate the behavior in a way that maximizes the welfare of the people engaged in the behavior, not the broader group of people affected by it. Commercial cartels are a 141 See RICHARD CRASWELL & ALAN SCHWARTZ, FOUNDATIONS OF CONTRACT LAW 333 (1994) (“[F]reedom of contract . . . arose from [a] laissez-faire context.”); see also BRUCE W. FRIER & JAMES J. WHITE, THE MODERN LAW OF CONTRACTS 425 (2d ed. 2008) (stating that very few statutes directly prohibit the enforcement of contracts). 142 See, e.g., Ami L. diLorenzo, Regulation B: How Lenders can Fight Back Against the Affirmative Use of Regulation B, 8 U. MIAMI BUS. L. REV. 215, 215 (2000) (discussing the implementation of the Equal Credit Opportunity Act in 1976). 143 See John M. Connor & Robert H. Lande, How High Do Cartels Raise Prices? Implications for Optimal Cartel Fines, 80 TUL. L. REV. 513, 514, 525 (2005). 2015/2016] Social Norms as a Substitute for Law 99 good illustration of this.144 These concerns, while eminently valid, are not fatal to the minimalist project. The present model in no way claims that all cooperative equilibria generated by repeated interaction will align with our social values. Indeed, the emergent order may be normatively repugnant. For example, social norms that institutionalized complex systems of racial, gender, religious, or sexual discrimination have in the past emerged as very stable, robust social equilibria.145 The answer to the problem of sub-optimal equilibria—whether this is defined in normative or economic terms— is simple. Norm-based minimalism is merely a point of departure— a conceptual baseline for policymakers. Minimalism does not in any way preclude legislators from stepping in and deploying massive doses of regulation to remedy sub-optimal equilibria where it is deemed necessary. The model outlined a range of legislative options available to policymakers captured by weak and strong versions of minimalism—noninterference, codification, modification, and topdown law. Policymakers adopting a minimalist approach can scale back regulation in interactive areas of law, wait and see if, and what kind of stable equilibria emerges, then fine-tune as needed. Where bottom-up order produces strong social norms that help create and sustain social order, policymakers need not regulate further. Where, however, bottom-up order produces unacceptable externalities, regulation is required. The degree of regulatory intercession necessary can simply be judged on a case-by-case basis. Ever higher degrees of legislative intervention may be implemented until the situation is sufficiently remedied. Nothing about a minimalist approach precludes taking action where bottom-up order proves suboptimal. 144 See John M. Connor & Darren Bush, How to Block Cartel Formation and Price Fixing: Using Extraterritorial Application of the Antitrust Laws as a Deterrence Mechanism, 112 PENN ST. L. REV. 813, 849–50 (2008). 145 See, e.g., Segregated America, SMITHSONIAN NAT’L MUSEUM OF AM. HISTORY, http://americanhistory.si.edu/brown/history/1-segregated/segregated-america.html (last visited Nov. 11, 2015) (noting how old customs contributed to segregated communities by 1900); Racial Segregation in the American South: Jim Crow Laws, GALE GRP., http://find.galegroup.com/gic/infomark.do?&idigest=fb720fd31d9036c1ed2d1f3a0500fcc2&type =retrieve&tabID=T001&prodId=GIC&docId=CX2831400031&source=gale&userGroupName= itsbtrial&version=1.0 (last visited Nov. 11, 2015) (discussing how slavery was a common practice in the South and many Southern plantations would not have survived without it). 100 Albany Law Review [Vol. 79.1 CONCLUSION In this paper I have argued that where parties engage in repeated interaction, policymakers can place greater reliance on informal social norms instead of top-down law. The value in this is that doing so may allow for less reliance on the coercive instruments of government, easing the enforcement and legislative burdens on the state. The argument draws on the folk theorem in game theory: parties that repeat interactions tend to police their own behaviour and self-order.146 It was, however, also acknowledged that repeated interaction is no guarantee that self-sustaining equilibria or even equilibria that is normatively palatable will emerge. However, the presence of repeated interaction provides a reliable and useful indication as to where minimalism is at least a possibility. The contribution of this paper to the law and norms literature is that it proposes a technique to identify the precise areas of social discourse where the possibility of decreasing regulation is most feasible. The paper serves a second, perhaps even more useful purpose in that it clarifies the limitations of minimalism, pinpointing the areas of law that top-down law remains indispensable. The idea of minimalism is an intriguing concept. Precisely because informal order is informal, it cannot be directly created. As a consequence, we naturally tend to assume that the only way order arises is by directly imposing it, state law being the paragon of such order. Yet this is not true. Bottom-up order arises all the time and may be made use of in a strategic manner. While State-imposed order remains crucial, we may benefit from learning how to conscript the untapped energy of social norms as a legislative tool—to the fullest extent that this is possible. 146 See supra note 33 and accompanying text.

RELATED PAPERS

RELATED TOPICS

Log In

Using Social Norms as a Substitute for Law

Using Social Norms as a Substitute for Law

Related Papers

RELATED PAPERS

RELATED TOPICS