Quarterly Journal of Austrian Economics

Legal Polycentrism, the Circularity Problem, and the Regression Theorem of Institutional Development

Legal Polycentrism, the Circularity Problem, and the Regression Theorem of Institutional Development

Volume 17, No. 4 (Winter 2014)

Legal polycentrism is the view that law and defense are, in relevant respects, no different from other goods and services normally supplied by the market, and that, in view of the generally acknowledged superior allocative properties of the market, freely competing protection and arbitration agencies would provide these goods at a much higher level of quality than territorial monopolies of force do (Tannehill and Tannehill, 1970; Rothbard, 1973; Molinari, [1849] 1977; Fielding, 1978; Friedman, 1989; Hoppe, 1999; Murphy, 2002; Stringham, 2007; Hasnas, 2008; Long, 2008). Legal monocentrism, on the other hand, is the term I use to designate the familiar position that law and defense are prototypical public goods, which have to be supplied by a territorial monopoly of force (otherwise known as the state) if they are to be supplied at all.

The former view developed in response to the latter, and since it is one of the newest theoretical developments in the field of political economy, its well-formulated criticisms are still comparatively rare. The one I find the most interesting, and the one I would like to address in the present paper, is centered around the so-called “circularity problem” (Morris, 1998; Lee, 2008; Buchanan, 2011), which focuses on the alleged inherent institutional shortcomings of legal polycentrism. The problem in question can be summarized as follows:

To show that competition between protection agencies would have beneficial consequences, [legal] polycentrists often cite results from price theory about market competition. But there is a circularity problem here: markets presuppose a legal framework; hence before polycentrists can employ price theoretic arguments about market competition, they must first show that the legal requirements of markets are satisfied, that is, that property rights and contracts are enforced. If these requirements are not satisfied, it is illegitimately circular to draw on market competition as an argument for legal polycentrism. (Wiebe, 2012, p. 1)

The implicit worry here is that price theoretic arguments about market competition and other efficiency-enhancing features of free-enterprise-based institutional arrangements depend on, but do not prove the existence of the requisite legal framework in a stateless environment.

In response, it has to be noted that the difficulties ostensibly raised by the circularity argument are by no means unique to legal polycentrism. An argument of a very similar structure can be deployed against, for instance, the supposed contractarian justification for governmental legal monocentrism: it might be claimed that if the social contract can be made in the state of nature, then the state (understood as a contract enforcer) is redundant, but if the social contract cannot be made in the state of nature, then the state is impossible. In sum, either the contractarian position is viciously circular, or the social contract, needing no meta-state to enforce it, effectively becomes a self-enforcing anomaly.1

The reason why I mention this parallel is because I believe that it illustrates the fact that the circularity alluded to in the context of both of the abovementioned legal systems does not point to any fundamental unworkability of either of them. In order to make this fact clearer, I will rely on the hierarchy of levels of social analysis proposed by Williamson (1998, 2000), which distinguishes between soft institutions—customs, traditions, norms, and religions—that emerge largely spontaneously and develop in an evolutionary manner, and hard institutions, whose purpose is to specify “the formal rules of the game” (ibid., p. 597), i.e., the ones referring to property rights, contract law, etc.

Where a proponent of legal polycentrism claims that “a functioning market and a functioning legal order arise together” (Long, 2008, p. 141), a proponent of coercive legal monocentrism can equally justifiably claim that a functioning state and a functioning legal order arise together. The parallel under discussion seems to me to point to the fact that incipient legal orders, regardless of their more specific characteristics, such as mono- or polycentricity, are necessarily grounded in the underlying soft legal institutions (customs, traditions, general social norms, etc.).2 In other words, if by “state of nature” one means a state of affairs in which there are no hard legal institutions, then legal systems do, in fact, emerge directly out of the state of nature.

This is illustrated, for instance, by the fact that “roving bandits” (Olson, 2000) can rely on the soft, informal institutions of trust and ostracism to solve their own prisoner’s dilemma, successfully police or eliminate free riders, etc., and eventually become stationary bandits. The same applies to “anarchic” communities that manage to survive for at least some time without being subjugated by roving-bandits-turned-stationary-bandits but also generate wealth sufficient to become attractive prey for them.

In sum, what ultimately accounts for the stability and workability of any given set of hard legal institutions (be it monocentric or polycentric, coercive or voluntary, monopolistic or competitive, etc.) is the underlying set of corresponding soft legal institutions, i.e., the ones rooted in custom, tradition, religion, etc., or, more broadly speaking, the ones ultimately dependent on preferences rather than incentives. They are, in an important sense, and for all practical purposes, an ultimate given, since they originate over very long time periods in an endogenous, evolutionary manner. The familiar message of de la Boetie, Hume, Mises, and their modern successors is that any territorial monopoly of force that fails to tap into these institutions or at least make peace with them is ultimately bound to collapse (Hume, [1742] 1971; Higgs, 1987; Mises, [1949] 1996, pp. 188–191; de la Boetie, 1997). According to the theory of legal polycentrism, the same goes for any rogue protection or arbitration agency, i.e., an agency whose mode of operation fails to reflect accurately the shared values and expectations of the society it purports to provide with protective or legal services (Boettke, Coyne, and Leeson, 2008).

One of the unique features of a competitive, contractual, polycentric legal order is that the role it assigns to hard legal institutions is merely to make the enforcement of rules based on the underlying soft legal institutions more effective. In other words, in such an order hard legal institutions do not establish the rules of social cooperation, but rather allow the process of their enforcement to benefit from specialization, division of labor, economic calculation, capital accumulation, greater incentive compatibility, and other efficiency- and welfare-enhancing features of free-enterprise-based arrangements. As I see it, the chief, non-circular claim of the legal polycentrist is that market competition in the area of law and defense provision generates precisely these beneficial features, while state monopoly in this area necessarily prevents their emergence. This is so especially insofar as the idea behind coercive legal monocentrism is essentially to override the aforesaid soft legal institutions rather than consult them, which results in the impossibility of making a logically meaningful distinction between the coercive monopolist enforcing the law and merely claiming to enforce it. This, in turn, makes the whole concept of law empty or arbitrary (Wisniewski, 2013).3

There exist numerous historical examples of the abovementioned process of the development of hard legal institutions on the foundation of their soft counterparts, some of which point to monocentric, and some to polycentric results. The list of examples belonging to the former category is admittedly much longer, but this should not be surprising, since it is a well-known and well-understood fact that overcoming the collective action problem is much easier for roving bandits than for aspiring market entrepreneurs (Olson, [1965] 1971).

Institutional efficiency is a function of the underlying incentive structure, which, in turn, is largely a function of the underlying ideological conditions. As it happens, the praxeological features of collective action imply that minorities can undertake it with much greater facility than majorities, since within small groups benefits are more highly concentrated, interests are more uniform, and effective monitoring of free-riders is more feasible. This, coupled with the iron law of oligarchy (Mosca, 1939; Michels, [1915] 1959), implies that it is much easier for bands of roving bandits to generate incentive structures favorable to undertaking successful collective actions than it is for peaceful members of extended societies to accomplish the same task. In other words, it is much easier to establish a territorial monopoly of force than to create a network of private, competing protection and arbitration agencies that could safeguard a given society from the depredations of such a would-be monopoly.

However, the difficulty of the latter task by no means implies its impossibility—instead, it points to the crucial link between incentives and preferences, and to the crucial fact that successfully modifying the latter can substantially alleviate the collective action problem (Hummel, 2001; Stringham and Hummel, 2010). Thus we get what might be called the regression theorem of institutional development, whereby the development of higher-level (hard) institutions is conditioned by the development of lower-level (soft) institutions.4 No alleged circularity seems to make this process inoperative.

In response, the defender of the circularity objection might suggest that the above solution is of limited value, since “informal institutions are limited in their ability to scale up”, i.e., in their ability “to function effectively as population size increases. For example, reputational mechanisms might break down in large anonymous groups, since communicating information about cheaters becomes prohibitively costly” (Wiebe, 2012, p. 8). I believe that this suggestion rests on a misunderstanding of the role of informal institutions in the process of establishing a robust system of polycentric governance. The scalability objection would work if the role of informal institutions was to replace their formal counterparts rather than to provide the necessary foundation for their emergence. This, however, is not the case—as soon as formal institutions emerge against the background of their informal counterparts, the scalability of the latter (or lack thereof) becomes irrelevant.

This is best illustrated by Menger’s ([1871] 1976) famous description of the bottom-up, evolutionary process whereby the most marketable commodity assumes the role of a universal means of exchange upon outcompeting all of its less marketable rivals, thus transforming a barter economy into a monetary economy. Now, while the initial existence of barter arrangements is necessary for the initiation of the abovementioned process that culminates in the emergence of money, together with the complex, formalized institutional framework that allows for its effective operation, the scalability of the former setting is in no way identical or even proportional to that of the latter. A barter economy is clearly limited in its ability to scale up, but a monetary economy is not. Likewise, “soft” reputational mechanisms might be limited in their ability to create a sufficiently scalable set of legal and protective institutions, but the same need not apply to “hard” frameworks that develop on their foundation. A different conclusion may be reached only if one mistakenly thinks of soft and hard institutions as substitutes rather than complements.

In sum, there does not seem to be any insurmountable problem of circularity confronting the position of legal polycentrism. Falling back on the hierarchy of levels of social analysis described in the new institutionalist literature (Williamson, 1998, 2000) allows us to generate a regression theorem of institutional development which disposes of any troublesome circularity in this context. Scalability also does not seem to be an overwhelming problem here. The reason why voluntary, competitive institutions in the area of law and defense have a harder time scaling up than coercive, centralized ones follows from the praxeological features of collective action, but this is a well-known observation, long appreciated by the theorists of legal polycentrism, who clearly recognize the indispensable role of ideology and preference change in rectifying the above asymmetry (Hummel, 2001; Stringham and Hummel, 2010). Thus, as I see it, the circularity problem fails to undermine legal polycentrism.5

  • 1It might be argued at this point that the claim that the social contract is vulnerable to the circularity argument is not valid. The reasoning is as follows: The social contract argument claims that the authority of the state rests on agreement in the “state of nature.” It does not assume that this agreement can be enforced in the state of nature, and this is the way out of the circle: The social contract can be made, but not enforced, in the absence of the state. However, since the social contract is not able to establish the state unless it can be enforced, and, by definition, nothing can enforce it in the state of nature, the social contract argument’s alleged ability to escape the circularity objection still does not make it capable of accomplishing its intended goals.
  • 2Unless a legal system is imposed on the inhabitants of a given territory by conquest, but even then its long-term survival depends on the ability of conquerors to integrate it with the local soft institutions to a sufficient degree.
  • 3It has to be stressed that this observation does not contradict my earlier claim that any territorial monopoly of force that fails to tap into the soft institutions existing in a given society or at least make peace with them is ultimately bound to collapse. This is because, first, overriding such institutions is not tantamount to ignoring them altogether, and second, as territorial monopolies of force consolidate their power over time, their dependence on such institutions may loosen.
  • 4A somewhat similar solution is suggested by Friedman (1996), who claims that the formal arrangements of a polycentric legal and protective order can piggyback on the pre-existing equilibrium generated on the basis of a past sequence of mutual threat games. I believe that the narrative described in the present paper complements and improves upon Friedman’s proposal insofar as it explicitly grounds it in the distinction and mutual relationship between incentives and preferences, as well as between qualitatively different kinds of institutions.
  • 5It might be worth adding in closing that even if the circularity argument were correct, the argument that the advantages of market competition make a polycentric legal order preferable to a monocentric one has not been refuted. The circularity argument would show only that a state is necessary to establish a legal system. It would not show that, once a legal system exists, the monocentric order should be retained.


Wiśniewski, Jakub Bożydar. “Legal Polycentrism, the Circularity Problem, and the Regression Theorem of Institutional Development.” The Quarterly Journal of Austrian Economics 17, No. 4 (Winter 2014): 510–518.

All Rights Reserved ©
What is the Mises Institute?

The Mises Institute is a non-profit organization that exists to promote teaching and research in the Austrian School of economics, individual freedom, honest history, and international peace, in the tradition of Ludwig von Mises and Murray N. Rothbard. 

Non-political, non-partisan, and non-PC, we advocate a radical shift in the intellectual climate, away from statism and toward a private property order. We believe that our foundational ideas are of permanent value, and oppose all efforts at compromise, sellout, and amalgamation of these ideas with fashionable political, cultural, and social doctrines inimical to their spirit.

Become a Member
Mises Institute