The origin of social institutions is a very old concern in social theory. Currently it has re-emerged as one of the most intensely debated issues in social science. Among economists and rational choice theorists, there is growing awareness that most, if not all, of the social outcomes that are of interest to explain are at least partly a function of institutional constraints. Yet the role of institutions is negligible both in general equilibrium theory and in most neoclassical economic models. Among other social scientists, there is a burgeoning substantive interest in institutions ranging from social movements, to formal organizations, to states, and even international regimes.
This chapter discusses the two principal approaches to the problem of institutional genesisāinvisible-hand and solidaristic. It further argues that the second of these is likely to afford us with a better means of attacking the problem than the first. Finally, one particular solidaristic explanation that holds promise for future research on institutional genesis is introduced.
The Concept of Social Institutions
Although the term institution is bandied about quite liberally in contemporary social science, no consensual definition of it has as yet emerged. The ambiguity of the term gives authors both the obligation and the license to adopt their favorite definition. At the most general level, I will take the existence of a social institution to be revealed by the appearance of some regularity in collective behavior. Collective behavior may be said to occur if different individuals behave similarly when placed in the same social situation;1 regularity, for its part, indicates that this collective behavior endures over some long but indefinite period of time.
If institutions are revealed by the appearance of collective behavioral regularities, then one naturally wonders both about their origins and about the mechanisms responsible for their persistence. In institutionally rich environments, new institutions can arise from old ones through modification or diffusion processes (White, 1981; DiMaggio and Powell, 1983). Such solutions to the problem of institutional genesis are limited, however, because they are exogenous and thus beg the question of the prime mover.
What is most challenging to account for theoretically is just how institutions emerge out of anarchy, that is, from a state of nature. How, in other words, do institutions ever arise from a noninstitutional environment? Two types of explanations have been advanced to address this hoary old Hobbesian problem.
The invisible-hand approach to institutional genesis, advocated to a greater or lesser degree by Menger [1883] (1963), Hayek (1973; 1976), and Nozick (1974), among others, views the emergence of institutions as a spontaneous by-product of the voluntary actions of self-interested individuals who share no common ends or values (see Hayek, 1976: 111). In such accounts, existing social institutions are usually conceived as Pareto-efficient equilibria; therefore they are self-sustaining (because no one who is subject to them has an incentive to change them), rather than dependent on some third-party enforcement apparatus (like the state) who existence, in turn, requires additional explanation.
Since invisible-hand arguments can offer an entirely endogenous explanation for the emergence of social institutions, they are to be admired for their parsimony and elegance (Nozick, 1974:18ā22; Ullmann-Mar-galit, 1978). Their principal advantage is that they rely on fewer assumptions than do other kinds of explanations.
The alternative approach to the problem of institutional genesis rests on quite different premises. Rather than emerging spontaneously among self-interested actors each pursuing their own ends, institutions in this view are a product of solidarity. Solidarity can only arise among individuals who share some common end (Hechter, 1987). To attain this common end, actors must establish a set of obligations as well as a mechanism that enforces compliance to these obligations (Hobbes, [1651] 1968; Durkheim [1897]; 1951; Blau, 1964:253; Hayek, 1976). From the solidaristic perspective, institutions persist not because they constitute self-enforcing equilibria, but because they are supported by consciously-designed controls.
There are two varieties of solidaristic explanations. On the one hand, institutions can be imposed upon a given population by some conqueror or overlord. Since it is easy to explain institutional emergence in the face of significant power differentials among individuals, this solution begs too many questions to be theoretically interesting (as Hobbes well understood). On the other hand, individuals with roughly equal power can create institutions voluntarily, in effect binding themselves to a joint project. This contractarian process is theoretically interesting precisely because it is such a problematic outcome.
Which approach is superior, the invisible-hand or solidaristic one? There is a great deal of debate in the literature on this question. Most of the advocates of invisible-hand explanations of institutional genesis rest their arguments on repeated game theory.
Yet, these arguments only suffice for the establishment and maintenance of conventions (Lewis, 1969)āsuch as the rule that we all drive on the right hand side of the road2ārather than for the establishment of n-person cooperative institutions. By cooperative institution, I refer to an institution, principally serving nonclosely related kin,3 that enables those who are subject to it to reap a surplus by agreeing on a jointly maximizing strategy that is otherwise unavailable due to the absence or inappropriateness of markets.
There is an essential difference between conventions and cooperative institutions. Cooperation is the dominant strategy in conventions because there is no free-rider problem. Compliance with a convention provides its own private reward: for example, drivers who ignore conventional rules of the road take their own lives in hand. Hence, conventions indeed can be conceived of as equilibria. In cooperative institutions (which resemble Prisonerās Dilemmas), however, defection is the dominant strategy. Hence, these institutions can persist only by precluding free riders, or by assuring would-be cooperators that they are not liable to be exploited by defectors.
Contrary to the rhetoric of Taylor (1976), Hardin (1982), and Axelrod (1984), repeated game theory offers no adequate solution to the emergence of cooperation among n players of a Prisonerās Dilemma super-game (Hechter, 1990). The inadequacy of repeated game theory in this respect is due to two separate problems. In the first place, there are multiple equilibria in the supergame, some of which are efficient and some inefficient (Aumann, 1985).4 Yet under most conditions it is difficult to determine which of these multiple equilibria will be realized. In the second place, unique cooperative solutions to the supergame rest on a most unrealistic assumptionāthat players are endowed with perfect monitoring capacity (Bendor and Mookherjee, 1987). This assumption limits the application of game-theoretic solutions to the evolution of cooperative institutions to the smallest of groups.5
In the wake of these current difficulties with the invisible-hand approach, it is best to consider the merits of solidaristic explanations, even though they require much stronger initial conditions. From a solidaristic point of view, the emergence of cooperative institutions requires individual agreement on some common end, acceptance of corporate obligations, and the establishment of formal controls to preclude free riding.
Can these admittedly strong initial conditions be explained on the basis of the typical self-interested behavioral assumptions of rational choice theory?6 I believe that the answer to this question is a qualified yes. Using the relatively weak assumptions that are traditional in rational choice, it is indeed possible to explain the emergence of cooperative institutions on the basis of solidaristic logic. The remainder of this chapter sketches out the basic argument, and then suggests that the argument can be applied to several types of empirical situations.
A Solidaristic Approach to the Emergence of Cooperative Institutions
Briefly, the genesis of cooperative institutions depends on the conjunction of (1) individualsā demands to provide themselves with jointly-produced private (that is, excludable) goods, as well as on (2) these individualsā potential control capacityāthat is, their opportunities either to dissuade each other from free riding, or to assure each other of their intent to cooperate. Both demand and control capacity are necessary for the emergence of cooperative institutions; without either, this kind of institutional genesis is doomed.
The demand for cooperative institutions arises from individualsā desires to consume jointly-produced private goods (hereafter termed joint goods) that cannot be obtained by following individual strategies. Cooperative institutions are generally formed to take advantage of positive externalities, such as increasing returns to scale, risk-sharing, and cost-sharing. The demand for joint goods is heightened by contextual events like wars, invasions, epidemics, and natural disasters, as well as by endogenous processes like rapid demographic growth. These events and processes are commonly experienced by a number of people, and on this account stimulate demand for goods that spread riskāsuch as the protection afforded by walls around a settlement, and the insurance provided the establishment of a mutual benefit society.
But the mere existence of demand for a joint good is insufficient to guarantee its production. One of the firmest conclusions of rational choice is that whereas the production of private goods is hardly ...