Ideas on Liberty

Spontaneous Order and the Case for the Free Market

BY ISRAEL M. KIRZNER

THE case for the free society, it has widely and correctly been understood, rests in large measure on its economic achievements. It is demonstrated, through reference to history and by economic reasoning, that the free market has enormous potential for stimulating economic growth and prosperity. The proponents of classical liberalism can therefore argue that their idea not only respects the dignity, humanity, and moral worth of the individuals who make up society—it also promotes their economic well-being. Were it the case—as the enemies of freedom have again and again erroneously claimed—that absence of central economic planning constitutes a sure recipe for economic chaos and failure, it would be much more difficult to argue persuasively for the classical liberal society.

Many thoughtful observers might conclude that such supposed economic failings add up to too high a price to pay for the moral virtues of freedom. So it is eminently understandable that classical liberals have seen economic science, with its demonstrations of the economic advantages of free markets, as an important element in the intellectual struggle on behalf of the free society. While economic science itself may be value-free and quite independent of its role in the case for freedom, the ideology of freedom is fully justified in deploying the teachings of economics in promoting its cause.

To a considerable extent, the relevant lessons from economics revolve around the concept of spontaneous order. Whereas the untutored view of society is likely to assume that absence of central control must inevitably generate hopeless discoordination and frustration—economics shows how the opposite of this view is in fact the truth. Economics shows how, from the independent decisions of many market participants, there emerges a systematic process of learning and coordination. The outcome of this process is the spontaneous order of the market economy. It is upon this spontaneous order that the unprecedented prosperity of market economies rests. In recent years, particularly under the influence of Hayek’s writings, a good deal of emphasis has come to be placed in the literature of liberty on this specific doctrine, the doctrine of spontaneous order.

The purpose of the present essay is to draw cautionary attention to a tendency to treat all examples of spontaneous order uncritically as equally powerful intellectual ammunition in developing the classical liberal case. I shall argue that this tendency is based on a serious misunderstanding. While it is emphatically true that the economic rationale for the free society rests largely on the notion of spontaneous order, it is not the case, I shall maintain, that every example of spontaneous order represents a useful weapon in the arsenal of freedom. In particular, I shall claim, a number of recent demonstrations that benign social conventions and institutions may have evolved organically and spontaneously have been thought to represent useful lessons pointing unequivocally towards the appreciation of the virtues of freedom. My position will be that it would be hazardous for proponents of freedom to rest their case substantially on such demonstrations. The critical policy issues facing modern societies are likely to be such, I believe, as at best to render quite irrelevant the lessons of social science concerning the long-run spontaneous emergence of benign social conventions. Proponents of freedom, it seems, would be well-advised to distinguish sharply between the kinds of spontaneous order that are generated in markets (against a given institutional background) and those other kinds of spontaneous order that may operate benignly to modify the institutional framework itself.

A prefatory word of clarification is perhaps called for. Nothing in what follows should be read as in any way critical of theories of long-run spontaneous order themselves. Such theories are highly stimulating, and a number of them may well be thoroughly valid and important. They may go far to illuminate significant aspects of human history. Our purpose here is to point out that reference to such studies may weaken, rather than strengthen, the economic argument for liberty. The economic argument for liberty rests on the propensity of markets to coordinate decisions and activities. The extraordinary power of arguments rooted in market theory should not be compromised by well-meaning but unhelpful reference to other kinds of spontaneous order. To point all this out is in no way to denigrate the profound scientific and human value of the long-run spontaneous order studies.1

Spontaneous Order: The Long Run and the Short

It is first necessary to articulate more sharply the distinction we wish to draw between two levels of, or dimensions for, spontaneous order. Traditional spontaneous order theorists from Ferguson and Smith onwards, generally assumed a society of individuals acting independently within a given institutional framework. That framework provided them with a relatively fixed background against which to operate. Economics then teaches that, under such conditions, a learning and discovery process is likely to be set in motion such as spontaneously to coordinate the independent actions of the participants. Without central control, participating individuals are led to anticipate each other’s decisions correctly, yielding an orderly situation in which mutually advantageous exchange opportunities tend to become discovered and exploited for mutual benefit. The prime example of this kind of spontaneous order is the ability of a market to generate a tendency towards the market-clearing price. Clearly, the demonstration that such a tendency towards spontaneous order exists is of great significance for a society contemplating the avoidance of central economic planning.

What has been emphasized, on the other hand, in an important recent strand of literature, is the plausibility of long-run social processes—proceeding by and large without a fixed given institutional background—generating the gradual emergence of beneficial social conventions and institutions. Here, too, the outcome of these spontaneous processes is a state of affairs in which individuals correctly and confidently anticipate the actions and responses of other members of society. (It is this confidence that individuals have concerning what others will do that constitutes the conventions and institutions that are the spontaneous outcomes of these social processes.) In this sense, therefore, these long-run processes are “orderly”—they generate coordinated outcomes. Standard examples of such spontaneously achieved social conventions are language, standards of weights and measures, and codes of behavior among members of social groups.

These two kinds of spontaneous coordination process are, despite their shared features, fundamentally different from one another. The emergence within society of a common language, a common set of standards for weight and measurement, and common codes of social behavior, differs sharply from the emergence of a market-clearing price for wheat or for unskilled labor in competitive markets. The former are the results of long-run processes, lasting centuries, during the course of which we can presume many long periods of discoordination. The latter are relatively rapid processes. The former generate institutions—social signposts which facilitate subsequent social intercourse, but which do not themselves fulfill the final goals of individuals; these long-run processes result in the fulfillment of intermediate objectives only. The shorter-run process of spontaneous coordination may apply to the actions taken directly to fulfill final consumption objectives. The former, longer-run processes generate coordinated actions and responses which do not constitute interpersonal exchange transactions. (That different market participants use common conventions concerning quality, units of measure, and even the use of a common medium of exchange may indeed facilitate exchanges, but are themselves merely ancillary to such exchanges.) The shorter-run processes are processes generating mutually gainful exchanges between participants.

We shall argue that, from the perspective of defenders of the free society, these important differences point to vitally significant normative implications separating long-run processes of spontaneous coordination from the short-run processes: The former do not, while the latter do, in general lend support to the case for a free society.

Spontaneous Order and Social Welfare

The term “spontaneous order” is almost invariably taken to connote an outcome that is socially benign. There is indeed a sense in which this is likely to be true, but this sense is a quite limited one. No doubt it is in general desirable that individuals be able to arrive at confidently accurate anticipations concerning the actions and reactions of others. This avoids much disappointment and frustration on all sides. So that it is indeed useful that the language my children learn at home overlaps with the language learned by other children in their homes. This permits social intercourse and facilitates education. But there is hardly—in the insight that such institutions emerge spontaneously—any implication that the emerging institutions are the best conceivable such institutions. There is no guarantee that the English language my children learn at their mother’s knee will be a “better” language for purposes of social intercourse than, say, French—or Esperanto. The demonstration that widely accepted social conventions can emerge without central authoritarian imposition does not necessarily point to any optimality in the resulting conventions.

What is demonstrated, on the other hand, by short-run spontaneous coordination theory (i.e., by the theory of the free market economy) is that there does exist a spontaneous tendency toward social optimality under the relevant conditions. The price that comes to be expected by all is not merely any price, but the market-clearing price, i.e., the price that stimulates the exploitation of the greatest possible volume of exchange opportunities. What is spontaneously tended to be achieved in these processes is not merely the avoidance of unrealizable anticipations, but the stimulation of anticipations that take advantage of conceivable opportunities.

Spontaneous Order and the Case for the Free Society

As noted at the outset of this essay, spontaneous order theory offers valuable ammunition for the proponent of the free society. Careful consideration of long-run processes of spontaneous coordination strongly suggests, however, that these processes not be cited as prime examples of what freedom can achieve. Awareness of these processes may indeed afford salutary refutation of naive prejudices concerning the chaos believed inevitably to ensue from the absence of central control. But such awareness is quite inadequate for building a general case for absence of centralized control. These processes cast much light on human history; but they offer little guidance for policy makers or social theorists pondering the foreseeable future. This is so for several reasons.

First, as noted in the preceding section, long-run spontaneous theory does not, in general, argue that evolved institutions are the best that might have come about. Second, long-run spontaneous theory does not, in general—there are exceptions—argue that deliberate centralized planning could not have arrived at equally benign (or better) social institutions. Third, because these processes are long-run in character, the benign character of the outcome may fall far short of justifying the process as a whole. After all, during the course of a centuries-long spontaneous process, there may occur (conceivably as a result of the absence of central control) a vast multitude of situations filled with human suffering and misery. It could be that the positive features of the resulting institution simply fail (no matter from whose perspective) to be judged worth these sacrifices.

Short-run processes of spontaneous coordination are far less vulnerable to such reservations. First, as noted, market processes tend, in general, to eliminate sub-optimalities. The relevance of Pareto-optimality to market processes is a systematic one. Second, market process theory not only shows how market coordination can be achieved spontaneously; it shows (as in the socialist calculation literature and related work) how such coordination could hardly be simulated by deliberate, central design. Third, because of the relative rapidity of market processes it is much more plausible to conclude that the optimality achieved by their outcomes justifies any sub-optimalities suffered (as a result of absence of control) during the process. (It must be recognized, however, that some of the institutions generated during long-run processes may continue to provide useful social services for a far longer period of time than can the market-clearing price achieved in a short-run market process.)

To recognize the limited relevance of long-run spontaneous processes for the theory of a free society is not to confess skepticism concerning the worthwhileness of such a society. After all, the traditional economic case for liberty rests not on possible long-term improvements in social institutions, but on the demonstrated advantages of liberty within a given institutional framework. A free society is prosperous, not because it generates a benign evolution in its laws, but because its given rule of law promotes innovative production and exchange. The classical case for freedom does not require that long-run spontaneous processes lead necessarily to better and better arrays of social institutions. In fact, for many proponents of the free society, its advantages depend on the fulfillment of limited but very definite assigned governmental functions. These governmental functions may well be assigned through explicit design. If the advantages of freedom require a framework of limited government, then one’s assessment of the advantages of a centuries-long Nozickian process of freedom without a government may well be distinctly unfavorable—even if that process can be expected eventually to generate such limited governmental institutions.

But if all this is granted, then it should be pointed out that emphasis by proponents of classical liberalism on long-run spontaneous processes not only constitutes a weak reed on which to support their case—but in addition may itself weaken that case. Classical liberalism assumes an array of given institutions. To conflate classical liberalism with ideologies that call for no such given institutional framework is to sow confusion and misunderstanding. Out of such confusion and misunderstanding may emerge, not sympathy and appreciation for the free market society, but profound and unjustified skepticism of it. Efforts to broaden public understanding of the economic virtues of classical liberalism should, it very much appears, be undertaken with well-nigh exclusive emphasis of market processes with a given institutional framework. Reference to the possible spontaneous emergence of social institutions, where appropriate, should be made with careful clarification of their limited relevance for the economic case for the free society.

Dr. Kirzner is Professor of Economics at New York University and is a Trustee of The Foundation for Economic Education.

2. Among the works dealing with the spontaneous emergence or evolution of social norms and institutions see especially: Edina Ullmann-Margalit, “Invisible-Hand Explanations,” Synthese 39, No. 2, (October, 1978), pp. 263–291; A. Schotter, The Economic Theory of Social Institutions, (Cambridge: Cambridge University Press, 1981); R. Axelrod, The Evolution of Cooperation, (New York: Basic Books, 1984); R. Sugden, The Economics of Rights, Co-operation and Welfare, (Oxford: Basil Blackwell, 1986). A good deal of Hayek’s more recent work, especially his Law, Legislation and Liberty, 3 volumes (Chicago: University of Chicago Press, 1973, 1976, 1979), is steeped in this evolutionary perspective (and is probably partly responsible for the tendency being cautioned against in this essay).

• 1. Among the works dealing with the spontaneous emergence or evolution of social norms and institutions see especially: Edina Ullmann-Margalit, “Invisible-Hand Explanations,” Synthese 39, No. 2, (October, 1978), pp. 263–291; A. Schotter, The Economic Theory of Social Institutions, (Cambridge: Cambridge University Press, 1981); R. Axelrod, The Evolution of Cooperation, (New York: Basic Books, 1984); R. Sugden, The Economics of Rights, Co-operation and Welfare, (Oxford: Basil Blackwell, 1986). A good deal of Hayek’s more recent work, especially his Law, Legislation and Liberty, 3 volumes (Chicago: University of Chicago Press, 1973, 1976, 1979), is steeped in this evolutionary perspective (and is probably partly responsible for the tendency being cautioned against in this essay).
• 2A prefatory word of clarification is perhaps called for. Nothing in what follows should be read as in any way critical of theories of long-run spontaneous order themselves. Such theories are highly stimulating, and a number of them may well be thoroughly valid and important. They may go far to illuminate significant aspects of human history. Our purpose here is to point out that reference to such studies may weaken, rather than strengthen, the economic argument for liberty. The economic argument for liberty rests on the propensity of markets to coordinate decisions and activities. The extraordinary power of arguments rooted in market theory should not be compromised by well-meaning but unhelpful reference to other kinds of spontaneous order. To point all this out is in no way to denigrate the profound scientific and human value of the long-run spontaneous order studies.