Knowledge, Discovery, and Prices

There are reasons to believe that Hayek derived this insight from his participation in the 1930s debate on the possibility of socialist calculation. As Don Lavoie has argued (Rivalry and Central Planning: The Socialist Calculation Debate Reconsidered [Cambridge: Cambridge University Press, 1985]), however, Hayek’s principal opponents, the market-socialist economists (including Oskar Lange, Abba Lerner, and Fred Taylor), misunderstood the Austrian criticism of comprehensive, government central-planning because they tried to compress it within a theoretical framework that assumed away what, in the Austrian view, were the essential problems to be overcome. (See, for example, Oskar Lange and Fred Taylor, On the Economic Theory of Socialism, ed. by Benjamin Lippencott [New York: McGraw-Hill, 1964], esp. p. 70, where prices are assumed to be “parametric,” that is, they are treated as simply “given” to market participants.) While, broadly speaking, the Austrians were pointing out the serious difficulties with information that a (by assumption, benevolent) central planner would face attempting to mimic the results achieved by a market economy, the socialists were arguing in terms of general-equilibrium models of the economy, in which most knowledge was assumed as somehow “given,” either to a fictitious Walrasian auctioneer or to his socialist equivalent, the central planner. This type of theoretical framework (both in its “partial” and “general” equilibrium variants), in which the “knowledge problem” disappears almost completely, came increasingly to characterize the economics profession from the 1930s onwards.

The Economics of Information

The knowledge problem has begun to attract greater attention in recent years, however, starting perhaps with George J. Stigler’s article “The Economics of Information” (Journal of Political Economy, Vol. 69, [1961], pp. 213-225). Since Stigler’s essay, the field that has come to be known as the “economics of information” has grown at an enormous pace. As Fritz Machlup put it, this new specialization studies “the complexities that may arise from the fact that information, new or old, may be inordinately uncertain, incomplete, partial, biased, misleading, costly, available to some but not to others, or giving rise to expectations, warranted or unwarranted, of various future developments” (Knowledge: Its Creation, Distribution, and Economic Significance, Vol. 3: The Economics of Information and Human Capital [Princeton: Princeton University Press, 1984]). The expanding literature on this subject deals with a great variety of issues, including the information problems of speculative markets, insurance markets, product markets, labor markets, technological innovation, public decisions, the formation of “human capital,” and more.

As a consequence of this new interest in the economics of information, economists have increasingly come to appreciate Hayek’s insight regarding the informational role of prices. Most economists, however, continue to operate primarily within an equilibrium framework. They concentrate their analysis almost exclusively on states in which, because all the relevant knowledge is already known to the parties concerned, all profitable opportunities have already been exploited and individuals’ plans made compatible. Accordingly, much of this literature treats knowledge as a costly commodity, which, like all other commodities, must be economized. Thus, the resulting equilibria will entail not “perfect” but only “optimal” knowledge: agents will be found to possess deliberately uneradicated ignorance. Such ignorance remains uneradicated because the cost of acquiring further information is not compensated by the benefits to be derived from it. Here is where this approach faces a problem: for such decisions to be made correctly—as must be the case in equilibrium—agents must know beforehand, among other things, what they are ignorant of and the costs and benefits of the knowledge that they could acquire; that is, they must know what they do not know. This is obviously a strong assumption—it is at best only slightly weaker than the assumption of perfect knowledge. I will take up this problem later in this essay.

From the perspective of the economics of information, then, Hayek’s point is frequently taken to be that prices are particularly effective (that is, low-cost) aggregators and transmitters of information. Andrew Schotter, for example, argues that the competitive price system is “an extremely efficient information-processing device that allows individually scattered bits of local information known only to specialists to be exchanged at minimal cost in order to coordinate social activity efficiently” (Free Market Economics: A Critical Appraisal [New York: St. Martin’s Press, 1985]). This interpretation of prices as “summaries of information” (or “sufficient statistics”) is starting to appear often in economics textbooks, and to a large extent underlies Thomas Sowell’s Knowledge and Decisions (New York: Basic Books, 1980), probably the most extensive development of Hayek’s insights. While for Sowell prices perform this role satisfactorily, a growing body of literature examines Hayek’s argument critically. Prominent among the authors producing this literature are Sanford J. Grossman and Joseph E. Stiglitz (see, for example, S. J. Grossman, “On the Efficiency of Competitive Stock Markets Where Traders Have Diverse Information,” The Journal of Finance, Vol. 31 [1976], pp. 573-585, his “Further Results on the Informational Efficiency of Competitive Stock Markets,” Journal of Economic Theory, Vol. 18 [1978], pp. 81-101, S. J. Grossman and J. E. Stiglitz, “Information and Competitive Price Systems,” American Economic Review, Vol. 66 [1976], pp. 246-253, and their “On the Impossibility of Informationally Efficient Markets,” American Economic Review, Vol. 70 [1980], pp. 393-408). Grossman and Stiglitz claim to have formalized Hayek’s ideas and to have shown that, although they may be correct in the standard context of perfect competition, they are not correct in situations of costly information, the only ones, according to the authors, in which the informational role of prices makes any sense. In such situations, they argue, prices will be “informationally inefficient,” that is, it will not be enough for agents to know only prices (in addition to their knowledge of their own preferences and the availability of resources and technology) in order for them to take “appropriate” actions. Some possible responses to this argument can only be briefly mentioned here: First, it may not be necessary, for Hayek’s position, to say that agents need to know only prices; it may be enough to state that prices reduce the amount of knowledge they need to know in order to make appropriate decisions. Second, Grossman and Stiglitz seem to deal mostly not with Hayek’s argument but with a different informational problem: they are concerned with whether or not agents may extract or infer information effectively from the prices they observe (for example, by observing a higher price for a piece of land they may infer that others have found out that there is something valuable, say mineral deposits, on it). But Hayek’s point is that agents’ economizing reactions to prices lead them to make the right adjustments without having to know their cause (as in the tin example, above). Third, it could be argued that Hayek had in mind not a situation where the only obstacle to perfect knowledge is that knowledge is costly, but rather situations of what I will call, for lack of a better name, “sheer” ignorance, a type of ignorance that characterizes disequilibrium states and that will be considered below.

Another body of literature has interpreted Hayek’s arguments in a similar way. The Hayekian problem has been examined through a series of highly abstract mathematical models, notably in the work of Leonid Hurwicz. This work has shown that, as long as the conditions of perfect competition hold, the price mechanism is “informationally best,” in the sense that “it uses the minimum number of variables for transmitting information between economic units.” But this work also asserts that this conclusion, attributed to Hayek, is not valid for nonperfectly competitive environments (that is, situations with externalities, public goods, economic indivisibilities, etc.), which are considered more realistic. This literature then attempts to design “informationally decentralized” mechanisms, combining planning and markets, for the allocation of resources—mechanisms that may achieve more efficient results under nonperfect competition, given “the difficulty of placing all the relevant information in the hands of a single agency because information is dispersed throughout the economy.” (See, for example, Leonid Hurwicz, “The Design of Resource Allocation Mechanisms,” American Economic Review Papers and Proceedings, Vol. 58 [1973], pp. 1-30, his “On the Dimensional Requirements of Informationally Decentralized Pareto-Satisfactory Processes,” in K. J. Arrow and L. Hurwicz, eds., Studies in Resource Allocation Processes [Cambridge: Cambridge University Press, 1977], pp. 413-424, and his response to Kirzner, “Economic Planning and the Knowledge Problem: A Comment,” Cato Journal, Vol. 2 [1984], pp. 419-425. See also S. Reiter, “Information and Performance in the (New)² Welfare Economics,” American Economic Review, Vol. 67 [1977], pp. 226-234.)

Bounded Rationality

Herbert A. Simon has also considered Hayek’s argument. Simon is the proponent of an approach that is often regarded as the alternative to the standard approach; I will call it my “bounded rationality” theory. Simon’s main dissatisfaction with mainstream economics is what he believes to be its lack of concern about the “computational limits” of economic agents as “information processors.” In Models of Man (New York: John Wiley & Sons, 1957) he states that “the capacity of the human mind for formulating and solving complex problems is very small compared with the size of the problems whose solution is required for objectively rational behavior in the real world.” These computational limits, he argues, make it impossible for man to be the optimizing agent described by the standard approach. The market is then seen as one among several arrangements that, supposedly, simplify the decisions faced by agents. In The Sciences of the Artificial [Cambridge, Mass.: The MIT Press, 1981] and Reason in Human Affairs [Stanford, Cal.: Stanford University Press, 1983], Simon explicitly agrees to a large extent with what he believes to be Hayek’s argument; he states that market prices are effective ways of summarizing large amounts of information in order that larger problems can come within the scope of agents with “bounded rationality.” Although “bounded rationality” theory is sometimes regarded as very similar, if not identical, to the modern Austrian approach, a few brief clarifications have to be made: on the one hand, Simon’s view of prices as summaries of information has more in common with the “economics of information” or the “rational ignorance” view described above than with the “market process” approach, about which more soon. (Of course, it is possible that Simon’s view of prices could be inconsistent with his own theory.) On the other hand—and admittedly, this is a difficult issue—it is not clear that he holds the same idea of “ignorance” as do the Austrian economists. The problem for Simon is not that information is costly, as in the mainstream view, but that there are “too many” variables for the achievement of “perfect knowledge.” This sidesteps an important problem, for even if the facts to be known were few and simple, they would still have to be discovered by alert, entrepreneurial agents.

Competition and Discovery: Optimal vs. “Sheer” Ignorance

There is a final interpretation of Hayek’s writings to be considered. Although the function of prices as exemplified by his tin example was the most explicitly developed argument in Hayek’s 1945 paper, it is also possible to find, implicit in his writings during the socialist-calculation debate and increasingly explicit in his post-1945 papers on competition, an additional insight about the informational role of prices and about competition. It is an insight yet to be absorbed by the approaches discussed above. The insight, which can also be found underlying the writings of Ludwig von Mises in the same period, consists in viewing market competition as a disequilibrium, rivalrous process, as a “discovery procedure,” which produces knowledge that no market participant would have possessed in its absence. This perspective, which even Mises and Hayek apparently only started to become aware of as a result of their debates with the market-socialists, has been developed more consistently and explicitly by contemporary followers of Mises and Hayek, particularly Israel M. Kirzner.

This approach also distinguishes the “optimal” ignorance of the “economics of information” from what can be termed “sheer” ignorance. As indicated, ignorance generally seems to have come to signify almost exclusively the result of a deliberate decision by an optimizing individual not to acquire further information, after weighing the costs and benefits of an additional “unit.” When Kirzner and others like him refer to ignorance and error and to the need for a discovery process, they have in mind “sheer” error. This is a situation in which agents have failed to correctly perceive the alternatives available to them not because they duly judged the effort unprofitable, but because they either have not noticed the alternatives or have not noticed the profitability of searching for them. (See, for example, I. M. Kirzner’s “Knowing About Knowledge: A Subjectivist View of the Role of Information,” in his Perception, Opportunity and Profit [Chicago: University of Chicago Press, 1979], pp. 137-153.) This ignorance is due to a lack of alertness on the individual’s part and not to his or her optimal response to costly information. This clarification is necessary also to understand that the idea of money profits as incentives to the discovery of new alternatives does not refer to incentives in the maximizing sense. Profits are not merely rewards that induce an individual to engage in a course of action—or in a search—whose possibility was known but that would have been found economically unattractive in the absence of the reward. Such a role may be present but does not exhaust the function of prices, for entrepreneurial profits can also be incentives that stimulate the perception and discovery of int previously unknown alternatives, which were not part of the agent’s framework of means and ends at all.

Ignorance, Discovery, and Normative Standards

In addition to the need for distinguishing the different views regarding the informational role of prices, there is also the problem of determining the appropriate standard by which to judge their effectiveness in performing that role. When judging the “informational efficiency” of the price system or, more generally, the efficiency of market outcomes, many mainstream economists usually use as a criterion the perfectly competitive (equilibrium) state, which under certain specified conditions can be said to be “Pareto Optimal.” These conditions are frequently not believed to hold in reality; thus, many criticisms of the market hold against it “imperfections,” such as the existence of externalities, public goods, “informational asymmetries,” and so forth. Harold Demsetz in “Information and Efficiency: Another Viewpoint” (The Journal of Law and Economics, Vol. 12 [1969], pp. 1-22) criticized this approach to efficiency, which he called “Nirvana economics.” Demsetz offered instead a “comparative institutions” approach, in which “the relevant choice is between alternative real institutional arrangements.” Demsetz’s important point, however, was placed within an equilibrium framework, which did not allow him to incorporate the entrepreneurial dimension discussed above.

On the other hand, to consider the world in disequilibrium terms is to accept not only the possibility of many of the problems pointed out by the critics of the market, but also the possibility of a normative standard that is not the Paretooptimal state. This standard would try to determine which institutional arrangements would be more likely to stimulate the discovery of, among other things, these very problems and solutions to them. In the Austrian, or market-process, disequilibrium view, the market system is, at any time, full of regrettable “inefficiencies” and mistakes (many of which will be in the process of being entrepreneurially discovered and corrected). But it is scientifically invalid simply to assume the existence of a government authority in possession of all the knowledge necessary for their solution. The analysis should, more appropriately, consider which social arrangement has the means for the discovery of such knowledge. Viewed in the light of such a standard, the market has at least one major advantage over other systems: through its translation of innumerable inefficiencies and mistakes into opportunities for pecuniary profit, it alone seems to have the power to awaken and mobilize the entrepreneurial alertness of market participants and thus to promote the discovery of these inefficiencies and their solutions. (Some of these themes are considered in a policy context in the second half of The Economics of Time and Ignorance by Gerald P. O’Driscoll, Jr., and Mario J. Rizzo [Oxford: Basil Blackwell, 1985]).

Conclusion

The elaboration of this insight about the role of prices in a world of ignorance poses a serious and exciting challenge to economists and, indeed, to all social scientists. Although models of “rational ignorance” have been used with varying degrees of success to analyze many social processes, few such analyses have incorporated the dimension of “sheer” ignorance discussed above. (For an exception in the field of history see B. L. Anderson’s “Entrepreneurship, Market Process and the Industrial Revolution in England,” in B. L. Anderson and A. J. H. Latham, eds., The Market in History [London: Croom Helm, 1986], pp. 155-200.) There are many opportunities for “academic entrepreneurship” in this area, which can provide the foundation for countless research papers, doctoral dissertations, journal articles, and books.

Esteban F. Thomsen is a graduate student in economics at New York University.