¹ For a representative sample of the conventional view, see Schumpeter, Joseph, Capitalism, Socialism, and Democracy (New York: Harper and Brothers, 1950), pp. 172–86.Google Scholar

² Totowa: Rowman and Allenheld, 1985.

³ Though not necessarily the Lange-Taylor model. See note 52.

⁴ Von Mises, Ludwig, “Economic Calculation in the Socialist Commonwealth,” ed. Friedrich, Hayek, Collectivist Economic Planning (New York: Kelly, 1967), pp. 85–130.Google Scholar

⁵ I believe that this interpretation of Mises is wrong, but a discussion of Mises's views would take us too far afield. See Lavoie, Don, Rivalry and Central Planning: The Socialist Calculation Debate Reconsidered (New York: Cambridge University Press, 1985), pp. 48–77.Google Scholar Lavoie's book is valuable for combating the conventional view of the socialist calculation debate.

⁶ Eugene Barone, “The Ministry of Production in a Collectivist State,” Collectivist Economic Planning, Appendix, and V. Pareto, Cours D'Economic Politique, Volume II, 1897.

⁷ Buchanan, p. 110.

⁸ Buchanan, pp. 16–17, 110. In what follows, I shall add to Buchanan's account where I think he has neglected some important aspects of Hayek's views. One reason Buchanan has done so is that his account relies for the most part on “The Present State of the Debate” [henceforth called PSD] in Collectivist Economic Planning, Hayek's 1935 article on this topic. Hayek's view is not fully appreciated unless one also looks at his 1940 article, written specifically in response to Lange, , “Socialist Calculation III: The Competitive ‘Solution’,” [henceforth called SCT] in Individualism and the Economic Order (Chicago: University of Chicago Press, 1948), pp. 181–208Google Scholar, as well as some later articles which illuminate Hayek's general framework. Two articles that are valuable in this regard are “The Use of Knowledge in Society,” in Individualism and the Economic Order, pp. 77–91 [henceforth called UKS] and “Competition as a Discovery Procedure,” in New Studies in Philosophy, Politics, Economics and the History of Ideas (Chicago: University of Chicago Press, 1978), pp. 179–90 [henceforth called CDS].

⁹ See UKS in particular. Hayek tends to use “knowledge” sloppily, not distinguishing it from true belief, where one's beliefs may not be justified, or even from justified belief, where one's beliefs may not be true. To avoid such problems, I shall generally use the term “information.” It should be assumed, unless otherwise stated, that such information is true.

¹⁰ ibid., pp. 81–83.

¹¹ ibid., pp. 81–82. Of course, this is not meant as a definition. The point is that these are two very important necessary conditions for something being an economic problem. (It's worth noting that Hayek's arguments in what follows can probably go through with the weaker claim that economic problems typically involve dispersed knowledge about constantly changing conditions.)

¹² At one point Buchanan says “in the market enormous amounts of complex information are utilized in the emergence and adjustments of prices over time…” (p. 16), but he really doesn't say much more than this.

¹³ UKS, pp. 85–87.

¹⁴ My use of “coordination” and related terms throughout this paper is different than the “coordination problem” of which game theorists speak. For game theorists, a coordination problem occurs when two or more people can get what they most prefer only if they choose the same course of action; the coordination problem arises in that there is no unique action that will produce what they most want. For example, in order to get a smooth flow of traffic, it doesn't matter whether we drive on the left or the right, but we must all drive on the same side of the road. The use of “coordination” in this paper concerns whether opportunities for mutually beneficial exchanges are grasped or recognized, or whether exchanges are made on the basis of misinformation about economic realities. See Section III for more detail.

¹⁵ Of course, those who make pricing and production decisions will make errors, and these will cause prices to be somewhat mistaken relative to changing economic realities. The structure of the argument does not require the assumption that prices will communicate and disseminate information perfectly. Rather it requires that a system which allows people to act on their beliefs about changing economic realities, and which tends to reward them if those beliefs are more often than not correct, will be more coordinated than a system which lacks such features. A similar point arises in Section IV in the discussion of business cycles.

¹⁶ Buchanan, p. 110.

¹⁷ ibid., p. 111.

¹⁸ I rely here on Lange's account, since his is more thorough. See Lange, Oskar, “On the Economic Theory of Socialism,” in On the Economic Theory of Socialism, ed. B.E., Lippincott (New York: McGraw-Hill, 1964), pp. 57–144.Google Scholar

¹⁹ In this paper, I employ an intuitive and common-sense notion of equilibrium where it means, roughly, a stable outcome where there are no endogenous disturbances. Hayek's conception of equilibrium will be discussed in Section III.

²⁰ Any standard introductory text in economics will show why a perfectly competitive market in equilibrium is Pareto-optimal.

²¹ Buchanan, pp. 110–11.

²² ibid., p. 111.

²³ ibid. The argument that Buchanan discusses is that, in the LT model, price readjustments would have to be so frequent and the cost of constantly accumulating data would be so high that the system would be less efficient. This is in effect an argument that, were the LT model to try to mimic market capitalism's flexible adaptation to change, it would be so costly as to be inefficient.

²⁴ It should be noted that Buchanan also discusses what he takes to be a motivational objection that Hayek raises against the LT model. I discuss this in Section II.

²⁵ SCT, pp. 188–89, 192–93. Hayek also notes that for goods made to order the notion of price fixing from above makes no sense.

²⁶ ibid., p. 193.

²⁷ Hayek wasn't as explicit about this as he should have been when he discussed the LT model, but remarks in PSD, pp. 212, 226–27, and SCT, pp. 188, 191 show that his benchmark of comparison is not a perfectly competitive market. UKS and, most importantly, CDS, pp. 179–86, give a systematic account of the misuse of the perfect-competition equilibrium model.

²⁸ For a history of the concept of perfect competition, see Stigler, George, “Perfect Competition, Historically Contemplated,” in Essays in the History of Economics (Chicago: University of Chicago Press, 1965), pp. 234–67.Google Scholar

²⁹ Some of the more important ones are: (i) Each agent is a utility maximizer. For consumers, this turns out to entail that they obtain the same marginal utility per monetary unit for every product they purchase. For producers, it turns out to entail the rules we discussed earlier: e.g., that output should be maintained at the point where price equals marginal cost, (ii) All goods are privately owned, (iii) There are no externalities, (iv) All utility functions are independent.

³⁰ There may be multiple equilibria in a perfectly competitive market, but this issue isn't important for the topic at hand.

³¹ In Walras's model of how perfectly competitive markets reach equilibrium, which Lange alludes to, the static nature of trial and error is particularly clear. An auctioneer calls out a set of tentative prices and the economic agents indicate precisely what they would supply and demand at those prices; if there is a shortage or surplus, the auctioneer calls out a different set of prices, and so on until the equilibrium set is found. No trading or exchange occurs while the tentative prices are called out; this occurs only when the right prices are found.

³² SCT, pp. 196–98.

³³ Buchanan, p. 112.

³⁴ A production function is a technical relationship, defined for a given state of technical knowledge, which sets out the maximum output capable of being produced by each and every set of specified inputs. Any standard introductory economic text will explain how given knowledge of production functions and factor prices, one can determine when price equals marginal cost.

³⁵ SCT, pp. 196–99, PSD, pp. 226–29, CDS, p. 185.

³⁶ Buchanan, pp. 15–16.

³⁷ CDS, pp. 185–86. This is the only passage I am aware of where Hayek seems to say that real markets tend towards Pareto-optimality.

³⁸ A referee for this journal thinks that the horizon of possibilities that Hayek is referring to is not the Pareto-optimality frontier. Two points seem to support the referee's interpretation: Hayek seems to refer to all the possible ways of producing goods, which is a different notion than Pareto-optimality (since utility or welfare is affected by distribution and production), and he refers to markets increasing the general pool, also a different notion than Pareto-optimality (since increasing the general pool may make some worse off). However, the evidence is mixed. Hayek also says that markets increase the chances that unknown people will get as large a real equivalent for their relative shares as possible; since he does not refer to the chance that markets may decrease certain people's real shares, a natural interpretation is that Hayek does not believe that real markets will make some better off while making others worse off. In any event, my aim in the text below is merely to show that if one thought that Hayek's criticism of the LT model rested on the claim that real markets approach or approximate Pareto-optimality, then Hayek's argument falls apart.

³⁹ Buchanan, pp. 42–44.

⁴⁰ Buchanan, p. 41.

⁴¹ Entrepreneurship and Competition (Chicago: University of Chicago Press), pp. 215–17. Hayek seems to suggest something like the plan coordination standard in CDS, p. 184. A number of Kirzner's essays on Hayek and issues arising out of the socialist calculation debate can be found in Perception, Opportunity and Profit (Chicago: University of Chicago Press, 1979).

⁴² The relationship between the plan-coordination standard and Pareto-optimality seems to be something like the following. One way to explain the sense in which a perfectly competitive market in equilibrium is Pareto-optimal is that there are no unexploited opportunities for mutual gain anywhere in the system. One important reason why real markets are never Pareto-optimal is that given scattered information and continual and continuos change, there are always unexploited opportunities for mutual gain. The plan-coordination standard of efficiency thus focuses on some of the salient reasons why real markets cannot be Pareto-optimal. However, unlike the Pareto-optimal standard, it does not use the situation where there are no unexploited gains from trade (e.g., a perfectly competitive market) as its benchmark by which to judge the efficiency of systems. Rather, the plan-coordination standard takes as a given that there is scattered information and continuous change – and accordingly unexploited gains from trade – and evaluates the processes or mechanisms 2 system has for coordinating the constantly changing scattered information.

⁴³ There are (at least) two reasons for this. First, it is desirable if an efficient system can be shown to be in equilibrium, e.g., shown to be stable, or at least to have equilibrating tendencies. Second, assuming that considerations of welfare or well-being count for something, there is a prima facie reason to promote and maintain an efficient system; and an understanding of those conditions which help to stabilize an efficient system can aid us in helping to bring about that system.

⁴⁴ Individualism and the Economic Order, pp. 33–56.

⁴⁵ ibid., p. 42.

⁴⁶ O'driscoll, Gerald and Rizzo, Mario, in The Economics of Time and Ignorance (New York: Basil Blackwell, 1985), p. 84Google Scholar, argue that Hayek's notion of equilibrium is mistaken since, under certain circumstances, correct foresight will prevent any equilibrium. They use an example, borrowed from Oskar Morgenstern, of Sherlock Holmes being pursued by his opponent Moriarty. If each perfectly anticipates what the other will do, this will lead to an endless series of plans, revisions of those plans, new plans, etc., and so no stability will ever occur. However, this example does not provide any problem for Hayek's analysis of equilibrium. Hayek says that, in order for any equilibrium to exist at a certain time, there must be a state of the world at a later time in which all the parties' plans could be carried out. Since Holmes and Moriarty's plans were mutually incompatible – the achievement of one of their plans requires the thwarting of the other's plan – in the first place, their plans couldn't be in equilibrium no matter what degree of knowledge they had. Thus Holmes and Moriarty's perfect foresight does not prevent an equilibrium.

⁴⁷ Buchanan, p. 5.

⁴⁸ Such a reply will involve interpersonal utility comparisons. For the sake of the argument in this section, I will assume that there is no problem with making such comparisons.

⁴⁹ Lange does provide an analysis of a socialist system with no markets in consumer goods and labor, but he does not favor such a system. See Lange, pp. 90–95.

⁵⁰ Of course, it doesn't follow that a coordinating process makes them worse off.

⁵¹ “The bourgeoisie… has rescued a considerable part of the population from the idiocy of rural life.” Karl Marx and Friedrich Engels, The Communist Manifesto, chapter II.

⁵² This is one of the primary differences between the LT model and more modern models of market socialism, wherein the economy largely consists of worker controlled firms. The workers in these firms hire and fire managers and have control over pay scales and differentials, worker relations, conditions in the firm, etc. The other principal difference between the LT model and more modern market socialist models is that, in the latter, there is no CPB which fixes producers' prices. There is a full-blown market for producer goods; the role of the state is largely limited to planning new investment. For a representative sample of models of modern market socialism, see Schweickart, David, Capitalism or Worker Control? (New York: Praeger Publishers, 1980).Google Scholar

⁵³ Lange's sole comment on this issue is that the managers are “public officials” (p. 74). This doesn't tell us who chooses them.

⁵⁴ See Lange, pp. 103, 105–6.

⁵⁵ In fact, on Hayek's view, business cycles arise from problems in producer markets, though the ultimate cause is the monetary policy of central banks. Hayek's business cycle theory is discussed in O'driscoll, Gerald P., Economics As a Coordination Problem: The Contributions of Fredrick A. Hayek (Kansas City: Sheed Andrews and McMeel, Inc., 1974).Google Scholar

⁵⁶ ibid., p. 106. When Lange said that mistakes can be localized in his system, perhaps what he meant was simply that there would be less mistakes in his system, since there would be less markets. The abolition of producer markets does not, however, mean that there will be less mistakes in the system. The mistakes in question concern mistaken beliefs or hunches about changing economic conditions, and the replacement of producer markets by CPB price-fixing does not mean that in capital, natural resources, the relation of inputs to outputs, etc., will cease or slow down. That decisions about producer goods will be made less often and by fewer people in the LT system as compared to market capitalism does not show that the amount or complexity of these decisions will diminish.