by Andrew Chamberlain
May 20, 1998
With the collapse of Communism in eastern Europe and the former Soviet Union, it is tempting to conclude that the idea of central economic planning had been laid to a permanent rest. Indeed, even self-described socialist Robert Heilbroner has conceded that “the contest between capitalism and socialism is over: Capitalism has won” (Heilbroner, 1989, p. 98). However, such epitaphs of socialist planning as an economic ideal apparently are premature as “market socialism” has proven to be one of the most enduring proposals in modern economics.
The earliest attempt at a theoretical framework for market socialism (i.e., centrally controlled firms that sell goods to consumers in competitive markets) was offered by Oskar Lange during what has come to be known as the Socialist Economic Calculation Debate (SECD). The debate involved several “neoclassical” American socialists such as Oskar Lange and Abba Lerner on one side, and a group of “Austrian” economists including Friedrich Hayek and Ludwig von Mises on the other. The focus of the SECD was whether a centrally planned economy under socialism would ever be able to achieve the economic goals of its proponents, with the Austrian camp arguing that it emphatically could not.
In this paper I will argue that the Austrian critique of central planning remains as a powerful argument against the feasibility of market socialism. The resilience of market socialist proposals today, I will argue, is due mainly to a failure to appreciate the insights of the Austrian argument in the SECD about the epistemological problems faced by central planners. This lack of understanding is not necessarily rooted in any ideological predilections or lack of rigor, but rather in the fact that the dialogue of the SECD was between two sets of scholars that were unwittingly operating within very different theoretical frameworks. As a result, the arguments of Hayek and others were misunderstood and largely dismissed within the economics profession at the time of the SECD, and only recently have begun to gain recognition as being of more than merely historical relevance.
The remainder of the paper is divided into three sections. First, an abridged history of the SECD is outlined and the major arguments of each side are summarized. Next, the Austrian position is analyzed in more detail, with emphasis placed on the theoretical differences between the two sides of the SECD. A final section then summarizes the relevance of the SECD for modern market socialist proposals as well as for economic theory and methodology in general.
I. The Socialist Economic Calculation Debate
Prior to 1920, the economics of socialism was largely unexplored. Marxists, following the lead of Karl Marx himself, paid little attention to the actual workings of a socialist economy (Vaughn, 1980). All of this changed dramatically in 1920 when Austrian economist Ludwig von Mises published his “Economic Calculation in the Socialist Commonwealth” (Mises, 1920) in which he plainly asserted that “rational economic calculation” (i.e., efficiency in production) would be “impossible” under socialism without a set of market-generated prices to guide the use of factors of production (Mises, 1920). From a starting premise that productive resources are owned by the state under socialism, Mises argued that there could be no market-generated prices for them. Without money prices to inform them of relative values, it would be impossible to choose among the vast array of technologically feasible input combinations. In effect, central planners would be left “groping in the dark,” unable to decide on the most efficient methods of production (Pasour, 1983). And without productive efficiency, the socialist economy would be unable to deliver on the promises of material well being made by its proponents.
Mises’ challenge infuriated many socialists and generated a large number of replies by Italian and German writers during the 1920s (Hoff, 1949). The definitive response to his initial thesis, however, was published some sixteen years later by socialist economist Oskar Lange (Lange, 1936). Lange, while agreeing with Mises that prices are necessary for rational calculation, argued that prices do not necessarily need to be generated in competitive markets. The Central Planning Board (CPB) could simply issue “quasi-prices” to the state-owned firms, who would then treat these as “given” parameters in decision making and proceed to minimize total costs and set prices equal to marginal cost. Lange argued that using a “trial and error” procedure that would adjust prices up or down in order to eliminate shortages and surpluses could reflect underlying input scarcities. As Henry D. Dickenson had shown earlier in 1933 (Caldwell, 1997), by solving a set of n Walrasian simultaneous equations in n variables the Central Planning Board could simply “find” the same prices as markets and mimic the workings of a perfectly competitive economy in all desirable aspects (Chaloupek, 1990). At the time it was published, Lange’s reply was widely seen as a refutation of Mises objections and as having “proved” the theoretical possibility of socialist calculation.
In the 1930s, Austrian economist Friedrich A. Hayek joined Mises in his attacks on socialist planning. Initially, Hayek focused on the difficulties of “the mathematical solution” proposed by Dickenson, including the huge amount of information that would need to be gathered, the difficulty of forming the correct system of equations, and the problem of solving hundreds of thousands of equations not once but continually as economic conditions changed. Hayek’s other objections included such things as incentive problems for managers, who would tend toward risk aversion when making investment decisions.
Hayek began his reply to Lange’s model of “trial and error” market socialism in 1940, and subsequently expanded his arguments in a series of articles up through the 1960s (Caldwell, 1997). The thrust of his argument was that Lange and his fellow socialists had become excessively preoccupied with the use of the static equilibrium models that were (and still are) the framework of neoclassical economic theory. Lange’s exposition of the workings of market socialism relied on all of the crucial “data” being “given” to the CPB, when in fact they are not known to anyone in their totality. While the models used by the socialists were not logically contradictory, Hayek argued that they were being misapplied as they failed to capture the actual process by which markets bring information about such things as least-cost production methods and available supplies to bear (Vaughn, 1980). Hayek’s arguments, however, were rarely taken seriously by the profession, and were seen as a “second line of defense” and a retreat to perceived “practical” problems with socialism in light of Lange’s theoretical “proof” that planning could achieve the same level of efficiency as free and competitive markets. Until the 1980s, the defeat of the Austrians in the SECD was accepted as conventional wisdom among most mainstream economists (Lavoie, 1981).
II. Understanding the Austrian Position
If for no other reason, the history of the SECD is a fascinating one because the arguments presented for and against socialism were made entirely on scientific grounds without any recourse to the profound moral and ethical questions involved. Because the economic issues involved were scientific in nature, it is remarkable that the two sides were unable to reach any sort of mutual understanding during some twenty years of argument (Kirzner, 1988). Professor Israel Kirzner has argued convincingly that this misunderstanding was due to the fact that the two sides of the SECD were unwittingly working within vastly different theoretical frameworks (Kirzner, 1988).
The socialist economists as well as nearly all of the economics profession at the time interpreted the Austrians’ arguments within the conventional neoclassical framework, in which assumptions of “given” prices and costs are then used to mathematically describe a stable equilibrium in detail. The framework of the Austrian economists, however, contrasts starkly with this view of the economy.
The central economic problem in the Austrian view is not how to allocate a fixed “pie” of resources among competing uses, but rather how to mobilize highly diffuse and incomplete knowledge in order to achieve economic coordination (Adaman & Devine, 1994). To the Austrians, the role of markets is one of discovery rather than allocation. Much of the knowledge that is utilized in production in a market economy is not scientific in nature, but rather is knowledge of particular time, places and circumstances. Many production techniques and possibilities simply do not exist until they are uncovered during the competitive process, which does not exist under socialism. So called “tacit” or qualitative knowledge about particular firms and resources presents additional problems, as they cannot be communicated objectively as statistics to the CPB. By its very nature, this crucial information is highly dispersed and fragmentary, and therefore is not ever known to any one agent in the economy (Hayek, 1945).
According to Hayek, the way that these facts are discovered and put to best use is through the process of entrepreneurs seeking out temporary profit opportunities. Without this competitive process, central planners have no way to discover, much less aggregate in statistical form for use, the data that is the foundation of neoclassical equilibrium analysis. Assuming that the “data” are simply “given” inputs into the CPBs calculations ignores the imperfections of knowledge, the dispersal of information, and effectively assumes away the entire economic problem (Hayek, 1982). During the SECD, Lange pointed out that the same information that guided production in a market economy would be available in a socialist economy (Lange, 1936). To Hayek, saying that this information was “available” was just the beginning of the problem of demonstrating the possibility of a non-market economy.
The model of “perfect competition” that is the core of neoclassical welfare economics was also seen by the Austrians as a misleading description of what actually occurs in a market economy. The concept of equilibrium, argued Hayek, “presupposes that the facts have already all been discovered and competition therefore has ceased. (Hayek, 1978a, p. 259)” In particular, the traditional model of perfect competition says nothing about how firms ever come to raise or lower prices, for example, when they are assumed to be externally determined constants. Within the Austrian framework, competition is understood as a verb rather than as a noun, and is seen as leading to the continual discovery of dispersed information that is by definition impossible to collect and transmit to any central planning authority, even in principle (Bradley, 1981).
Other Austrian arguments against the possibility of central planning included the slower speed of adjustment to a changing environment and the problem of specifying exactly what a “product” is for accounting purposes. Hayek argued that only if prices under socialism were adjusted immediately when imbalances were perceived could they approximate market outcomes. The likelihood that they would instead be changed only at accounting intervals implied that shortages and surpluses would be unavoidable. Because there are more dimensions to products than simply price and quantity, central planners would also be unable to accurately take all characteristics of products into account, especially for physically unique goods (Vaughn, 1980). All of this led the Austrians to conclude that a coherent socialist economy was a “purely fictional” construct of its proponents (Hayek, 1982).
III. Implications of the Austrian Critique
The Austrian focus on disequilibrium behavior and information imperfections was (and still is) difficult to interpret within the neoclassical framework of perfect information that was popular at the time of the SECD. As a result, the sophisticated insights of Hayek in particular were interpreted in a very simplistic way within the economics profession, and did not receive the careful attention that they in fact deserved until decades later. Because neoclassical economics and market socialism share a common set of analytical tools and assumptions, it was inevitable that the unfamiliar approach of the Austrians would be seen as ineffective and unconvincing.
The following section examines two key implications of the Austrian arguments from the SECD. The first implication is that the Austrian arguments highlight serious errors within existing proposals for market socialism that are based on a supposed refutation of the Austrian position. Until the early 1980s, the standard account of the SECD held that the Austrians had failed to prove their case. This is not surprising given that all of the major discussions of the debate until that time were from a decidedly neoclassical view. While leading scholars such as Joseph Schumpter and Frank Knight were responsible for popularizing this “standard account” of the SECD, their focus on the non-essential arguments of Hayek and Mises fails to capture the uniqueness of the Austrian approach and interprets their objections through the lens of neoclassical assumptions.
In 1978, Friedrich Hayek reflected on this misunderstanding when he wrote, ”...after vainly waiting for upwards of 40 years to find a respectably intellectual defense against objections raised to socialist proposals, I am becoming a little impatient.” (Hayek, 1978b, p. 124)
Indeed, the most important objections to central economic planning raised by Hayek and Mises have yet to be adequately answered by proponents of market socialism (Lavoie, 1981). Most of the attempts to answer the Austrians’ claims have focused on the non-essential parts of their critique of central planning. By pointing to recent advances in computer technology, for example, advocates of market socialism claim to have refuted Hayek’s entire position by showing that data transmission and “equation solving” would not pose serious problems under socialism (Cottrell and Cockshott, 1993). Hayek’s central argument, however, was not so much that a socialist economy could not transmit the necessary data, but rather that it could not generate it to begin with. Without the competitive process of discovery and innovation, a socialist economy would have available only a small fraction of the knowledge that is utilized in a competitive economy. The task faced by proponents of market socialism is to explain exactly how spontaneous discovery is to occur within a planned economic system.
In a somewhat more sophisticated attempt to address this argument, Cottrell and Cockshott (1993) have suggested that innovation and “discovery” are no more likely to occur in competitive markets than under a species of market socialism featuring an “innovation budget” that would allocate “experimentation funds” to those with new ideas for products and processes. While a complete discussion of the issues raised by this proposal is beyond the scope of this paper, the “planning budget” solution clearly fails to address Hayek’s emphasis on spontaneous accidental discovery, and is based on rather unlikely assumptions regarding incentives and political allocation mechanisms.
Many other replies from branches of socialist economics such as “planometrics” have retained essentially static assumptions in their descriptions of what Hayek and others saw as a fundamentally dynamic process. The Austrian argument about the dispersed, incomplete, and subjective nature of economic information is one that has not to date been fully addressed by proponents of market socialism.
The second implication of the SECD has to do with the methodology of neoclassical economics in general. It is no coincidence that (1) market socialism was developed by neoclassical economists, (2) that free-market neoclassical economists were unable to produce a theoretical case against central planning, and that (3) neoclassical economic theory has shown to be of limited value in reforming the former Communist states (Murrell, 1991). The common strain between these facts is that there are important institutions and processes in a competitive economic order that are assumed away within most general equilibrium models. The static approach of these models and their fixation on a unique and stable equilibrium misrepresents some very important aspects of reality. For example, only in the early 1980s did mainstream neoclassical economists come to recognize how the existence of historical accidents and “path dependence” may prevent actual competitive markets from reaching a unique “best” solution as the models predict. The work of Paul David and Brian Arthur on the “economics of QWERTY” was seen as a revolutionary insight into the way actual economies function (Krugman, 1994). Within the discovery and process-oriented Austrian framework, however, there is nothing at all remarkable about the way accidents of history can alter the course of economic development. The point is not that the approach of Austrian economics is without fault, but rather that the dynamic approach of Austrian economics is able to capture some key aspects of economic reality that are ignored in the static neoclassical worldview.
Another methodological insight of the Austrians is their treatment of knowledge as being dispersed and incomplete. The modern significance of this insight can be seen in the growth of the “economics of information” within neoclassical economics. Although Caldwell (1997) makes it clear that Friedrich Hayek did not directly participate in the development of the “economics of information,” a listing of his criticisms of market socialism—the failure to account for adjustment processes, the misunderstanding of the decentralized and incomplete nature of information, the lack of adequate incentive structures—reads like a research program for general equilibrium theorists since the 1950s (Vaughn, 1980). It is interesting and somewhat ironic that many mainstream advocates of the “economics of information” approach such as Joseph Stiglitz emphasize the inadequacy of general equilibrium theory in much the same way as the Austrians did (Caldwell, 1997). Although the contributions of Hayek in this area of economic theory have been categorized by some as “fuzzy” and incomplete, this is largely the product of an attempt to view the SECD backwards through the lens of contemporary agency and mechanism design models (Caldwell, 1997). Doing so results in a classic example of a “Whig” interpretation of history that fails to recognize that the Austrian theorists took a path that really was different from their neoclassical contemporaries (Caldwell, 1997).
The arguments of the Austrian economists during the SECD were not fully made explicit at the outset of the debate. Only as Hayek and others began to recognize the theoretical rift between themselves and their neoclassical counterparts did they refine their critique of market socialism. The result was a complex and subtle epistemological argument about the very nature of knowledge, competition, the price system, and economic methodology. This approach to economic theory is not merely an obscure branch of neoclassical economics, but rather represents a fundamentally different way of viewing the economy. The Austrian case against market socialism is grounded in an entirely different paradigm than that of neoclassical economics, to use the language of Thomas Kuhn, (Kuhn, 1962).
Clearly, there is a persistent communication failure between modern Austrians and the rest of the economics profession. The empirical failure of central economic planning in the late 1980s has led to a recent rethinking of socialist economics, and a renewed interest in the apparently prescient arguments of the Austrians made in the SECD (Barkhan and Roemer, 1993). The arguments of Hayek and Mises, however, have been interpreted by would-be reformers of market socialism within modern mechanism design and principle-agent models. These interpretations suggest that the Austrian critique was concerned mainly with the coordination of private information and on monitoring the behavior of agents. This is not, however, the most subtle reading of the work of the Austrians, and is typical of the communication problems. The core argument of Hayek and others that deserves attention is that market socialism as a method of organizing production would be unable to discover and make socially useful the dispersed, tacit, and ultimately subjective knowledge that is available for use within a competitive economic system based on private property rights. The inability of modern market socialists to answer this argument casts serious doubt on the practical workability of existing proposals for market socialism.
Adaman, F. and Devine, P. (Spring 1994). Socialist renewal: lessons from the calculation debate. Studies in Political Economy 43, p. 63-77.
Bardhan, P. and Roemer, J., eds. (1993) Market socialism: The current debate. New York: Oxford University Press.
Bradley, R. (Winter 1981). Market socialism: A subjectivist evaluation. Journal of Libertarian Studies 5 (1), p. 23-35.
Caldwell, B. (December 1997). Hayek and socialism. Journal of Economic Literature 35, p. 1856-86.
Chaloupek, G. (1990). The Austrian debate on economic calculation in a socialist economy. History of Political Economy 22 (4), p. 659-73.
Cottrell, A. and Cockshott, W. P. (1993). Calculation, complexity and planning: The socialist calculation debate once again. Review of Political Economy 5 (1), p. 73-112.
Hayek, F. A. (September 1945). The use of knowledge in society. American Economic Review 35 (4), pp. 519-530.
Hayek, F. A. (1978a). Competition as a discovery procedure, in Nishiyama, C. and Leube, K., ed. (1984). The Essence of Hayek. Stanford: Hoover Institution Press.
Hayek, F. A. (1978b). Socialism and science, in Nishiyama, C. and Leube, K., ed. (1984). The Essence of Hayek. Stanford: Hoover Institution Press.
Hayek, F. A. (1982). Two pages of fiction, in Nishiyama, C. and Leube, K., ed. (1984). The Essence of Hayek. Stanford: Hoover Institution Press.
Heilbroner, R. (January 23, 1989) The triumph of capitalism. New Yorker 64 (49), p. 98-109.
Hoff, T. (1949). Economic calculation in the socialist society. London: W. Hodge.
Kirzner, I. (Autumn 1988). Some ethical implications for capitalism of the socialist calculation debate. Social Philosophy and Policy 6 (1), p. 165-82.
Krugman, P. (1994). Peddling Prosperity: Economic sense and nonsense in the age of diminished expectations. New York: W. W. Norton & Co.
Kuhn, T. S. (1962) The structure of scientific revolutions. Chicago: University of Chicago Press.
Lange, O. (1936). On the economic theory of socialism: Part I. Review of Economic Studies 4 (1), p. 53-71.
Lavoie, D. (Winter 1981). A critique of the standard account of the socialist calculation debate. Journal of Libertarian Studies 5 (1), p. 41-73.
Mises, L. (1920). Economic calculation in the socialist commonwealth, in Hayek, F. A. (1935). Collectivist Economic Planning. London: George Routledge and Sons.
Murrell, P. (Autumn 1991). Can neoclassical economics underpin the reform of centrally planned economies?. Journal of Economic Perspectives 5 (4), p. 59-76.
Pasour, E. (Spring 1983). Land-use planning: Implications of the economic calculation debate. Journal of Libertarian Studies 7 (1), p. 127-38.
Vaugn, K. (October 1980). Economic calculation under socialism: The Austrian contribution. Economic Inquiry 18, p. 535-52.