Economic calculation problem

Since capital goods and labor are highly heterogeneous (i.e. they have different characteristics that pertain to physical productivity), economic calculation requires a common basis for comparison for all forms of capital and labour.

As a means of exchange, money enables buyers to compare the costs of goods without having knowledge of their underlying factors; the consumer can simply focus on his personal cost-benefit decision. Therefore, the price system is said to promote economically efficient use of resources by agents who may not have explicit knowledge of all of the conditions of production or supply. This is called the signalling function of prices as well as the rationing function which prevents over-use of any resource.[5]

Without the market process to fulfill such comparisons, critics of non-market socialism say that it lacks any way to compare different goods and services and would have to rely on calculation in kind. The resulting decisions, it is claimed, would therefore be made without sufficient knowledge to be considered rational.[6]

The common basis for comparison of capital goods must also be connected to consumer welfare. It must also be able to compare the desired trade-off between present consumption and delayed consumption (for greater returns later on) via investment in capital goods. The use of money as a medium of exchange and unit of account is necessary to solve the first two problems of economic calculation. Mises (1912) applied the marginal utility theory developed by Carl Menger to money.

Marginal consumer expenditures represent the marginal utility or additional consumer satisfaction expected by consumers as they spend money. This is similar to the equi-marginal principle developed by Alfred Marshall. Consumers equalize the marginal utility (amount of satisfaction) to the last dollar spent on each good. Thus, the exchange of consumer goods establishes prices that represent the marginal utility of consumers and money is representative of consumer satisfaction.

If money is also spent on capital goods and labor, then it is possible to make comparisons between capital goods and consumer goods. The exchange of consumer and capital/labor goods does not imply that capital goods are valued accurately, only that it is possible for the valuations of capital goods to be made. These are foundational elements of economic calculation, namely that it requires the use of money across all goods. This is a necessary, but not a sufficient condition for successful economic calculation. Without a price mechanism, Mises argues, socialism lacks the means to relate consumer satisfaction to economic activity. The incentive function of prices allows diffuse interests, like the interests of every household in cheap, high quality shoes to compete, among buyers, with the concentrated interests of the cobblers in expensive, poor quality shoes. Without it, a panel of experts set up to "rationalise production", likely closely linked to the cobblers for expertise, would tend to support the cobblers interests in a "conspiracy against the public". However, if this happens to all industries, everyone would be worse off than if they had been subject to the rigours of market competition. The latter forces producers to produce superior products at appropriate prices to please their consumers.

The Mises theory of money and calculation conflicts directly with Marxist labour theory of value. Marxist theory allows for the possibility that labour content can serve as a common means of valuing capital goods, a position now out of favour with economists following the success of the theory of marginal utility.

The third condition for economic calculation is the existence of genuine entrepreneurship and market rivalry.

According to Israel Kirzner (1973) and Don Lavoie (1985), entrepreneurs reap profits by supplying unfulfilled needs in all markets. Thus, entrepreneurship brings prices closer to marginal costs. The adjustment of prices in markets towards equilibrium (where supply and demand equal) gives them greater utilitarian significance. The activities of entrepreneurs make prices more accurate in terms of how they represent the marginal utility of consumers. Prices act as guides to the planning of production. Those who plan production use prices to decide which lines of production should be expanded or curtailed.

Entrepreneurs lack the profit motive to take risks under socialism and so are far less likely to attempt to supply consumer demands. Without the price system to match consumer utility to incentives for production, or even indicate those utilities "without providing incentives", state planners are much less likely to invest in new ideas to satisfy consumers' desires. Entrepreneurs would also lack the ability to economize within the production process, causing repercussions for consumers.[7]

The fourth condition for successful economic calculation is plan coordination among those who plan production. The problem of planning production is the knowledge problem explained by Hayek (1937, 1945), but first mentioned and illustrated by his mentor Mises in Socialism (1922), not to be mistaken with Socialism: An Economic and Sociological Analysis (1951). The planning could either be done in a decentralised fashion, requiring some mechanism to make the individual plans coherent, or centrally, requiring a lot of information.

Within capitalism, the overall plan for production is composed of individual plans among capitalists in large and small enterprises. Since capitalists purchase labour and capital out of the same common pool of available yet scarce labor and capital, it is essential that their plans fit together in at least a semi-coherent fashion. Hayek (1937) defined an efficient planning process as one where all decision makers form plans that contain relevant data from the plans from others. Entrepreneurs acquire data on the plans from others through the price system. The price system is an indispensable communications network for plan coordination among entrepreneurs. Increases and decreases in prices inform entrepreneurs about the general economic situation, to which they must adjust their own plans.

As for socialism, Mises (1944) and Hayek (1937) insisted that bureaucrats in individual ministries could not coordinate their plans without a price system due to the local knowledge problem.Opponents argued that in principle an economy can be seen as a set of equations. Thus, using information about available resources and the preferences of people, it should be possible to calculate an optimal solution for resource allocation. Friedrich von Hayek responded that the system of equations required too much information that would not be easily available and the ensuing calculations would be too difficult. This is partly because individuals possess useful knowledge but do not realise its importance, may have no incentive to transmit the information, or may have incentive to transmit false information about their preferences.[8] He contended that the only rational solution is to utilize all the dispersed knowledge in the market place through the use of price signals.[9] The early debates were made before the much greater calculating powers of modern computers became available but also before research on chaos theory. In the 1980s, Alexander Nove argued that the calculations would take millions of years even with the best computers.[10] It may be impossible to make long-term predictions for a highly complex system such as an economy.[11]

Hayek (1935, 1937, 1940, 1945) stressed the knowledge problem of central planning, partly because decentralized socialism seemed indefensible. Part of the reason that Hayek stressed the knowledge problem was also because he was mainly concerned with debating the proposal for market socialism and the Lange model by Oskar R. Lange (1938) and Hayek's student Abba Lerner (1934, 1937, 1938) which was developed in response to the calculation argument. Lange and Lerner conceded that prices were necessary in socialism. Lange and Lerner thought that socialist officials could simulate some markets (mainly spot markets) and the simulation of spot markets was enough to make socialism reasonably efficient. Lange argued that prices can be seen merely as an accounting practice. In principle, claim market socialists, socialist managers of state enterprises could use a price system, as an accounting system, in order to minimize costs and convey information to other managers. However, while this can deal with existing stocks of goods, providing a basis for values can be ascertained, it does not deal with the investment in new capital stocks. Hayek responded by arguing that the simulation of markets in socialism would fail due to a lack of genuine competition and entrepreneurship. Central planners would still have to plan production without the aid of economically meaningful prices. Lange and Lerner also admitted that socialism would lack any simulation of financial markets, and that this would cause problems in planning capital investment.

However, Hayek's argumentation is not only regarding computational complexity for the central planners. He further argues that much of the information individuals have cannot be collected or used by others. First, individuals may have no or little incentive to share their information with central or even local planners. Second, the individual may not be aware that he has valuable information; and when he becomes aware, it is only useful for a limited time, too short for it to be communicated to the central or local planners. Third, the information is useless to other individuals if it is not in a form that allows for meaningful comparisons of value (i.e. money prices as a common basis for comparison). Therefore, Hayek argues, individuals must acquire data through prices in real markets.[12]

The fifth condition for successful economic calculation is the existence of well functioning financial markets. Economic efficiency depends heavily upon avoiding errors in capital investment. The costs of reversing errors in capital investment are potentially large. This is not just a matter of rearranging or converting capital goods that are found to be of little use. The time spent reconfiguring the structure of production is time lost in the production of consumer goods. Those who plan capital investment must anticipate future trends in consumer demand if they are to avoid investing too much in some lines of production and too little in other lines of production.

Capitalists plan production for profit. Capitalists use prices to form expectations that determine the composition of capital accumulation, the pattern of investment across industry. Those who invest in accordance with consumers' desires are rewarded with profits, those who do not are forced to become more efficient or go out of business.

Prices in futures markets play a special role in economic calculation. Futures markets develop prices for commodities in future time periods. It is in futures markets that entrepreneurs sort out plans for production based on their expectations. Futures markets are a link between entrepreneurial investment decisions and household consumer decisions. Since most goods are not explicitly traded in futures markets, substitute markets are needed. The stock market serves as a ‘continuous futures market’ that evaluates entrepreneurial plans for production (Lachmann 1978). Generally speaking, the problem of economic calculation is solved in financial markets as Mises argued:

The problem of economic calculation arises in an economy which is perpetually subject to change [...]. In order to solve such problems it is above all necessary that capital be withdrawn from particular undertakings and applied in other lines of production [...]. [This] is essentially a matter of the capitalists who buy and sell stocks and shares, who make loans and recover them, who speculate in all kinds of commodities.[13]

The existence of financial markets is a necessary condition for economic calculation. The existence of financial markets itself does not automatically imply that entrepreneurial speculation will tend towards efficiency. Mises argued that speculation in financial markets tends towards efficiency because of a "trial and error" process. Entrepreneurs who commit relatively large errors in investment waste their funds over expanding some lines of production at the cost of other more profitable ventures where consumer demand is higher. The entrepreneurs who commit the worst errors by forming the least accurate expectations of future consumer demands incur financial losses. Financial losses remove these inept entrepreneurs from positions of authority in industry.

Entrepreneurs who commit smaller errors by anticipating consumer demand more correctly attain greater financial success. The entrepreneurs who form the most accurate opinions regarding the future state of markets (i.e. new trends in consumer demands) earn the highest profits and gain greater control of industry. Those entrepreneurs who anticipate future market trends therefore waste the least amount of real capital and find the most favorable terms for finance on markets for financial capital. Minimal waste of real capital goods implies the minimization of the opportunity costs of capital's economic calculation. The value of capital goods is brought into line with the value of future consumer goods through competition in financial markets, because competition for profits among capitalist financiers rewards entrepreneurs who value capital more correctly (i.e. anticipating future prices more correctly) and eliminates capitalists who value capital least correctly. To sum things up, the use of money in trading all goods (capital/labor and consumer) in all markets (spot and financial) combined with profit driven entrepreneurship and Darwinian natural selection in financial markets all combine to make rational economic calculation and allocation the outcome of the capitalist process.

Mises insisted that socialist calculation is impossible because socialism precludes the exchange of capital goods in terms of a generally accepted medium of exchange, or money. Investment in financial markets determines the capital structure of modern industry with some degree of efficiency. The egalitarian nature of socialism prohibits speculation in financial markets. Therefore, Mises concluded that socialism lacks any clear tendency towards improvement in the capital structure of industry.

Some academics and economists argue that the claim a free market is an efficient, or even the most efficient, method of resource allocation is incorrect. Alexander Nove argued that Mises "tends to spoil his case by the implicit assumption that capitalism and optimum resource allocation go together" in Mises' "Economic Calculation in the Socialist Commonwealth". Joan Robinson argued that many prices in modern capitalism are effectively "administered prices" created by "quasi monopolies", thus challenging the connection between capital markets and rational resource allocation.[14]

Socialist market abolitionists argue that whilst advocates of capitalism and the Austrian School in particular recognize equilibrium prices do not exist in real life, they nonetheless claim that these prices can be used as a rational basis when this is not the case, hence markets are not efficient.[15][16] Robin Hahnel further argued that market inefficiencies, such as externalities and excess supply and demand, arise from buyers and sellers thoughtlessly maximizing their rational interests, which free markets inherently do not deter. Nonetheless, Hahnel commended current policies pursued by free market capitalist societies against these inefficiencies (e.g. Pigouvian taxes, antitrust laws etc.), as long as they are properly calculated and consistently enforced.[17]

Milton Friedman agreed that markets with monopolistic competition are not efficient, but he argued that it is easy to force monopolies to adopt competitive behavior by exposing them to foreign rivals.[18] Economic liberals and libertarian capitalists also argue that monopolies and big business are not generally the result of a free market, or that they never arise from a free market; rather, they say that such concentration is enabled by governmental grants of franchises or privileges.[19][20] That said, protectionist economies can theoretically still foster competition as long as there is strong consumer switching. Joseph Schumpeter additionally argued that economic advancement, through innovation and investment, are often driven by large monopolies.[21]

Allin Cottrell, Paul Cockshott and Greg Michaelson argued that the contention that finding a true economic equilibrium is not just hard but impossible for a central planner applies equally well to a market system. As any universal Turing machine can do what any other Turing machine can, a central calculator in principle has no advantage over a system of dispersed calculators, i.e. a market, or vice versa.[22]

In some economic models, finding an equilibrium is hard, and finding an Arrow–Debreu equilibrium is PPAD-complete. If the market can find an equilibrium in polynomial time, then the equivalence above can be used to prove that P=PPAD. This line of argument thus attempts to show that any claim to impossibility must necessarily involve a local knowledge problem, because the planning system is no less capable than the market if given full information.

Don Lavoie makes a local knowledge argument by taking this implication in reverse. The market socialists pointed out the formal similarity between the neoclassical model of Walrasian general equilibrium and that of market socialism which simply replace the Walrasian auctioneer with a planning board. According to Lavoie, this emphasizes the shortcomings of the model. By relying on this formal similarity, the market socialists must adopt the simplifying assumptions of the model. The model assumes that various sorts of information are given to the auctioneer or planning board. However, if not coordinated by a capital market, this information exists in a fundamentally distributed form, which would be difficult to utilize on the planners' part. If the planners decided to utilize the information, it would immediately become stale and relatively useless, unless reality somehow imitated the changeless monotony of the equilibrium model. The existence and usability of this information depends on its creation and situation within a distributed discovery procedure.[23]

One criticism is that proponents of the theory overstate the strength of their case by describing socialism as impossible rather than inefficient.[24][25][26] In explaining why he is not an Austrian School economist, anarcho-capitalist economist Bryan Caplan argues that while the economic calculation problem is a problem for socialism, he denies that Mises has shown it to be fatal or that it is this particular problem that led to the collapse of authoritarian socialist states. Caplan also states the exaggeration of the problem; in his view, Mises did not manage to prove why economic calculation made the socialist economy 'impossible', and even if there were serious doubts about the efficiency of cost benefit analysis, other arguments are plentiful (Caplan gives the example of the incentive problem).[4]

Joan Robinson argued that in a steady-state economy there would be an effective abundance of means of production and so markets would not be needed.[27] Mises acknowledged such a theoretical possibility in his original tract when he said the following: "The static state can dispense with economic calculation. For here the same events in economic life are ever recurring; and if we assume that the first disposition of the static socialist economy follows on the basis of the final state of the competitive economy, we might at all events conceive of a socialist production system which is rationally controlled from an economic point of view."[1] However, he contended that stationary conditions never prevail in the real world. Changes in economic conditions are inevitable; and even if they were not, the transition to socialism would be so chaotic as to preclude the existence of such a steady-state from the start.[1]

The purpose of the price mechanism is to allow individuals to recognise the opportunity cost of decisions. In a state of abundance, there is no such cost, which is to say that in situations where one need not economize, economics does not apply, e.g. areas with abundant fresh air and water. Otto Neurath and Hillel Ticktin argued that with detailed use of real unit accounting and demand surveys a planned economy could operate without a capital market in a situation of abundance.[28][29]

In Towards a New Socialism's "Information and Economics: A Critique of Hayek" and "Against Mises", Paul Cockshott and Allin Cottrell argued that the use of computational technology now simplifies economic calculation and allows planning to be implemented and sustained. Len Brewster replied to this by arguing that Towards a New Socialism establishes what is essentially another form of a market economy, making the following point:[30]

[A]n examination of C&C's New Socialism confirms Mises's conclusion that rational socialist planning is impossible. It appears that in order for economic planners to have any useful data by which they might be guided, a market must be hauled in, and with it analogues of private property, inequality and exploitation.[31]

In response, Cockshott argued that the economic system is sufficiently far removed from a capitalist free-market economy to not count as one, saying:

Those that Hayek was arguing against like Lange and Dickinson allowed for markets in consumer goods, this did not lead Hayek to say : Oh you are not really arguing for socialism since you have conceded a market in consumer goods, he did not, because there remained huge policy differences between him and Lange even if Lange accepted consumer goods markets. It is thus a very weak argument by Brewster to say that what we advocate is not really socialist calculation because it is contaminated in some way by market influences.[32]

Leigh Phillips' and Michal Rozworski's 2019 book The People's Republic of Walmart argues that multinational corporations like Walmart and Amazon already operate centrally planned economies in a more technologically sophisticated manner than the Soviet Union, proving that the economic calculation problem is surmountable.[33] There are some contentions to this view however, namely how economic planning and planned economy ought to be distinguished. Both entail formulating data-driven economic objectives but the latter precludes it from occurring within a free market context and delegates the task to centralized bodies.[34]

Karras J. Lambert and Tate Fegley argue that artificial intelligence systems, no matter how advanced, cannot assume the role of central planners because they do not fulfill the prerequisites of effective economic calculation. This includes the ability to convert the ordinal preferences of producers and consumers into commensurate cardinal utility values, which are available and agreed upon, and forecast future market interactions.

One reason includes how they are dependent on Big Data, which in turn is entirely based on past information. Hence, the system cannot make any meaningful conclusions about future consumer preferences, which are required for optimal pricing. This necessitates the intervention of the programmer, who is highly likely to be biased in their judgments. Even the manner by which a system can "predict" consumer preferences is also based on a programmer's creative bias. They further argue that even if artificial intelligence is able to ordinally rank items like humans, they would still suffer from the same issues of not being able to conceive of a pricing structure where meaningful pricing calculations, using a common cardinal utility unit, can be formed. Nonetheless, Lambert and Fegley acknowledge that entrepreneurs can benefit from Big Data's predictive value, provided that the data is based on past market prices and that it is used in tandem with free market-styled bidding.[7]