Mises's Socialist Calculation Problem
What if giving the public the power to decide who crunches numbers meant that you totally destroyed the possibility of finding out what numbers are to be crunched in the first place?
We've all heard the cliché, "Socialism works well in theory, but can't be implemented in practice because of human failings." One twentieth century economist, Ludwig von Mises, proved that socialism can't even work in theory. In 1920 he published his most famous article, "Die Wirtschaftsrechnung im Sozialistischen Gemeinwesen" ("Economic Calculation in the Socialist Commonwealth;" Archiv für Sozialwissenschaft und Sozialpolitik 47 (1920), 86-121), sparking an intermittent but bitter debate that remains more than eighty years later. In 1932 he expanded his argument to a full-length treatise of over five hundred pages, entitled Die Gemeinwirtschaft; in 1936 this was translated into English as Socialism, and this is the title by which it can be found today.
But before we begin, I'd like to make a brief note on the scope of this argument. Mises's calculation problem applies to single agent, or any incorporated body or congress which acts in concert, insofar as they attempt to exercise discretion over the whole of an economy's resources – land, natural resources, machinery, labour, etc. – in order to reach any specific ends. For most purposes the aim in question is the common wealth or the improvement of the material standard of living of as many citizens as possible with as little sacrifice as possible, and in modern political terminology this is usually called "communism," because it is the system in which capital is owned by all citizens in common. Mises called it "socialism," because it is the socius or social body that is the owner of all goods. Mises's argument applies to any political economy that operates according to the principle of social ownership of all capital, whether it be of the Soviet model, social democracy, anarchism, or any other kind of arrangement.
Today we normally reserve the name "socialism" for mixed economies, where the public body reserves the right in principle to intervene in the economy but generally delegates it to private citizens. The calculation problem can apply to mixed economies in some cases. Insofar as the public body in a "mixed" economy actually attempts to exercise the discretion it has reserved for itself, it suffers from both the calculation problem and from the results of undermining the market process. Insofar as the public body instead delegates decisions to private citizens who also bear complete civil and pecuniary responsibility for the decisions they implement, calculation is no problem and the commonwealth can benefit from the market process.
For reasons that will be explained below, this theorem does not really apply to management of the small communes, kibbutzim and cooperatives that became popular in the 20th century ce. These are essentially extended domestic firms or households, and although managing a household of twenty is different from managing a household of five, the complexity of managing households is different from managing a whole nation – a "household" of thousands or millions.
1: Value, cost, and price
The science of finance, accounting and the full benefits of economics can only be realized when we've got something mathematical to deal with. While a small nuclear, extended or communal household or farmstead could be more or less competently managed without recourse to mathematics by someone who knew what they were doing, any kind of manufacturing or more intensive agricultural enterprise requires bookkeeping and assessment – capital accounting. But what is it that we account for? What are the terms that are used in these mathematical formulas and equations? Where do the numbers come from?
Before we can appreciate the problem of economic calculation, we must have a full understanding of what, exactly, it is that we're doing when we're performing economic calculation. For such an understanding, we need to know the difference between value, cost, and price.
"Value," in economics, doesn't refer to philosophical ideal values that supposedly transcend human interests and desires. Instead, what economists have to consider is human value insofar as it has economic effect.* Value, as an economic phenomenon, has certain properties. Four of these are important for us:—
1: Value is immanent, which means that although it is real and has concrete effects, and although we know our own values, the values of others can be known only through phenomena they cause; value can't be empirically measured or comprehended.
2: It is subjective, meaning that the value attached to any given object or objective is relative to the desires of particular agents or subjects whose constitutions determines it. Values are different for different people. Values can also change for one person over time.
Both of these properties would seem to spell doom for our prospects of making a science of economics, accounting or management. Fortunately, value also has two other extremely important properties:
3: Value is an intensity and is known through intension, not extension. It can't be measured with cardinal numbers and therefore can't factor directly into any kind of mathematical calculations. However, value can, to use the language of mathematics, be mapped ordinally. Given a set of mutually exclusive (though not necessarily exhaustive) possible situations, either an agent's desire will attach equally to all (either the agent values all equally, or is indifferent to them), or else will attach with different intensity to different options. Choice occurs when one object or objective is valued above all others with which it is incompatible, put together.
When an agent pursues one alternative to the exclusion of the rest, the chosen course is the "good" and those that weren't compatible with it are the "cost." Cost can be known extensionally, in foregone alternatives, missed opportunities, and declined options. The cost, at the time of action (though not necessarily in retrospect, when knowledge and values change), is always inferior to the good that is chosen over it, because even though an agent doesn't always choose what is in his best interests, he nevertheless always chooses what he desires. However, although these costs can be known, they can't be measured because they have no common terms; they're themselves only intensities of value or desire. We still don't have economic calculation yet.
4: Finally, it is transmissible; there are some objects and objectives such that value only attaches to them insofar as they is a means to other valued objectives. An object or activity which has no inherent desirability for the agent may nevertheless have value by its function in contributing to the acquisition of another object or the accomplishing of some other activity that the agent does value (either inherently or medially again).
Now, as I pointed out in my writeup "money," this is the origin of a general currency. Value can be transmitted along extremely extended chains of causality, sometimes attaching to otherwise useless items or activities. There may be a good which is acquired, not because its consumption is useful to the agent's objectives or fulfills the agent's desire, but because someone else will take it in exchange for some good that they have that is useful to the agent. Thus that good becomes a medium of exchange. If one good becomes more common (or "current") as such a medium, it becomes money – a substance capable of being measured in cardinal units. This is price: the ratio of the units of money that can be exchanged against a given number of units of specific capital.
Prices are the numbers with which we calculate.
2: The socialist calculation problem
A human being or committee of human beings can roughly manage the chores of a large household or small commune in their heads without too much trouble. However, management becomes progressively more difficult as the tasks required of a firm or company extend in scope. The complexity of a country's manufacturing and distribution is exponentially beyond the powers of even the most intelligent party's powers of qualitative comparison. A national economy can't be run without calculation. Millions of people depend on agriculture, industry, and commercial distribution to live. Committing different resources (including labour) to different activities cannot be done on spot assessments; calculations with cardinal numbers is required. In short, prices.
As I said above, prices are ratios of interconvertibility; they are the assessment of different pieces of capital that could be converted to different tasks, reckoned in common terms. The cardinal numbers associated with each resource can't be arbitrary; they must reflect the contribution that resource is perceived to make to the pursuit of the tasks in question – and ultimately, to social prosperity in general.
In the market economy, there is a feedback mechanism whereby the ratios of interconvertibility are determined by the balance of demand – which is what we're trying to fill – and supply – which is, as far as goods requiring management are concerned, limited.** Units of currency are proportioned against units of the different resources based on the utility of the different lines of production to which those resources could be applied and the prioritizations of the different lines of production that could make demands on a resource that can't be applied to all lines at once, and on the amount of capital that we can actually use. In other words, ratios of exchange on a stock market (including a labour market).
However, within a single corporate entity – be it one person's holdings, a firm's holdings, or a government's holdings – no exchange takes place. It can't take place, because those goods aren't for sale and don't need to be bought – the entity already owns them. So these goods are never exchanged, and therefore never assessed against a medium of exchange.
This isn't much of a problem for a firm within a market society managing intrafirm transfers and applications, because the firm can compare their resources to those that are being traded on the market. The managers of the firm compare their capital to similar capital that is being bought and sold on the market, and use the prices that are finalized in those transactions as data for calculation. But the larger the owning body and the more heterogeneous the resources under its control, the more difficult this becomes, because the resources being managed are too different from the resources being assessed on the market. This is especially true of land, whose soil can have different compositions and which are subject to different climates in different parts of the world; it's also true of any machinery of different models or made of different alloys with different stresses, labour forces with different skills and different interests – even buildings of different sizes and different shapes.
In Communist countries, like China, Cuba, or the former Soviet Union, bureaucrats have a rough idea of what an industrialized nation needs: It needs some farms, it needs some railways, it needs some factories. But how many farms, growing what? How many railways, going where? How many factories, producing what? To figure this out, they have no way to prioritize, because their capital is not subject to evaluation on a market. So they looked at the proportions in America and England – how many farms do they have, and what are they growing? How are their rails organized? What do their factories make? Well, they say, we'll just try to do roughly the same thing. That's how they operated, and it worked to some degree – it wasn't pleasant, but it kept the state machinery going.
The reason it wasn't pleasant is because the numbers they were trying to use in their calculations were geared to different land, different geography, different climates, different resources and different labour pools. The crops that are good to grow in one country, the railway connections that are efficient and profitable, and the commodities manufactured there, are geared to the circumstances and values of that country; those numbers, which are prices, arise in a market in which only citizens of that country participate and only their goods are being assessed.
That means that when you try to apply those numbers to Cuban or Chinese or Soviet circumstances, because of the differences in climate and soil qualities and different lays of the land and natural resources and even skills in the labour pool, you're trying to compare apples and oranges. And that means you are totally unable to take advantage of your circumstances and opportunities. In fact, the greater the difference between your own circumstances and those of the country from whose market you're trying to steal cheat figures, the worse the results will be when you try to apply those figures in your own decisions. You have no mechanism for telling what are good figures and what are bad ones. You just kinda have to guess.
You can run a household on guesswork and vague valuations. But you can't run a country.
Under a single world socialism, or in completely autarkic socialist countries larger than a few households that attempt to manage their resources in total contempt and ignorance of market data, the effect is a total breakdown of all human productive enterprise. Even the most benevolent party would lack the omniscience required to commit the right resources to the right lines of production in the absence of any kind of data regarding what the citizens want and what we have on hand to provide them with that, which is what prices are.
Ratios of interconvertibility can only be determined by actual conversions. Conversions, the barter of capital against currency, can only occur between owners. No conversions, no exchange; no exchange, no ratios; no ratios, no cardinal numbers; no cardinal numbers, no calculation; no calculation, no management; no management, no economy. If financial calculation and capital accounting become impossible, people starve, because society can't mobilize to grow food, process it, or get it to them; those that survive will often be left to the elements because the society can't mobilize to build them houses or get them clothes. The basic necessities of life can't be provided – how much less so any sort of luxury or leisure.
3: Some implications and consequences
I've already dealt with the matter of the benefits of the market process and the harmful consequences of trying to tinker with that process. In this node, I've dealt with the disaster that follows when a society tries to replace the market with something else. This strips all controversy from the matter of public policy on political economy. We have only two real choices, public or private discretion over scarce goods; to the extent that the former is exercised, we know that harm and disaster follow, and to the extent that the latter is a society's guiding principle, cooperation and mutual benefit follow.
An interesting but little-explored consequence of the calculation problem is that it puts to rest one of the main criticisms of laissez-faire – the objection that there is no natural limit to the size of a firm's holdings or the power it can exercise over property it accumulates. The fear is that one corporation will become so powerful, by buying up the land and maintaining its property rights over that land to issue edicts to its tenants, that it would essentially become a new totalitarian government.
This argument is fatally deficient, because even in the absence of legal restrictions on how large a firm could become, in the absence of special statutory privileges firms do have a natural limit to how big they can become.† Firms can be enormous, but they can't function as absolute monopolists. The calculation problem doesn't just apply to governments; it applies to any single agency that attempts to exercise discretion over all capital. As a single owner begins to gobble up resources – especially land but also machinery or mineral resources – it comes increasingly to suffer from the calculation problem, because it slowly loses reference to its assessment mechanism. The numbers it uses for its cost center accounting become more and more detached from the reality of supply and demand, and therefore its ability to exploit its resources begins to wane. Eventually it would reach a point at which it was more profitable to disintegrate the company than to try to press on; even if it tried, it would eventually lose the ability to continue gobbling up resources. There are a couple of resources that could, theoretically, be monopolized – the rarer mineral deposits, for example – but these have minimal impact on the economy. The resources that do have an impact on the economy would be excruciatingly difficult, if not impossible, to monopolize. If it were monopolized, the disaster that would follow would be so all-encompassing that it would liquidate the standard of living of even the owner himself. It just wouldn't be worth it.
Mises's calculation problem was specifically targetted to socialists, who believed that public ownership of all capital would improve the standard of living of the worker. However, it doesn't just apply to those who have the commonwealth in mind. It doesn't matter if the agency that attempts to exercise its control is a totalitarian bureaucracy, a democratic parliament, an oligarchical board of directors, or a single king. It doesn't matter whether the goals of the directing body is the general prosperity or their own prosperity. The only goal that could be attained is the absolute impoverishment of everyone, the utter breakdown of all human industry and the deaths of untold millions who depend on the division of labour to live – a group which could include the decision-maker himself.
The socialist calculation problem in no way reflects the problems that might face a group who decided to live or work communally within a market economy. Worker's cooperatives and communal households do not suffer from the socialist calculation problem, so long as they arise from the discretion of private owners of whom the co-op or commune is composed. They may suffer other problems, but not the socialist calculation problem. This problem also does not apply to businesses which happen to be state-owned but which are not legally monopolized by the state. It begins to apply in any industry that the government attempts to monopolize for itself, including agriculture or medicine. The damages are proportionate to the degree to which the monopolizing party attempts to exercise its control; it reaches the point of total economic destruction when it encompasses all capital.
* Although a complete consideration of market-influencing values must necessarily include whatever philosophical ideal values the market participants themselves may hold, economists can only consider such values insofar as the consumers act on them, not insofar as they are "values."
** If the supply in question weren't scarce, management wouldn't be a problem, because everyone would have far more of the good than they would ever know what to do with anyway.
† Modern corporations do have special statutory privileges, especially governing the administration of property and liability, and therefore the laws by which firms operate do not match the social function of firms.
Sources
Mises, Ludwig von. Human Action. Auburn, AL: Ludwig von Mises Institute, 1998.
—— Socialism. Indianapolis: Liberty Fund, 1981.
|