The economic calculation problem (ECP) is a classic argument in favor of markets and prices and against central planning. It’s a powerful argument, not because it’s a reasonable or understandable argument, but because it’s so complex in its awfulness that it’s difficult to produce a convincing critique of it. The basic thrust of the argument is that rational provision cannot occur without some collection of market institutions (private property, quid pro quo trade, prices, finance, business). The ECP is essentially a dense collection of rationalizations for the price system, and it presumes that a price is a sort of “god numeral”, something that can inform one who reads it of many independent quantities simultaneously.
The process of “economic calculation”
The most basic statement of the economic calculation argument is that markets alone are able to perform a sort of “calculation,” which is the most critical function of the economy. The economy is supposed to be the sphere of social organization that is responsible for acquiring, using, transforming and provisioning resources. Like all practical problems, there are certain constraints on the solution space, and according to the calculation problem, markets are needed to deal with all of these constraints. These constraints all relate to steps in the process of fulfilling a human desire. I think the ECP, to some extent, attempts to relay these steps and constraints. However, I think that extent is as a thin veil over a politically-motivated attempt to affirm capitalism and deprecate socialism.
The first step in fulfilling a desire is the identification of that desire. Economists are primarily concerned with this step, how it affects the behavior of the economy, how it affects prices, how prices affect it, and so on. Markets are supposed to respond to the changing desires of people, and direct producers towards fulfilling those desires. This happens, according to economists, through the equilibrium process of supply and demand. According to the ECP, markets, and only markets, can do this. The primary constraint here is the “knowledge problem”, the problem of producers and consumers being unable to unwilling to share information about themselves needed for the production process.
The next step is validation, which is where the economic system needs to provide some constraints on the possible outcomes. Economists are most concerned with the constraint of scarcity—economists phrase this as “the fundamental problem of having limited resources and unlimited human wants”. Whether human wants are fundamentally and immutably unlimited is not self-evident, nor has this idea ever been experimentally or empirically verified. Regardless, there are indeed limited resources, and they must be marshaled towards fulfilling some subset of all human desires. A useful economic system must be able to constrain the uses of scarce resources and validate the ends to which they are going, within what is currently technically possible. However, I would argue that scarcity should not be the only constraint, that rationality and ethicality are further constraints that should be involved in this step, and I would argue that markets make rational and ethical decision-making nearly impossible.
Once the identified desire is validated, there is mobilization—what economists refer to as “incentivization”. Fulfilling desires using resources takes some human mobilization to realize those desires. It’s not clear one way or the other whether it is necessary for the economic system to actually mobilize people; the typical rationale that it is necessary is not based on empirical fact, but classist or anti-humanist prejudice; I believe the importance of “incentives” to modern economists is derived from the widespread belief of classical political economists in the need for statesmen to whip the indolent peasants into useful industry. Whether or not the economic system needs to actually mobilize people, it certainly needs to coordinate that mobilization into a coherent outcome.
Those who have been mobilized engage in the next step, realization. This is what’s typically known as “production”, but not all desires are for products. We need and desire people to maintain, to repair, to improve, to watch, to discover, and so on. The person who has been mobilized, who I will call the realizer, realizes that which is desired by the desirer. This is really the step which should need the least amount of control over it; it should be primarily shaped by the realities and technicalities of the problem being solved. However, in the market system, it is overwhelmingly shaped by the constraints of the market.
Finally, once the object of desire has been realized, there must be a process of resolution, which can involve different things, depending on what was realized. This could perhaps be the distribution of a product to its desirer, or it could be simply some sort of notification that the realizer is finished with their task or share of their task.
I would consider “economic calculation” to be anything that can successfully and rationally accomplish this process. This will look different according to scale. On the smallest possible scale, you would identify that you are hungry, walk to your backyard, pick a piece of fruit, and eat it. On the large scale, this could involve a computerized request for a product that goes through a complex series of messaging to different parts of industry in order to coordinate the production, assembly, and delivery of the component parts into the whole and to the desirer. On the grand scale, it involves coordinating the realization of all possible desires over the long run. The framers of the calculation problem would likely not agree with my statement of the problem, as we will see later.
The assumptions of the calculation problem
It’s not only the content of the economic calculation problem that supports this idea of requiring market institutions, but the wider framework of liberal economics that the calculation problem is couched in. The content itself is little more than: markets and only markets can calculate. We believe that content because we in turn believe that markets do indeed calculate. This belief is an assumption of the economic calculation problem and not an argument that is subsequently supported with reasoning or evidence. Rather, the problem states that market institutions are solely capable of fulfilling the set of criteria posed by the problem, and because other systems aren’t markets, they do not fulfill those criteria and thus couldn’t possibly perform economic calculation. Before we attack any of the assertions of the argument, we need to address its underlying assumptions.
The first and foremost assumption is that markets are capable of economic calculation. That markets can calculate is plausible to us because we are Western, our societies use markets and have grown prosperous over time. In contrast, other societies that were socialist or otherwise have failed to match the prosperity of western nations. For one, the failure to match Western prosperity is greatly overstated (along with Western prosperity itself); socialism has taken many nations from feudal societies to modern industrial societies in a much shorter time than capitalism has. Two, this argument depends on the false dichotomy between the polity and the economy, in other words, the idea that provision is apolitical. Economics and politics are so deeply interactive and interdependent I don’t think there’s much use in distinguishing them. For example, much of the reason that the West is so much more prosperous than elsewhere is because of centuries of colonialism, which is largely ongoing through modern forms of “free trade imperialism”. Three, many socialist countries used markets, yet this apparently does not count. The link between the failures of socialist nations and their lack of markets is far from being proven and settled.
That markets are capable of calculation is also plausible to us because of “supply and demand”. The idea of supply and demand posits that prices will respond to relative changes in scarcity or abundance. Consequently, people will choose to produce or consume things that are less or more abundant thereby increasing prosperity. The first problem here is that we already know for sure that not all prices are based on supply and demand—price controls, subsidies, and state administered prices exist in nearly every economy. In fact, this is what most prices are like—in studies by Means and Berle and subsequent researchers, 75% or more of businesses sampled set their prices according to something other than supply and demand.
Clearly, the logic of pricing goods and services is not some unalterable law of economics; it is variable, and so it is individual, social, or cultural. And if at most a quarter of prices are based on the logic that is supposedly responsible for markets’ ability to calculate, it becomes much harder to accept that there is a rational calculation occurring. On top of that, we know for a fact that people do not respond rationally to prices. Every single person reading knows an example: Charm prices. When something is $10, we perceive it to be significantly more expensive than something that’s $9.99. There are many ways that people do not deal with prices rationally, but I’ll leave the rest out of here.
Even for those prices that actually are based on supply and demand, are they actually participating in any sort of “calculation”? How does supply and demand actually affect a price? An increase in scarcity causes an increase in price, which causes people to use less of it, except when it causes people to use more of it (Veblen goods, Giffen goods), and vice-versa. If a price increasing by some indeterminate amount is caused by an increase in scarcity, then we are capable of measuring scarcity to at least some degree. If that is true, then why do we even need the price for this? The price can’t very useful in measuring scarcity or abundance, because, as economists believe, prices also measure value. The convolution of these two completely independent measurements into a single number, price, does not result in a value useful for rational calculation; this is a fallacy of composition.
Even more confusing is the supply-side contention of supply and demand. An increase in price, which somehow just happens (presumably due to the continuous bidding war of the infinitely many perfectly or “boundedly” rational buyers that economists assume in this model), causes producers to flock to the potential source profit. The producers, which are always capable of engaging in meaningful competition with the other producers, because there is no real product differentiation, add supply to the market, driving the price back down but supplying them with profit in the short term. Thus, profits are transient phenomena based primarily on the fluctuations of material and preferential realities, at least in this highly synthetic conception of price. It seems very evident merely from existing within the market economy that none of these things are the case. Businesses don’t generally compete over the same products, and people who study business learn to actively avoid doing so through product differentiation and branding.
We are all well-aware that real prices are not the result of a cosmic-scale bidding war, and so the only way “demand” is really measured is through purchases. In other words, production occurs before demand is measured, making the initial run of production based on an estimate. There is no reason that estimating the amount of things people want and then adjusting that guess based on the actual desire for them is a process exclusive to markets. Likewise, producers don’t need the price to tell them to produce more of something, because they can observe that people are buying it. Like demand, the observation that a large or growing number of people are using something is reason enough to supply more of it.
There is actually a third factor that we are all familiar with, which determines price to a large extent, and that is the input costs. Only a small minority of prices are less than or equal to their input costs, because without adding to the price, there is no accumulation, and accumulation is the goal of business. Businesses are continuously trying to reduce their costs, but this isn’t necessarily taking the form of being more efficient or judicious with the use of resources. It can equally involve using cheaper materials that reduce the quality or lifespan of their product, producing more pollution, lowering wages, or using slave labor. The ECP sometimes includes the statement that profit is a reward for producing something of value. Of course, we could come up with numerous activities that produce profit and not value, such as stealing a car and selling its parts. What businesses accumulate is not value, but power. Money is a token obtained either through force upon those less powerful or through subjugation to those more powerful. Those who get it obtain the privilege to access things that are valuable to them, such as consumption goods or influence over those less powerful.
Markets are not shared by equally powerful actors passively accepting the outcome of an impartial, apolitical, equilibrium process that gives us a view into the underlying material realities of industry and preferences of consumers. They are dominated by businesses that hold power over industry, individuals, and politics. Their ability to charge more for scarce and valuable things does not enable some sort of calculation process for rational coordination of industry and provisioning. Instead it enables a calculation process for accumulating power. The evidence is heavily in favor of this framework and not the economic calculation framework:
- Provision and the structure of industry is far from rational in reality. Market-driven economies are producing existential threats to humans, mass extinctions of animals and plants, rampant depletion of valuable resources, buildup of toxic pollution, continuous soil loss and nutrient loading, and a long, growing list of other iniquities. This is not only because of “distortions” by the state, but because of the inherently irrational nature of satisfying the whims of whoever has the wealth to back them, rather than using some rational means of deciding which desires are valid.
- Inflation has occurred continuously for as long as we’ve had a modern market economy. Companies that receive the most funding by financiers are those that best engage in inflation. Inflation is actually a redistributive process—when a price is inflated, the seller gets more of your money. Since all prices do not rise simultaneously, this still represents an increase in purchasing power for the seller.
- Nothing increases a company’s value to its investors more than mergers & acquisitions (M&A). If the ECP is to be held true and central planning is unequivocally worse than decentralization, then M&A activity should decrease the value of a company, because it will be less efficient at economic calculation.
- Dominant companies actively shape policy in their favor through lobbying, the placement of high-level employees into government bureaus, interstate/international presence, and non-governmental trade regulations.
- Companies lower their costs by using cheaper labor. Today, it isn’t the scarcity or abundance of labor that results in lower wages, but the greater or lesser institutional power between highly-organized and legally-protected workers and disorganized and unprotected workers. A Mexican worker is not cheaper because of supply and demand, but because there is comparatively less labor power and law in Mexico.
- Supply and demand is one of many possible logics behind pricing. It is a subset of power logic: Scarcity increases the power of those who possess the scarce resources. However, many other things can increase one’s power, such as command over larger worker hierarchies or favorable regulations.
- Actual capitalists are not concerned with bettering their fellow man, or producing the best product, or directing industry most efficiently. Their main focus is to accumulate wealth. Anything else that they achieve is merely instrumental to the accumulation of wealth.
The case for price being a stand-in or “signal” for logistical or industrial realities is weak. It has persisted so long only because of the institutional force pushing the idea to be true, to legitimize the institutions.
The failure of economics to make the case for this impartial equilibrium process invalidates most of the other claims of the ECP. It’s actually quite absurd to suggest that prices are the result of anything but a person deciding what they will be. Price is not a natural phenomenon, there is no way they could possibly result from a non-arbitrary process. Economists try to rationalize their claim through models, models which assume completely ridiculous conditions, such as infinitely many buyers and sellers, perfect information, perfect competition, no product differentiation, and rational actors with totally-ordered and known preferences. They ignore the significant overlap between producers and consumers, and treat a reduction in the price of a good to be identical to an increase in income. And despite all these inane assumptions, their models still don’t prove what economists want them to.
We can also examine the real-world problems with suggesting that competition shapes arbitrarily decided-upon prices into useful material-logistical-industrial “signals” that guide production to the most efficient and effective outcome. As suggested by the listed assumptions of economists, actual competition is very limited—businesses actively try to avoid competing over anything but revenue and capitalization. There are not infinitely many alternative buyers and sellers, so there is only a limited range of possible price outcomes, and these are not necessarily rational outcomes. Products are highly differentiated and products A & B are not in competition over price alone. Buyers and sellers are not rational, far from it, in fact: buyers do not have ordered or known preferences, they are not able to calculate the best possible decision for themselves, and they often make mistakes and have regrets; sellers do not know exactly what buyers want, they do not respond instantly to changes in cost or demand with price changes, and they exhibit risk-averse behavior which appears irrational to economists. Producers and consumers are not a completely distinct, bipartite set, a producer of one thing is a consumer of many other things. There is also a finite subset of all goods that any one person will consume, and that consumption is not necessarily indefinitely recurring. A reduction in price is therefore not equivalent to an increase in income; one may have already bought the thing whose price was reduced, one may not ever buy it, and one may be a laborer in the thing’s production, which raises the (likely) possibility that the reduction in price was due to a reduction in wages.
There are, of course, even wider assumptions that are more difficult to discuss here. One such assumption is that a non-market society must necessarily take the same shape as a market society. That is, industry is structured as a collection of producers and consumers, where each group is distinct and non-interactive, whose goal is to produce commodities for individual consumption to fulfill individual wants in a never-ending hedonic treadmill. Consumers are essentially mysterious black boxes, and to a large extent, so are producers. Each have selfish, conflictual interests that prevent any meaningful sharing or cooperation from taking place, even if said sharing or cooperation would achieve a substantially better outcome.
The arguments of the calculation problem
I’ve addressed many of the assumptions underlying the problem. By the standards of refuting an argument, I think showing the underlying assumptions to be false is sufficient. However, I once read (I believe in Capital as Power) that to defeat an enemy, you must attack its weak points, but to defeat an ideology, you must attack its strong points. I think the calculation problem represents a major part of the ideology supporting capitalism and other marketeer theories. So, to ensure I hit its strong points, I will simply attack from all sides.
Comparison of Heterogeneous Goods
The ECP states that prices are necessary to compare heterogeneous goods. The idea that prices are useful for comparing heterogeneous goods such as capital and labor is, like other claims, dependent on prices being the outcome of an impartial process. This comparison of heterogeneous goods, therefore, is based on the arbitrary choices of price-setters. There is no reason to believe that the arbitrary choices of a small subset of people, fed through a complex, abstruse, and highly-flawed process of numeric indirection is any more capable of comparing heterogeneous goods than people are themselves.
It’s not as if price really lets us circumvent the comparison of heterogeneous goods. The number of cases where two exactly identical items are on the market for different prices is almost or completely nonexistent. Even for something extremely standardized, such as a screw or bolt, there are differences in strength, weight, corrosion resistance, tolerance, and probably others I don’t know of because I am not a mechanical engineer. Every set of goods that we would practically compare are heterogeneous.
When we compare things based on price, we are not making any materially, industrially, or logistically useful decision; prices are based on power differentials far more than they are based on other factors like how much material they use, how beneficial or harmful they are, and so on. Most people call these non-material/industrial/logistical factors “distortions”, “externalities”, and so forth, depending on their character. Really, though, the political dynamic is the actual main factor in price, and the real-world factors such as material or logistic constraints are the “distortions”.
The behavior of a company in changing its prices in response to increasing or decreasing costs is a good example of this. Because companies almost always have an advantage over consumers, we can predict what will happen in either case: If production costs go up for the company, the price of their products will go up to pass the costs on to the consumers; If the production costs go down for the company, the price of their products will stay the same to pass the profits on to their executives and shareholders. And since every cost is eventually paid to a person, an increase in cost to some is always an increase in revenue to another.
This is really what inflation is: Whenever a cost increases, the market responds by distributing the increases in cost to those least able to resist it, which is invariably the consumer class. The factual reality of continuous, uninterrupted inflation for as long as it has been measured shows us that prices are always increasing, for everything, without end. The only thing that explains this is that prices and costs are, more than anything else, a matrix of relative power differences in an unending struggle to get ahead in the market.
So, when we compare “capital” (as in technomass, what economists and ordinary people consider to be capital) to labor using their respective prices, what we are really comparing is the amount of control needed to obtain a machine that does something (what is called the “subjective value” by neoclassicals or austrians) versus the amount of control needed to make workers do that thing themselves (what is called the “labor value” by classicals or marxists). This is certainly useful to the capitalist system, but its usefulness here has nothing do with any material, industrial or logistical reality. The “cost” of the technomass was, likewise, based on the amount of control needed to obtain its input materials and the amount of control needed to realize the product through labor. This cost comparison, therefore, is finding how to use the least force to gain the most power.
The influence of products’ heterogeneity on prices is but a mere “distortion”, one that is often erased in the actual price: Businesses are likely to follow the precedent of the market rather than account for differences in the outcome. Products end up in price classes, low-end and high-end, and rarely are the differences between low-end and high-end products too subtle to differentiate without a magic number being assigned to them.
Relating Utility of Consumption Goods to Capital Goods
For the above reasons, we should also reject the idea that the price system relates the utility of “capital” with the goods it produces. Prices are simply not measures of utility; they certainly cannot simultaneously be intelligible measures of utility, supply and demand, and the underlying costs of production. There are instances in the real world of several different quantities being formed into a single metric quantity. But these metrics would cease being useful if they were not formed according to a completely standardized definition, and were instead formed by the arbitrary decisions of unconnected and sometimes conflicting parties.
There are other reasons to reject the idea, as well: This argument is circular. A good cannot exist without having been produced, and if something is required to produce it, then we certainly can’t price the thing that produced it based on the price of the good it produces. We also, as I have shown, cannot simply defer this problem to a corrective equilibrium process. A number does not necessarily confer superior qualities to a corrective process, and so, again, if it were possible for this to happen, there is no reason to believe it could not occur without a god numeral.
What actually determines the “value” (as in the price) of technomass is capitalization: It is transformed into capital, by estimating the expected future value, and discounting it by the estimated risk, to determine the value in the present. This is not done in order to “maximize utility”, but to maximize revenue. In order for this to be equivalent to maximizing utility, there would have to be a direct relationship with revenue and utility. But, since we know that someone can steal a car and sell it for parts and earn revenue, we know this cannot be entirely true. In Capital as Power, Nitzan and Bichler make a convincing case for business revenue being entirely dependent on “strategic sabotage of industry”.
What capitalization accomplishes is finding those types of strategic sabotage that best increase control. Stealing a car carries a lot of risk, but using slave labor in a poor country on the other side of the world is not nearly as risky. Polluting is even less risky—in fact, the state defines the amount of pollution you’re allowed to emit for free. Companies will often emit more than this, because the expected gain might be far higher than the risk discounts. This strategy has transformed industry from being relatively decentralized, localized, low-cost (in terms of labor time), and aimed toward quality, to one that is heavily centralized, globalized, high-cost, and aimed toward output (and therefore revenue).
Coherent planning is somehow part of the ECP’s critique. I really don’t know how this made it past the draft stage, nor how anyone could take it seriously, especially today. Somehow central planning is simultaneously bad because it’s based on the decisions of a small group of bureaucrats and doesn’t represent the interests of everyone, and also bad because they cannot possibly form a coherent plan for production. The market, on the other hand, which has no central guidance whatsoever, is characterized primarily by ruthless competition, and is shaped solely by the god numeral, and is the only possible social structure that can form a coherent plan of production. Bureaucratic planners face a knowledge problem, the inability for central planners to obtain sufficient knowledge of the economy (this was of course posited before the development of today’s pervasive, global surveillance system where all your preferences are in fact stashed away in a database somewhere, but somehow it is still a critique that is taken seriously) but of course, the god numeral solves this problem as well, thanks to the magic of price equilibria.
A capitalist myth must of course include entrepreneurship: A socialist economy could never work because it lacks entrepreneurship. This argument states, essentially, that under socialism, there would never be any novelty—in other words, the classic “capitalism is responsible for innovation” argument. Without the profit motive, nobody would be willing to take risks to develop anything new, so everything would stagnate into inefficient, ineffective industry, bread lines, and empty toilet paper shelves. It’s hard to critique this argument seriously because it’s hard to take it seriously at all. Capitalist investment is highly risk-averse and highly results-driven, so innovation actually tends to be incremental, marginal, and limited to those that increase capitalist power. In fact, the ECP ironically states exactly this in another argument, which I will cover subsequently. Entrepreneurs may try to produce radical innovation, but it doesn’t matter how innovative they are if they never get permission from capitalists to scale up through investment. Further, the risk of incurring significant personal costs through a failed or unsuccessful capitalist venture, a risk that is actually very large (80% of start-ups fail within a few years), is far greater than that of a system where no such personal cost exists. There is no reason to believe that entrepreneurship could be a superior innovative mechanism.
Financial markets are another asserted requirement for effective economic calculation. Here it is supposed that financial markets are indispensible in avoiding errors in “capital investment” and coordinating planning through speculation. Again, given that more than three-quarters of businesses completely fail within years is some pretty strong empirical evidence against this claim. Moreover, this argument requires the conflation of “capital” and physical industry—for financial markets to relate to effective and error-free industrial planning, the “capital” that financial markets control must represent the industrial technomass they are supposed to be provisioning effectively. However, as Nitzan and Bichler show in Capital as Power, capital is formed through the strategic sabotage of industry, because capitalization is a measure of control and not production. In fact, the value of a business’s capital actually moves inversely to the value of its capital goods. In other words, the businesses with less technomass are more highly capitalized. While finance is risk-averse and aims to reduce risk, its intention is not to reduce the risk of misallocating industry in the production of goods, but to reduce the risk of misusing control in the accumulation of more control. In reality, finance leads to lots of useless production, such as the housing boom, and does not lead to intelligent planning decisions, such as maintaining the infrastructure necessary for society to function.
Based on the contradictions of empirical reality to the assertions of the ECP and the wider economic claims it depends on, as well as the incoherence of the argument in making contradictory claims, I don’t believe there is any reason to accept it as a valid criticism. Its arguments, once you strip away the liberal economic mysticism, all come down to: Systems that aren’t markets do not have prices; prices are necessary for economic calculation; therefore, systems that aren’t markets cannot perform economic calculation. In fact, I believe to accept it as a valid criticism would be to reject the Church-Turing thesis: If economic calculation can take place, then that calculation is computable. If economic computation is impossible, then so is economic calculation. Of course, if you read my other material, you will see that I do believe economic computation to be possible. Other forms of calculation or computation would in fact be far superior at coordinating industry.