Alchian’s Calculation Problem

Production, Information Costs, and Economic Organization,” by Armen Alchian and Harold Demsetz, is probably the former author’s most cited academic article (638 cites according to In it, he advances a theory of the firm that both builds from and adds on to Ronald Coase’s transaction cost theory of the firm. It may be surprising to find out that, in the same sense, Alchian and Demsetz’ theory adds on to Ludwig von Mises’ theory of economic calculation. The argument advanced in Alchian and Demsetz (1977) is that it can be cheaper to organize a firm as a means of calculating the marginal productivity of inputs, rather than rely on exchange across markets (i.e. the pricing process).

Before explaining Alchian and Demsetz’ idea in further detail, allow me to review Mises’ theory of economic calculation. Hopefully, my explanation will help emphasize the similarities in both theories. If you read this blog often, you will probably find the (next) paragraph on Mises repetitive.

A big question that early Marginalists were interested in answering is: how are factors of production (inputs) valued? We know that consumers’ good values are derived from their potential to directly satisfy ends. But, inputs are means to an end — they are not the final product. The answer is that the value of an intermediate good is derived from the final product. Austrians refer to this as imputation, whereas others may know it as derived demand. The follow-up question is, how are these values conveyed to us? For example, how does a producer really know how much each unit of input is really worth? Mises’ answer is that a money-using economy, with a competitive pricing process — where input owners and input buyers can bid and bargain against each other — will establish prices that help convey information on the value of the factors of production. In other words, competitive prices will approximate the marginal value of inputs; at least, they give an idea of what these marginal values might be.

This is exactly the same starting point for Alchian and Demsetz. Oftentimes, production requires what they refer to as the “team use” of inputs. They explicitly define “team use” as a situation where the product of two or more inputs is not a simple additive function (i.e. the marginal product of input plus the marginal product of the second input), but where the product is more than what could be produced by the inputs separately. In these situations, it’s very difficult to know how much any given input is responsible for the output. Further, if we’re talking about labor, or non-labor inputs with multiple owners, this difficulty in separating marginal product creates an incentive to shirk, because it’s difficult to adjust incomes to those input’s marginal product. One method of alleviating the problem is to organize production within a firm, where the purpose of the firm is to help meter (measure and control) the marginal product of the input.

The costs of measuring and/or controlling the marginal productivity of an input in team production are transaction costs. In a frictionless world, the price you pay for a good or service represents the other person’s opportunity cost — the opportunity they forgo to enter into exchange with you. Transaction costs, then, are costs that are not related to the other person’s opportunity costs. They are costs to using the pricing process. In the case of Alchian and Demsetz (1977), the relevant transaction costs are informational. Because the marginal productivity of an input isn’t just known — and a competitive price won’t reflect this value accurately, because of the incentive to shirk (free ride) —, in team production there exist positive transaction costs, which is the price of metering marginal productivity. The firm organizes in response.

One solution the authors offer — what they call the “classical firm” — is the manager, or someone who “specialize[s] as a monitor to check the input performance of team members” (p. 158). How does the firm monitor the monitor, so to speak? It can offer that person the residual between total output and the total costs of the inputs. There is therefore an incentive to maximize team production. Based on this person’s observations then, the return to each input can be changed, and the inputs themselves can be changed. Apart from the emphasis on (a different kind of) transaction costs, we see the Cosean element to the argument: the firm organizes to allocate resources without use of the pricing process (because, in some cases, this kind of organization is less costly, on net, than a decentralized market). In reality, firm hierarchy can be more complicated than this (and the authors do discuss some different, modern firm dynamics, under the lens of their theory), but the general concept remains the same.

Mises presented the pricing process as the solution to a specific information problem: calculating the rewards input users had to distribute back to the factors of production. Alchian and Demsetz, it’s true, are arguing that the pricing process, under certain conditions (i.e. team production), is not the most beneficial method of calculating marginal productivity. The incentive to shirk is a free riding problem, and this creates an opportunity for an externality: some of the costs of shirking are externalized to the input users or to the other inputs. Competitive prices may not reflect those goods’ marginal productivity. Therefore, alternative methods of economic calculation are organized: the firm. It, in my opinion, makes the theory of economic calculation that much richer, because originally the theory did not tackle certain real world frictions.

There is another Austrian theme that I interpret in Alchian and Demsetz (1977): the role of entrepreneurship. In Competition and Entrepreneurship, Israel Kirzner defines entrepreneurship as the task of being alert towards opportunities to profit, where profit is the residual between revenues and costs (i.e. a disequilibrium between the prices of outputs and the prices of inputs). The reality of disequilibrium — one friction Austrians have never shied away from — and the role of the entrepreneur is a central element in the original theory of economic calculation. Because prices will never fully represent a good’s marginal rate of substitution, there is an entrepreneurial role in bearing the risk to search for and take advantage of profitable opportunities. The entrepreneur acts as a coordinator. Within the firm, the concept remains the same. For Alchian and Demsetz, using Kirzner’s terminology, the monitor is an entrepreneur who earns a profit by best using a given pool of inputs in team production. The monitor is either rewarded or punished, through profit and loss, based on his ability to measure and control each input’s marginal productivity.

I’m sure others have already made the same inferences as me, but I have never read something that explicitly connects these three theories (Mises’ econ. calculation theory, Alchian’s and Demsetz’ theory of the firm, and Kirznerian entrepreneurship). I think it helps cast the development of transaction costs theories, including theories of the firm and, more broadly, New Institutionalism, in a new light, and it presents this latter work as an extension to important Austrian themes. It also goes beyond “core” Austrian theory — the pricing process — by providing a much more complete theory of the division of labor and exchange, where there are alternative institutional frameworks of allocation that the individual can rely on. It’s also interesting to note how the theory of entrepreneurship is present in all of these alternative frameworks.

  • Aleksandar Maksimović

    You might be interested in this paper that deals with an another Austrian interpretation of the firm contra Coase, using some Alchian and Demsetz insights among others.

    The Austrian School’s conventional theory of the firm is based on an attempt to synthesize Coase’s concept of the firm as a centrally planned hierarchy with the Austrian theory of entrepreneurship and monetary calculation. This paper is a critique of that program as well as an attempt to outline the alternative theory of the firm, one based on the synthesis of the contractual agency theory of the firm (Alchian-Demsetz, Jensen-Meckling) with the same Austrian arguments about entrepreneurship and calculation. The firm in this paper is defined as a nexus of various markets for goods as well as for labor and managerial services rather than as a hierarchy or “organization.” Both the neoclassical and Austrian critiques of the latter concept are utilized to prove that a clear distinction between the market and the firm cannot be established. That distinction is based on the misunderstanding of the firm’s dynamics as exclusively tied to the managing/transaction costs ratio as well as on the mischaracterization of inter-firm relations as commanding ones (Demsetz-Alchian, Jensen, Meckling, Fama, Cheung). On the other hand, the central planning view of the firm is equally at odds with the key Mises’s argument that rational economic planning is impossible in the absence of market prices (Mises, 1990). If this is so, the firm, as understood in a Coasian paradigm, would not have any reason to exist, or any reason to contribute positively to economic efficiency, because it would simply represent a centrally planned “island of incalculability” in a wider market setting (Rothbard, 2004). Since the firm is a nexus of various markets, its operation is contrary to the Coaseian assumptions led by the price signals. Only insofar as the internal firm’s operation is driven by the price signals can the firm be efficient.

    • JCatalan

      Thanks for the link. It’s always dangerous to comment on a paper after just reading the abstract, but something that comes to mind is Hayek’s theme of the communication of knowledge. Prices are an important method of communication, but there are other ones like language. There may be other ways of allocating resources other than the pricing process, and the relative efficiency may have to do with the size of that “market” (i.e the size of the firm).

    • daniilgorbatenko

      I think there is a way to reconcile the two views of the firm (the Coasean and the one Jancovic outlines), and that way is to get rid of their most questionable parts.

      We basically need to say that firms are hierarchies but not from the economic (but rather sociological) standpoint. From the economic standpoint, strictly speaking, there is no allocation of inputs within the firm because employees basically figure inside the firm as inputs in the production process, not as decision-makers of the use of the resources. On the other hand, a firm isn’t a nexus of contracting, it works by means of contracting but there is strictly speaking no contracting “within” the firm.

      The explanation for the existence of firms essentially needs to be an explanation for the existence of labor contracts (because this is the thing that differentiates a firm from a group of independent producers). And this explanation comes from the nature of the production processes which are organized as firms. They require high-level decision-making. This explanation is distinct from transaction costs because it implies that these processes can’t conceivably be organized otherwise (not as a firm) because economic calculation should proceed in a limited number of heads (perhaps, it’s always ultimately done in one head per firm). It can’t happen in the heads of thousands of employees.

      • genecallahan

        “if allocation is understood is change of the decision maker of the use of the input”

        But why in the world would we want to view allocation in that way? Typically, it means, “This coal can be used to make steel, or to make electricity: somehow this must be decided.”

        Firms must do this. Why isn’t this “allocation”? Given your re-definition, the socialist commonwealth NEVER allocates resources, and so has no calculation problem to confront!

        • daniilgorbatenko

          I use this word in such a way because the only way I can make sense of the Coasean idea that firms exist because market allocation of resources is ripe with transaction costs is to interpret it as confusing hierarchy in economic sense with a hierarchy in the sociological sense.

          If Coaseans mean by allocation what you outlined then the question becomes what Coaseans mean by purely market allocation of resources which would supposedly exist without firms. I submit that this idea is meaningless and so is the whole Coasean explanation of the existence of firms.

    • genecallahan

      “Only insofar as the internal firm’s operation is driven by the price signals can the firm be efficient.”

      I bet you have never worked in a firm, have you? Straight from school to academia, right?

  • Pingback: Kirzner and Alchian/Demsetz on the Firm | Econ Point of View

  • Brian Albrecht

    Another great post, although I disagree on some points regarding Kirzner and Alchian. The paper mentioned by Aleksandar seems of interest, but I think the synthesis of these authors is, if not impossible, way beyond my intellect. I started writing a comment here, but it turned into a whole post @ (sorry for the plug)

  • Pingback: Alchian’s Calculation Problem « Economics Info

  • genecallahan

    Very interesting, thought provoking post.

    However, I am skeptical of the idea that when there are transaction costs, then the price doesn’t reflect opportunity cost. Aren’t the transaction costs just the opportunity cost of engaging in the bargaining, rather than just hanging on to the damned thing? (Or foregoing having it, on the other side?)

    • JCatalan

      You’re right, all costs are opportunity costs. But, the MC schedule in a price model represents the other person’s opportunity cost (that you have to satisfy in the exchange). Transaction costs would be the opportunity cost of your own time/resources/etc.