Calculational chaos

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #18846


    I’ve studied a bit on Mises’ economic calculation argument and I understand how prices coordinate resources in the economy and provide the info needed to allocate them efficiently. I recently came across arguments by socialists that scientists were able to invent technology that refuted Mises after he wrote his arguments. I have some limited knowledge of the calculation debate that arose after Mises wrote but perhaps I don’t have enough to provide the arguments to Completely refute this obviously flawed claim by socialists. Here’s an example of an argument I ran across:

    “In the 20s and early 30s when Mises first advanced his arguments, no such algorithmic techniques were known. But in 1939 the Soviet mathematician V Kantorovich came up with a method which later came to be known as linear programming or linear optimisation, for which he was later awarded the Nobel prize. This mathematically proved that it was possible to efficiently calculate the allocation of resources in a command economy.

    Open source software which will easily run a USSR scale economy on any personal computer, is now freely available online. Here’s a link:

    Mises was clearly, demonstrably, and unequivocally wrong about economic calculation, something only ancaps and libertarians still fail to acknowledge. It is intellectually dishonest to claim otherwise”

    So apparently they claim that there is technology that can run calculation in an economy through linear programming and alogorithms. This seems impossible but I can’t quite put into words why it is wrong. Perhaps it’s because the output target must be set in advance and this is completely arbitrary and should be subject to calculation as well. Or perhaps it’s because machine can’t act as entrepreneurs and innovate by forecasting and dealing with uncertainty of the future?

    Any help on the proper way to refute this claim would be very much appreciated!
    Thank you



    There are two lines of argument that were advanced against central planning in the calculation debate that began the 1930s: one from Mises and another from Hayek.

    Mises argued that without private property in the means of production there can be no prices for the factors of production and without prices for the factors of production the efficient technique of production in a given line of production cannot be determined. Different technically possible methods of production which use different machines, natural resources, and labor skills cannot be compared in the units of the inputs themselves to determine which technique uses the least inputs. An automobile, for example, can be built with a more capital-intensive or an more labor-intensive process, it can be powered by an internal-combustion or electric engine each of which requires a different configuration of inputs to produce, etc. The number of machines cannot be added to the number of labor hours to determine minimum input use. Linear programming does nothing to solve this problem. And, as Mises pointed out, this is a fundamental conceptual problem that exists regardless of the complexity of production across the entire economy.

    The other fundamental problem that Mises raised with central planning, which also exists regardless of the complexity of production throughout the economy, is the uncertainty of the future. Efficient use of resources cannot rely solely on existing knowledge. Production aims at the realization of ends in the future. Efficiency, therefore, requires accurate entrepreneurial anticipation of how given courses of action will play out into the future. This is why central planners cannot rely on the existing prices of capitalism on the day of the revolution to guide their production decisions. Linear programming, obviously, does not solve, or even address, this problem.

    On the calculation argument, take a look at Joe Salerno’s epilogue to Mises’s original 1920 article:

    Hayek introduced a second line of argument, namely, the problem of obtaining the information sufficient for the central planners to make efficient production decisions. His main point was that such information is, at least partially, tacit. It consists of production know how that is contingent on person, place, and time. There is no process, Hayek argued, except market prices that can compile and transmit this information to production decision- makers. Linear programming does not address this problem.

    Hayek also stressed that the knowledge necessary for central planners to make efficient production decisions is complex and constantly changing. There is no known method to compile such information into useful form except the pricing process of the market. The advocates of central planning took this last point, on which they thought faster, bigger computers could provide a solution, and asserting a theoretical solution declared victory over Mises and Hayek.

    Lucas Engelhardt has called into question the veracity of such a procedure on its own grounds:


    The way we calculate GDP is just plain silly. First off, income should be taken out of GDP altogether. What counts is the money people pay for services or the money they pay for products. When a product is paid for the full value of that product should be calculated as the GDP for the point of origin. This will require a much better tracking and labeling system, but it would be well worth it. The result would be a GDP lower than GNI, which I am sure is true.

    Statistics we need.

    Hypothetically speaking, archaeologists of a future civilization looking back on our own would track standard of living with 1 essential metric: the size per bed of farmhand dormitories. This is the standard of living of the poor.

    What they frequently substitute, since farmhand dorms disintegrate in the archeological record, is the size of churches. Churches are frequently made of durable material.


    Thank you so much for the great answer and resources.

Viewing 4 posts - 1 through 4 (of 4 total)
  • You must be logged in to reply to this topic.