Books / Digital Text

III. Policy

11. Knowing and Entrepreneurship by Mateusz Machaj

We* understand knowledge as an acquaintance with various facts and natures of objects in the real world. By studying and investigating aspects of our lives we get to “know” certain things and we classify these inquiries into disciplines. We can widen knowledge in total by different methods. In order to achieve progress in gained knowledge we use dissimilar frameworks to learn mathematics, physics, economics, social relations, characters of our friends, or languages. It is also important that we can learn some of these things through different methods, especially different methods for different people, or different methods for the same people over time. One term “knowledge” is being used to deliberate in general about all those disciplines, yet this should not cloud first and foremost feature of knowledge: its heterogeneity.

The Austrian school has been mostly successful in economic theorizing because it realistically emphasizes heterogeneous nature of the world. Whereas various neoclassical schools, or their siblings, tend to homogenize economic phenomena, the Austrians tend to do the opposite. The prime example of the case is theory of capital, which in the Austrian version is built on the notion that capital goods do not have a common physical denominator (which could theoretically express its aggregated “amount”). Starting from such basic observation the Austrians were able to build their own theory of socialism and theory of the business cycle. As Roger Garrison notes (1992, p. 171, emphasis added),

If capital goods were wholly non-specific, if the collection of them were fully homogeneous such that any one capital good is a perfect substitute for any other, then production processes could proceed as if time ran both ways. A half-finished performance hall could be completed — with no effects on cost or construction time—as a bowling alley; the production process that yields musical instruments could — with an eleventh-hour change of mind — yield bowling pins and bowling balls instead.

Under homogeneous circumstances the issue of proper allocations would never have to arise, since every process would already be fully integrated and properly coordinated. The problem of the trade cycle would be nonexistent, since any inconsistency in the various diverse stages of production would be absent. Similarly any socialist economy would not fail at the basic problem of equilibrating the capital goods market, because optimal allocations of them would have already been chosen.1

Other important Austrian contributions are also more or less related to the issue of heterogeneity. For this reason it could even be seen as a typical feature of the modern Austrian economist’s toolbox. Austrians are different, because Austrians heterogenize.

The same approach to heterogeneity applies for different types of “knowledge.” A typical model breakthrough comes from Hayek’s example of a breakaway from the neoclassical approach. Hayek’s famous contribution comes from the analysis on how knowledge is “used in society” (Hayek 1945). Yet even though this analysis of complexity of economic phenomena is fruitful and worth of deeper studying, it (along with others) created a lot of side debates about the “knowledge” problem under hypothetical socialist order. We will attempt to refrain from settling those debates here. Our goal is to follow Hayek’s footsteps and to try to distinguish several types of knowledge. The goal can allow us to settle the definitional importance of knowledge for Mises’s argument about the impossibility of the rational allocation of resources under socialism.

Here we offer our (arbitrary) classification of knowledge, which, though not very rigorous, helps to navigate through the usages of the term in the calculation debate. It is important to keep in mind that we don’t want to completely classify various types of knowledge, but to envision how it relates to the socialist puzzle.

Objective “Technological” Knowledge2

The word “objective” seems suitable, because the main feature lies in the interpersonal aspect of this knowledge, which can be simply transmitted from one person to another. It is knowledge which is coded in textbooks and countless publications.3 Due to its specific “objectivity” it can be communicated between the people with the use of alphabet, algebra and other symbols. Without those symbols there would be no abstract thinking, and consequently man would still live in caves (Cassirer 1944, pp. 46–47). Objectivity is here to be understood as the possibility to be (potentially) universally recognized by any intelligent being, no matter what place and time one lives in. Due to language and objectivity of those statements knowledge can be transmitted (sometimes through the painful process of learning) between all intelligent (and sufficiently capable) individuals.

Such knowledge can include statements from all developed sciences be they empirical or non-empirical; mathematics and logic, physics and chemistry, climatology and biology, economics and sociology, politics and history, etc. Even though all those disciplines differ and use radically dissimilar methods, they can be grouped into one big family of objective Science. There are multiple examples of that knowledge such as (geology) “earth is not flat,” (biology) “spiders eat flies,” (physics) “the speed of light is constant,” (mathematics) “In Euclidean geometry parallel lines do not intersect,” (climatology) “Earth is warmer than it was 40 years ago,” (economics) “minimum wage leads to higher unemployment,” (history) “Julius Cesar did not invent the caesar salad,” and so forth.

The important fact is that none of those statements has to do with distinct characteristics of the particular being who is proposing them. They are as general as possible and can be presented by a male teenager in Africa, a female doctor in Germany, or retired astronaut in the Moon. Also they are conditioned by the concept of Wertfreiheit. They are value-free. Their most important feature is correctness or incorrectness, no matter the values, opinions and views of the person proposing them. During the socialist calculation debate such knowledge was seen as easily obtainable and possessed by socialist bureaucrats.

Hayekian Knowledge

Human knowledge does not end with such universal and communicative observations. Not all the data can be effortlessly gathered in objectified and interpersonal form. Some information is hard or costly to transfer, so perhaps it seems sensible to use the name “transfer problem.” There exist two main reasons causing the transfer problem to arise. The first one is a subjective nature of individually “witnessed” data, which become a part of “tacit knowing.” Hayekian knowledge is perceived by an individual. At the same time it is being used by the individual even though she or he cannot formulate it explicitly and communicate it to another person. Tacit information is beyond textbooks and often beyond personal recognition of it (Polanyi 1966, p. xviii). Since personal boundaries are difficult to overcome such knowledge remains hidden behind individual barriers of the mind (Huerta de Soto 2010, pp. 27–28).

The second reason for the transfer problem is decentralized nature of Hayekian knowledge. At first it may seem that the reason is no different from the first one. Nevertheless the difference is important, because in the first case barriers have more to do with individual’s limits. In the second case scantiness of the data is an objective fact important for practical reasons. Because countless individuals are working with complex data, it is practically impossible for any isolated individual to gather their knowledge and unify it into one objective formula (even without admitting the “tacit” element of it). Hayek wrote extensively about its economic importance (see his illustrations in Hayek 1945, p. 522). He also made it an important part of the argument against market socialism model (Hayek 1940, pp. 192–93).

The examples of that knowledge could be “John knows unspoken local customs,” “Jack is the only one who knows how to talk to Mary,” “Martin knows how to start that machine,” etc.

Misesian Knowledge

An important question that arises with the title of the section is: why make a difference between “Hayekian” and “Misesian” knowledge? We are inclined to do so, because Mises emphasized the role of prices in the economy, whereas Hayek attempted to go further and focus on something underneath prices: production functions. For the former, prices per se were of interest. For the latter something more substantial had to be hidden behind those prices. Hence local conditions and knowledge about them was named by us as “Hayekian.” In the case of Mises, all aspects associated with calculation and prices will be seen by us as “Misesian” knowledge.

Therefore Misesian knowledge is strictly associated with monetary prices, and has three interrelated features in different time dimensions:

1. past prices and praxeological recapitalizations undertaken in the past,
2. current price offers,
3. “current allocation activities” (Salernian “social appraisement process”4).

Strictly speaking prices are ratios of exchange between sovereign owners in a realized transaction. In that sense they are phenomena of the past. Currently existing, though not yet realized, price offers are also often seen as “prices” of the present circumstances. Competing and cooperating owners of the factors of production establish a nexus of contracts that allows them to create the price structure. The phenomena of price activities arise in all instances of economic calculation — realized past prices , past actions undertaken to correct them, current price offers, and current actions based on calculation outcomes and expectations about future prices. Clearly, at every point in time part of the existing Misesian knowledge is objective and known, but part of it is always beyond human recognition, because it will be determined in the future: allocation activities undertaken after the acquaintance with price offers. That is why entrepreneurship consists of a combination of knowledge and ignorance.

Past prices can be observed and expressed in the form of statistics, therefore they belong also to our first category of knowledge (as we emphasized in the beginning we are not searching for fully non-overlapping definitions). Nevertheless past prices are only the beginnings of calculation, since they only reflect past choices conditioned by outdated anticipations (see Mises 1966, p. 330). The next constituents are price offers, which in the Misesian sense are not yet “prices.” They are offers formed today under current market conditions, which are different from the conditions under which past prices had been formed. Therefore in contrast to realized prices they convey some form of current information and views about the future. If someone theorizes about prices as information signals, currently available price offers perform this function (they are not strictly speaking prices as exchange ratios).5

Price offers and past prices close the category only of existing Misesian knowledge. Economic calculation involves economic activity under uncertainty, what results in changes of economic conditions and unexpected outcomes (with price changes). It is one thing to know past prices and current price offers, but it is another to act upon those prices. Past prices inform entrepreneurs about past events. Current price offers inform entrepreneurs about today’s conditions and expectations about the future. Potential, not realized, prices “transmit” correct and incorrect entrepreneurial anticipations about possible marginal valuations of resources they own. That is why they do not transmit strictly Hayekian “knowledge,” but can include entrepreneurial perspectives on Hayekian knowledge.

All knowledge associated with various past and present instances of monetary calculation is not sufficient for the market process to happen. The driving forces for it are allocation activities (part of yet non-realized Misesian knowledge of what would private entrepreneurs do). These are actions undertaken by entrepreneurs after recognition of current price offers (with considerations on past prices and recapitalizations). The central owner under socialism has precisely the following problem: he cannot know allocation activities based on current price offers.6 He is not in a position to recognize what private owners would do, and how they would exclude each other from the market process. He is able to gather data on past prices, or even price offers right before the complete nationalization of resources, but he cannot know which allocation activities would have been performed under private property. Even if he or she knew all the relevant Hayekian knowledge, it would not suffice to solve allocation problems under socialism, since all of the Misesian knowledge would have to be known. The activity of entrepreneurs is something which cannot be implicit in the informational parameters of any system of equations, or any prices based on past or current data (see Salerno 1994, p. 120).

Three distinctive examples of Misesian knowledge could be: (1) “Lemons sold for 3 dollars per kilogram yesterday,” (2) “This flat is for sale for a million dollars,” (3) “Martin decided to produce 30 uniquely designed cars and price them at $3 million per car.”

“Full” Economic Knowledge

Complete economic knowledge is not anything “real,” but it is one of the assumptions in the possible “mathematical” solution to the calculation problem (which was never consequently defended by anyone). It boils down to knowledge of all possible “production functions” available to human beings. Hayek had this type of knowledge in mind when he theorized about allocation problems after postulating many ifs; if we possess all relevant information, all preferences, all knowledge of available means, then the problem of allocation is “purely one of logic” (Hayek 1945, p. 519).

In the neoclassical analysis, production functions are very simple (they have to be) and easily subjected to mathematical formulation. They use only a few variables as factors of production. Their coefficients are given and their influence on production is established and well known. At the same time, since the equations are simple and use few variables, “marginal rates of substitutions” can be inferred from those equations. They can become sorts of shadow prices, which could in theory substitute real world monetary prices and entrepreneurial assessments.7 Those substitution levels can demonstrate, for example, “how much more is being produced when x amount of factor A is substituted for y amount of factor B?” Such contingent tradeoffs could be used for rational allocation.8

In reality such full economic “knowledge” cannot be achieved for two main reasons. Firstly, as Austrian economists have emphasized, production functions9 are complex and each one of them is extremely specific. Production functions consist of many factors of production, which cannot be constricted and grouped into such macroeconomic (or microeconomic) variables as “K” (capital goods) and “L” (labor), or additionally “H” (human capital) and “A” (technology, or “total factor productivity”). Real world production functions have many more variables and their coefficients are not stable numbers. Due to complexity of those functions, simultaneous equations of production functions cannot in fact be “solved” even in “theory.” Walrasian equations can surely be solved, because they are simple and have as many equations as unknowns with known coefficients (Walras 1954, p. 238).10 They appear to be mathematical tasks. By assuming such a trivial world of flat production functions, one is assuming away essential problems of complex economic reality.

The second reason for the lack of such “full” knowledge of the real world is uncertainty and human creativity. However precise the production functions are, they are never accurate, because people are never in a position to fully determine the future. They cannot “close” production functions and make them “complete,” because they would have to include all possibilities about the future.11 Assumptions about the knowledge of those functions implicitly embrace the notion that future is largely foreseen, and that man can anticipate what he or she will learn in the future. Human beings are not omniscient and the future is purely uncertain (in the Knightian sense). It cannot even be subjected to calculus of class probabilities, because in the course of economic events case probability prevails. By assuming away the uncertainty of the future, the fundamental problems of entrepreneurship are also assumed away. Change implies necessity for economic decision making (Mises 1966, p. 212).12 With full knowledge of the future, human beings do not face the problem of proper judgments, since all of them are optimal and efficient. Henceforth “full” economic knowledge (which would allow “shadow prices” instead of monetary prices) is impossible to be achieved, because production functions are too complex and because people can never have a complete list of “correct” functions (which would include information about future events).

The last few sentences seem too trivial and obvious to be mentioned, but there is an interesting consequence of them for the Hayekian concept of knowledge. The complete full economic knowledge is not split up and partitioned between the individuals, therefore it does not become “Hayekian knowledge” when decentralized. If we somehow summed up all the Hayekian knowledge we would still not achieve “full knowledge.” In referring to the hypothetical concept of full economic knowledge Mises writes “no single man can ever master all the possibilities of production, innumerable as they are,” and so the entrepreneurs are divided between their tasks in the environment of monetary calculation (Mises 1990, p. 17). Hayek has a footnote to that Mises’s passage when he refers to the “division of knowledge” (Hayek 1937, p. 50). Yet this is not what Mises had in mind, since clearly full economic knowledge, “all the possibilities of production, innumerable as they are,” cannot be either known or divided between individuals just as infinity cannot be divided into finite numbers. Mises’s point was that “full knowledge” can never be achieved, not that it is in some way divided between the people (compare with Horwitz 1998, p. 430).

As we see, full economic knowledge is unachievable because of the “complexity” and “indeterminacy” of what we sometimes call “production functions.” Indeterminacy problems were to be avoided only if man could turn into a sort of “Laplace’s demon” — entity capable of gaining knowledge about “everything,” meta-knowledge, which would allow the possessor of it to project reality in any way he or she wanted. Fortunately we deal in this article with humans, not gods; henceforth we can set such issues aside for philosophers and theologians. The theoretical economic system can never be “complete” in such sense.

Knowing, Guessing and the Market Process

Perfect Laplacian knowledge leads to perfect forecast. All-knowing man possessing features of the Laplacian “demon” could notice and understand the position of any molecule (even a social “molecule”) in the (social) universe. Such recognition would allow for the planning of every future step ahead and effectively adjust actions to any desirable and possible state of affairs. No mistakes would be committed and the equilibrated Utopian dream could be realized. Any step away from such perfect knowledge results in uncertainty. In order to cope with uncertainty people try to forecast future events.

Beyond the point of perfect knowledge the strict connection between knowledge and forecast breaks. At the extreme, perfect knowledge allows for perfect forecast.13 Once we move away from perfect knowledge we also move away from perfect foresight. Moreover, under the circumstances of uncertainty more knowledge does not always mean better forecasts. It may be truer for cases of natural sciences. The more we know about physics, or chemistry, the better we can forecast “behavior” of the matter. It is slightly different with knowledge of social sciences, where knowledge to some extent improves our understanding of the social world (not necessarily forecasting abilities). More Hayekian, or more current Misesian knowledge, does not necessarily lead to a better economic forecast.

Portions of social knowledge do not guarantee that foreseeing will be in a better shape. Entrepreneurs might be equipped with Hayekian knowledge, but this does not guarantee their success. They can gain a lot of Hayekian knowledge in the market, but still these gains will not automatically transform themselves into entrepreneurial successes. Even the elements of Misesian knowledge do not assure that. Entrepreneurs can acquaint themselves with past prices (realized exchanges) and price offers (currently existing ratios). Knowledge of those is not a formula for commercial accomplishments. When the entrepreneur starts to gather all the price data and gets to know current and previous price offers, it is still not enough to bring him good foresight. Moreover, it is almost nothing. The entrepreneur can gather all that knowledge, and still lose money.

Additionally, gains in knowledge per se do not reap entrepreneurial gains. The effective entrepreneur is not someone who knows “more” than others. There are many entrepreneurs who accomplish a lot even though they were less knowledgeable than their rivals. Especially in the light of the fact that many huge entrepreneurial successes work like in the romantic Schumpeterian story of the entrepreneurs, who break the existing social structures. Sources of triumphs for any entrepreneur do not lie in the typical knowledge build-up, but often in envisioning what is unseen and most likely cannot be seen. All those actions are subjected to revisions and to praxeological recapitalizations in the form of losses and profits, as well as changing asset ownership. Good choices are indicated by correct monetary imputation, and do not have to be correlated with gains in information, or any type of “knowledge” acquisition (Salerno 1990a, pp. 59–60; 1990, pp. 42–43).

Naturally, it does not follow that “knowledge” has nothing to do with forecasts and entrepreneurship. Nevertheless, the entrepreneurs are not spreading Hayekian “knowledge” in their calculations. First of all, in the case of the unfortunate word “transmission,” they are transmitting some things, but these are not Hayekian knowledge and not in the form of prices. Entrepreneurs are transmitting their judgments, and they do it mostly in the form of price offers conveying this information. Whether correct or incorrect, price offers given by sellers of goods and services inform us about how market conditions are currently perceived. The yet to be successful entrepreneur is the one who is capable of “spotting” false prices, a discrepancy between current price offers for factors of production and prices for consumer goods which will be created in the future. “Spotting” is a metaphor, since technically we can only “spot” what already exists. “False prices” do not exist yet. They shall only materialize once the future becomes present. Hence the reason why Kirznerian “profit opportunities” are blurred by clouds of uncertainty and they do not exist yet. Current price offers inform us how entrepreneurs envision today future market conditions. Precisely that kind of “information” is hidden behind prices, not information about proper ways of adjusting “production functions.”

In the neoclassical framework entrepreneurial choice is given by the intersection of the marginal revenue curve and marginal cost curve. The main oversimplification in such an apparatus comes from the coincidence of the two and presupposed incidental existence. In reality one can get to know marginal cost curves by searching for price offers (more or less). Nevertheless the marginal revenue curve does not exist; it cannot be spotted and properly acted upon. We cannot be alert to the marginal revenue curve because it is not there yet. Instead of one marginal revenue curve there is virtually unlimited number of potential non-realized marginal revenue curves. Each of them has case probability assigned to it, thus strictly speaking it has no numerical probability at all. Whoever is more successful in picking the “proper” curve, wins. The “proper” solution is offered with the future being realized. In order to foresee the demand, one does not need to “know more” than others. One needs to make a proper judgment (Hülsmann 1997, p. 35). The “selection” mechanism cannot be reduced to gains in any mentioned type of knowledge.

In other words, the market process is not driven by entrepreneurs who know more, but by entrepreneurs who deliberately select arbitrary types of information and act upon them. A real world forecast is based on those selections of information. Information is interpreted, understood and used.14 What types of information are available to various entrepreneurs? As we saw in the process of economic calculation there is lots of it: realized transactions, which inform us about habits; and recapitalizations, which inform us about the extent of past mistakes. On top of that there are current price offers, which inform us about competitive potential in the market e.g., in which field we can be outcompeted by others and in which fields can we rely on the division of labor. Finally, there are undertaken actions and reallocations by other owners. All this Misesian type of knowledge is generated by the market, based on praxis, and can be referred to as the social appraisement process.

Not only is the world and its information heterogeneous, but so too are individuals. Each entrepreneur is different and has his unique entrepreneurial vision, which can be expressed through the use of property. Entrepreneurs differ in their judgments and disagree on what is economical, and what is not (Lavoie 1985, p. 123). Whoever performs well enough in this task outcompetes his rivals in the market process.

Let us take the case of an entrepreneur producing machines with the use of steel. He can notice past prices for finished products (machines) and past prices of steel. They can inform him about past exchanges and demonstrate past market conditions. He can evaluate them and engage in Verstehen. Any information he gets by contemplation can be useful for current price considerations. Equally useful are “present prices,” price offers for steel. (The entrepreneur also tries to anticipate future prices of the machines). Steel prices inform the entrepreneur how steel is being valued by sellers and by his competitors, other entrepreneurs who alternatively employ steel (to produce something else or similar). Henceforth current prices (offers and transactions from the immediate past) at least inform the entrepreneur of how valuable alternative employments for various factors are, or how other market participants envision the markets of goods produced with steel (compare with Yeager 1994, pp. 95–96). This notification of how much factors are expected to be worth, is a relevant part of the market process and entrepreneurial division of labor.

Accurate anticipation of future prices based on individual understanding of selected information leads to profits. In everyday life we notice how new information changes the prices and actions of market participants. The person acquiring new knowledge cannot be sure that its spread should change prices in a particular way. In some cases we can be almost close to certainty what the effect should be. But it can never be “fully” known in advance. If new fields of oil are discovered, the anticipation is that the price of oil should go down. Nevertheless it need not to, and we can envision scenarios in which the opposite happens. Successful entrepreneur is the one who can “interpret the information” correctly, but only in the ex post sense. He acts very often against the tide and the rest of the market.

The crucial side of the competitive process is its legal aspect. The mechanism of entrepreneurial selection is based on property shifts, which result from monetary calculation. This works despite psychological motivations of the participants, or their “knowledge,” or their “ignorance.” It does not matter what entrepreneurs’ incentives are, or what kind of information they possess. They can know a lot, or little, they can be motivated in their actions by their personal skills, or act upon an ideological bias. Whatever they know, and whatever their incentives are, as economists we do know that those who satisfy consumers most survive in the market. We do not even have to assume that entrepreneurs are interested in “maximizing” profits (Alchian 1950, pp. 212–13).15 Their personal interests and motivations are not important. Profits are the link between consumer satisfaction and entrepreneurial decisions acknowledging them. That is why the market process “works” — because calculation has consequences for allocations.

In the economic analysis of socialism we can assume many things. If we assume that planners have “full knowledge,” then we “solve” the problem with an unrealistic assumption. In the real world planners can only gain other types of knowledge. They can possess all the necessary technological knowledge, and even the more specific Hayekian knowledge of time and place. We can even add that planners could possess scatters of Misesian knowledge: they could accurately know past prices and price offers right before the imposition of the socialist order. Yet even this knowledge does not solve the main socialist deficiency: the central owner does not know what are, or would be, the allocations of private owners. He cannot substitute them, or even hire them as bureaucrats, because tangible entrepreneurial skills are manifested in the realms of praxeological boundaries conditioned by asset ownership. When the central owner nationalizes the resources, all entrepreneurial skills are outlawed and simply lost.16 They cannot be recovered by any bureaucratic structure, because there is no real world competition set in the property regime.


As we have seen, in economics “knowledge” can have many different meanings. In assessing economic systems one has to be careful in making particular assumptions about “knowledge,” because any discussion may turn out to be blurred by definitional barriers. Depending on what we exactly mean by the term “knowledge” various conclusions about its possession or non-possession can be reached. It all comes down to what exactly we understand by this term.


Alchian, Armen A. 1950. “Uncertainty, Evolution, and Economic Theory.” Journal of Political Economy 58(3).

Barone, Enrico. 1935 [1908]. “The Ministry of Production in the Collectivist State.” In F.A. Hayek, ed., Collectivist Economy Planning. London: Routledge and Kegan Paul.

Cassirer, Ernst. 1944. An Essay on Man: An Introduction to a Philosophy of Human Culture. New Haven, Conn.: Yale University Press.

Garrison, Roger. W. 1992. “The Limits of Macroeconomics.” Cato Journal 12(1) (Spring/Summer).

Hayek, Friedrich August. 1937. “Economics and Knowledge.” In Individualism and Economic Order. Chicago: University of Chicago Press.

——. 1948 [1940]. “Socialist Calculation III: The Competitive Solution.” In Individualism and Economic Order. Chicago: University of Chicago Press.

——. 1945. “The Use of Knowledge in Society.” American Economic Review 35(4).

——. 1984. “Two Pages of Fiction.” In C. Nishiyama, K. R. Leube, ed. The Essence of Hayek. Stanford, Calif.: Hoover Institution Press.

Horwitz, Steven. 1998. “Monetary Calculation and Mises’s Critique of Planning.” History of Political Economy 30(3).

Huerta de Soto, Jesús. 2010. Socialism, Economic Calculation, and Entrepreneurship. Cheltenham, UK: Edward Elgar.

Hülsmann, Jörg Guido. 1997. “Knowledge, Judgment and the Use of Property.” Review of Austrian Economics 10(1).

Kirzner, Israel M. 1996. “Reflections on the Misesian Legacy in Economics.” Review of Austrian Economics 9(2).

Lavoie, Don. 1985. Rivalry and central planning. The socialist calculation debate reconsidered. Cambridge: Cambridge University Press.

Mises, Ludwig von. 1966. Human Action: A Treatise on Economics. Chicago: Contemporary Books.

——. 1990. Economic Calculation in the Socialist Commonwealth. Auburn, Ala.: Mises Institute.

Polanyi, Michael. 1966. The Tacit Dimension. Garden City, N.Y.: Doubleday.

Salerno, Joseph. 1990. “Ludwig von Mises as a Social Rationalist.” Review of Austrian Economics 4.

——. 1990a. “Postscript: Why a Socialism Economy is ‘Impossible’.” In Ludwig von Mises, Economic Calculation in the Socialist Commonwealth. 1990. Auburn, Ala.: Mises Institute.

——. 1994. “Reply to Leland B. Yeager on “Mises and Hayek on Calculation and Knowledge.” Review of Austrian Economics 7(2).

Stigler, George, and Gary S. Becker. 1977. “De Gustibus Non Est Disputandum.” The American Economic Review 67(2).

Walras, Léon. 1954. Elements of Pure Economics, or The Theory of Social Wealth. London: American Economic Association and the Royal Economic Society by Allen and Unwin.

Yeager, Leland B. 1994. “Mises and Hayek on Calculation and Knowledge.” Review of Austrian Economics 7(2).

  • *. Mateusz Machaj is assistant professor at the Institute of Economic Sciences at the University of Wroclaw, Wroclaw, Poland. I would like to thank to Professor Joseph Salerno for many years of his invaluable help. This article is an outcome achieved due to indispensable long-term academic guidance of Professor Salerno. My intellectual development would have not been possible without his personal support, and without the study of his masterful works on monetary theory and general economic theory.
  • 1. Mises notes (1966, pp. 206–07) that under perfect substitutability of capital goods would imply that “all means of production ... would be as if only one kind of means — one kind of economic goods of a higher order existed.” Therefore in a socialist economy one could calculate according to the usage of the one universal higher order good (e.g., kilograms of such good), and avoid the problem of valuation of heterogeneous factors of production (non-perfect substitutability of capital goods).
  • 2. Although other types below could also be seen as objectively existing.
  • 3. During the socialist calculation debate the term “technological” knowledge was used (see Mises 1966, p. 699).
  • 4. On the appraisement process see Salerno (1990, p. 42; 1994a, p. 120). It is of course debatable to call activities as “knowledge.” But, as we explain below, we will stretch a little bit and name them “knowledge,” because from a certain perspective this is what the central planners would need to “know” — the actions of private owners — in order to act efficiently.
  • 5. They also include current understandings of past trends in prices. The information on past prices visioned as valuable is being reflected in the current appraisal.
  • 6. At some point Hayek suggested this is not the main problem, because “price expectations and even the knowledge of current prices are only a very small section of the problem of knowledge” (Hayek 1937, p. 51). In the other paper he suggested otherwise. See Hayek (1984, pp. 57–58).
  • 7. Stigler and Becker (1977, p. 77) use the term “shadow price” to label a valuation for a good, which is not sold or purchased in the market. They use it for a different type of a discussion, but the idea to use the concept of “shadow price” is similar as in here. A “shadow price” is something which is to be inferred from subjective valuations and can substitute market pricing. Yeager uses “shadow price” in the analogous sense (Yeager 1994, p. 101).
  • 8. From the equations we can know how much of an additional amount of one factor of production is needed to replace decreased amount of the other factor if one wishes to maintain the level of output. These types of rate can be known only if production function is simple and known.
  • 9. Actually the word “function” is a doubtful name, but it is a topic for another discussion. There is not much typically “functional” about production processes.
  • 10. Walras later on (when he deals with progress) allows for adjustable coefficients, but still the system contains “as many equations as there are unknowns to be determined” (p. 384).
  • 11. This is why a neglected Barone stated that “it is frankly inconceivable that the economic determination of the technical coefficients can be made a priori” (Barone 1908, p. 287). Ironically he later became to be quoted for having “solved” the problem of economic calculation under socialism, even though he did not believe so and actually argued the opposite.
  • 12. As Hayek (1945, p. 94) notes “economic problems arise always and only in consequence of change.”
  • 13. “It may be added that knowledge, in the sense in which the term is here used, is identical with foresight only in the sense in which all knowledge is capacity to predict” (Hayek 1937, p. 51). It might be stated that we need calculation, because we can never possess enough knowledge.
  • 14. As Kirzner points “possessing all this information is not the same as having assimilated it” (Kirzner 1996, p. 150). In this sense “assimilation” process is always subjective (both for the entrepreneur and hypothetical central owner under socialism).
  • 15. Actually “maximization” is also an improper word, since it would imply we have a particular “function” to be maximized. In reality, entrepreneurs choose between various rates of profits and case probabilities associated with them.
  • 16. Mises (1990, p. 38) brilliantly emphasized this in his initial article: “Unfortunately ‘commercial-mindedness’ is not something external, which can be arbitrarily transferred. … The entrepreneur’s commercial attitude and activity arises from his position in the economic process and is lost with its disappearance.”