Academic article of the week: The use of knowledge in society


Today we publish a translation of the famous article by F.A. Hayek, published in the American Economic Review in 1945 and quoted in several scholarly works since then.

What is the problem we are trying to solve when trying to build a rational economic order? Based on some broadly accepted assumptions, the answer is quite simple.

If we detained all the relevant information, if we could take as a starting point an established system of preferences, and if we had full knowledge of the available means, the rest of the problem would simply be a matter of logic. That is, the answer to the question as to what is the best use of available means is implicit in our assumptions.

The conditions that must be satisfied to solve this ideal problem have been thoroughly analyzed and can be better explained in a mathematical model: we would say that the marginal substitution rates between any two goods or factors must be the same regardless of their different uses.

This, however, is decidedly not the economic problem facing society; and the economic calculation we have developed to solve this logical problem, although it is an important step towards solving the economic problem of society, does not yet provide an answer for it. The reason for this is that the total "data" of society from which economic calculations are made are never "given" to a single mind so that it could analyze its implications - and never will.

The peculiar character of the problem of a rational economic order is characterized precisely by the fact that the knowledge of the circumstances under which we have to act never exists in a concentrated and integrated form, but only as scattered pieces of incomplete and often contradictory knowledge, spread over several individuals. The economic problem of society, therefore, is not merely a problem of how to allocate "certain" resources - if by "determinate" we understand something that is available to a single mind that can deliberately solve the problem on the basis of this information.

Instead, the problem is how to ensure that any member of society will make the best use of the known resources, for purposes whose relative importance only these individuals know. Or, putting it succinctly, the problem is the use of a knowledge that is not available to anyone in its entirety.

The fundamental character of this problem has unfortunately been obscured, not illuminated, by many of the recent refinements in economic theory, and in particular by the varied uses of mathematics. Although the problem which I would like to deal primarily with in this article is the problem of the organization of a rational economy, to follow this path I will need repeatedly to draw attention to the intimate connections that this problem has with certain methodological questions.

Many of the arguments that I intend to present are, in fact, conclusions reached through different paths of reasoning that unexpectedly converged. But, as I nowadays understand these questions, this convergence is not a coincidence. It seems to me that many of the divergences that arise both in the field of economic theory and in economic policy have a common origin in a misunderstanding of the nature of the economic problem of society. This misunderstanding, in turn, is due to an improper application of mental habits developed to deal with problems of nature with social phenomena. In common language, we define the word "plan" as the set of interrelated decisions concerning the allocation of available resources. All economic activity in this sense is planning; and in any society in which several people collaborate, planning, regardless of who does it, will have to be based on certain knowledge; and this knowledge will not be available in the first instance to the planner, but rather to someone who should relay them to the planner. The various ways in which knowledge reaches the people who use it to draw up their plans is a crucial problem for any theory that seeks to explain the market process; and the problem of how best to use the knowledge that is initially dispersed among several independent people is at least one of the main problems for economic policy - or for any attempt to devise an efficient economic system. The answer to this question is intimately related to another issue that emerges here: that of who is planning. The whole disagreement over "economic planning" is part of that. It is not a question of whether or not to plan, but whether planning should be done centrally, by a single authority for the entire economic system, or whether it should be divided among several individuals. In the specific sense in which the term is used in contemporary controversies, planning necessarily means central planning - directing the entire economic system of accord with a unified project. Competition, on the other hand, means a decentralization of planning that will be carried out by many independent people. The middle way between these two positions - much talked about, but little appreciated when seen in practice - is the delegation of planning to certain organized industries, that is, the institution of monopolies. The question of which of these systems will be most efficient depends mainly on the question of which one can expect a more complete use of existing knowledge. And this, in turn, depends on whether we have a greater probability of being able to put all the knowledge that is dispersed among various individuals at the disposal of a central authority, or of giving individuals sufficient additional knowledge so that they become capable of to integrate their plans with those of others. It will immediately become evident that at this point the answer will differ according to the different types of knowledge; and the answer to our question will consequently turn to the relative importance of different types of knowledge; those most likely to be available to particular individuals, and those that we would be more certain to find in the possession of a body made up of well-chosen specialists. If it is so widely accepted today that the latter is preferable, this is because a type of knowledge - scientific knowledge - occupies so prominent a place in the public imagination today that we have forgotten that it is not the only type of relevant knowledge. It may be admitted that, in relation to scientific knowledge, an organ with a handful of well-chosen experts is the best choice to better master the available knowledge - although this is obviously merely a matter of exchanging one problem for another: the problem of how to choose these specialists. What I want to emphasize is that even assuming that this problem could be solved immediately, it would only be part of a larger problem. Today it is almost a heresy to suggest that scientific knowledge does not correspond to the totality of knowledge. But a little reflection will show that there is undoubtedly a very important body of disorganized knowledge that can not be called scientific, understanding "scientific" as the knowledge of certain general rules: the knowledge of certain particular circumstances of time and place. It is in this respect that virtually every individual has a comparative advantage over all others, for he has unique information about what kinds of beneficial uses can be made with certain resources; which will only happen if the decision on how to use them is left in the hands of this individual or is taken with his active cooperation. Just remembering how much we need to learn in any profession after we have completed our theoretical training, how great is the part of our professional life in which we spend learning specific skills, and how valuable, in all circumstances of life, is the knowledge of people, local conditions and certain special circumstances. Knowing and operating a machine that was not being adequately exploited, or exploiting the ability of someone who could be better exploited, or being aware of a surplus of reserves that can be used during a temporary interruption of supply is as socially useful as knowledge of the best alternative techniques. The conveyor who earns his living discovering how best to seize his empty cargo space, the real estate agent whose knowledge consists almost exclusively of finding temporary opportunities, or the individual who does arbitrage, who profits from the local differences between the prices of certain goods - they all carry out eminently useful works which are based on a special knowledge of the circumstances of a fleeting moment, unknown to others. It is curious that today this type of knowledge is widely overlooked, and that people who make use of it to attain privileges over people with better theoretical or technical training are seen almost as if they were doing something dishonorable. But while conquering privileges by using superior knowledge of the conditions of communication and transportation is seen as almost dishonest, the truth is that it is almost as important for society to make the best use of these opportunities as it is to make the most of the latest scientific discoveries. prejudice has a considerable influence on the fact that people tend to adopt a more unfavorable attitude towards commerce than in relation to productive activities. Even economists who believe they are totally immune to the shallow materialistic fallacies of the past constantly make the same mistakes in practical knowledge - and the reason for this seems to be that, according to the way they see the world, this kind of knowledge should already be "given" instead of being something that needs to be sought. The most common idea today seems to be that all such knowledge should be constantly available to everyone, and since that is not the case, the current economic order is criticized for being supposedly irrational. This conception ignores the fact that the method of making this knowledge widely available is precisely the problem that we need to solve.If today it is fashionable to minimize the importance of knowing the particular circumstances of time and space, this is largely due to little importance given to the question of uncertainty itself. Indeed, some of the assumptions (which are generally only implicit) adopted by the "planners" differ from their opponents both in terms of the capacity for unforeseen changes to cause substantial changes in the production plans and in relation to the frequency with which this occurs. Of course, if it were possible to make detailed economic plans for significantly lengthy periods beforehand, and then to follow them closely, so that no other major economic decisions were needed, the task of drawing up a complete plan for all economic activity would not unattainable. Perhaps it is worth stressing that economic problems arise always and exclusively as a result of changes. As long as things continue exactly as they were before - or at least when they proceed according to what was expected of them - then no new problems will arise that require solutions, so there is no need for a new planning. The belief that change - or at least small daily adjustments - has become less important in modern times assumes that containment of economic problems has also become less important. For this reason, people who tend to overlook the importance of uncertainty are the same ones who argue that economic issues are no longer as important as technological knowledge. It is true that, thanks to the sophisticated apparatus of modern industry, one only has to make economic decisions at long intervals; such as when deciding whether a new plant should be built, or should a new procedure be introduced? Is it true that, once a factory has been built, the rest is more or less mechanical, determined by its characteristics, leaving little to be changed to adapt to the eternal fluctuations of every moment? I know it, it does not support this widely accepted belief. At least in areas that are competitive - and only those areas serve as a model for that - the task of stopping the costs of going up requires a constant struggle that absorbs much of the manager's energy. It is easy for an inefficient administrator to spend the small leftovers from where the profits go; it is a commonplace of business experience that, under the same technical conditions, the same output can be made within a huge range of costs - but this is not equally known to those who study only economics. The very desire - often stated by producers and engineers - to be allowed to do their projects without financial considerations is an eloquent testimony to the power these factors exert over their daily work. One reason for the growing inability of economists to heed the constants small changes that make up the whole of economic activity is probably the fact that they are increasingly concerned with statistical data, which pass a much more stable picture of the economy than the small daily movements. However, the relative stability of the large statistical data can not be explained - as statisticians often want to believe - by the "big numbers laws" or by mutual compensation for small random changes. The number of elements they deal with is not large enough for these accidental forces to produce stability. The continuous flow of goods and services is maintained by deliberate and constant adjustments, by new decisions taken daily in the light of circumstances that were unknown until the previous day, by B's decision to enter the scene when A ceases to perform its role. Even the largest and most mechanical of the factories goes on in large part because of an environment that can provide you with all of your unexpected demands: new roof tiles, papers for your documents, and all the thousand and one types of equipment you do not can be produced by the plant itself but, in order for it to continue to function, need to be readily available on the market. At this point, I should briefly note that the kind of knowledge I have been dealing with is a a type which, by its very nature, can not be transposed to statistical data and therefore can not be made available to a central authority which decides on the basis of statistical surveys. The statistics that this authority would have to use would come precisely through the abstractions of the small differences between things, joining as elements of a single type items with different characteristics of place, quality, and other particular characteristics that would be very important to take specific decision. Consequently, central planning based on statistical information, by its very nature, can not directly take into account the circumstances of time and place, needing to find some way for these decisions to be left to someone who is in the place. If we can agree that the economic problem society is basically a matter of adapting quickly to changes in the particular circumstances of time and place, it seems clear that, as a consequence, fundamental decisions should be left to those who are familiar with those circumstances, and the resources immediately available to deal with them. We can not expect this problem to be solved by transmitting all this knowledge to a central directory that, having integrated all that knowledge, issues an order. We need decentralization because only then can we ensure that knowledge of the particular circumstances of time and place is readily used. But the man who is within a particular situation can not make decisions based solely on his knowledge of the facts concerning his immediate surroundings, for although this is an intimate knowledge, it is also limited. However, the problem remains of how to convey to this man enough information so that he is able to fit his decisions into the general pattern of changes in the economic system as a whole. How much knowledge does he need to be successful in this? Which of the events that will happen beyond your immediate horizon of knowledge are relevant to your immediate decision, and how well does he need to know these events? There is practically nothing that happens in the world that can not influence the decision that he needs to make. But he does not need to know these events in themselves, nor does he need to know all their effects. For him, it is not important to know why a certain type of screw is being most sought after at a specific time, or why paper bags are more easily available than canvas bags, or why specialized workers or specific machines momentarily have become difficult to find. All he needs to know is how much more or less difficult is the acquisition of certain things in relation to other things that also interest him or if the demand for other things that he produces or uses is more or less urgent. He is always concerned with the relative importance of particular things, while the factors that alter this relative importance are of no interest to him except insofar as his own effects on the concrete things of his environment are concerned. I called "economic calculation" helps us, at least by analogy, to understand how this problem can be solved - indeed, as it is already being solved - by the price system. Even if there were a single controlling mind that possessed all the data on a small and narrow economic system, it would not bother to go through all the relationships between ends and means that might be affected every time some small adjustment in resource allocation was done. Indeed, one of the great contributions of the pure logic of choice is to have conclusively demonstrated that even a single omniscient mind could only solve this kind of problem by constructing and constantly using equivalence rates (or "marginal values" substitution "), that is, by assigning a numerical index to each type of resource which, without being derived from any property of that particular thing, would still reflect or condense its relevance in the total structure of means and ends. For each small change, she would have to consider only those quantitative indices (or "values") in which relevant information would be concentrated; and by adjusting the quantities one by one, it could rearrange all the elements without having to retake the whole puzzle from the beginning nor need to stop at each step to re-analyze all the elements and their ramifications.Basically, in a system in which the knowledge of the relevant facts is dispersed among several people, prices can serve to coordinate the different actions of several people in the same way as the values would help that omniscient mind to coordinate the different parts of its plan. It pays to contemplate for a moment a very simple and common example of the pricing system in action to see exactly what it can do. Suppose for a moment that somewhere in the world a new opportunity to use some raw material came up - let us take the tin as an example - or that some of the tin sources have been eliminated. For our example it does not matter - and it is very significant that it does not matter - which of those two causes have increased tin shortage. All that tin users need to know is that some of the tin they used to consume now is being used more profitably elsewhere, and as a result they need to be more economical in their use. know where this demand most urgently came from, not even for what they will save these resources. It is enough that some of them directly know of the existence of the new demand and transfer resources to it, that some other people perceive the emptiness that was then created and act to fill it with resources from other sources, and then the effect will quickly spread throughout the economic system, influencing not only all the uses of tin, but also the uses of its substitutes, and the substitutes of these substitutes, as well as the supply of all things made of tin, and that of its substitutes for these things, against; and all this occurs without the vast majority of those who carry out these substitutions knowing anything about the original cause of these changes. The whole acts as if it were a single market, but this does not happen because each of its members could analyze it as a whole, but rather because the limited fields of vision of each one had sufficient scope so that, through numerous intermediaries, the relevant information was communicated to all. The mere fact that there is a price for each good - or, rather, that each local price is linked in some way with the cost of transporting it to that location, and so on - brings the same solution as a single mind endowed with all the information (although it is only an imaginary possibility) would have reached, even if this information is actually dispersed among all the people involved in the process. We need to understand the price system as a mechanism for transmitting information in order to understand its a function which it obviously fulfills with less perfection as prices become more rigid. (But even when tabulated prices become extremely rigid, the forces that would normally act to cause price changes remain active, exerting considerable influence on changes in other aspects of contracts.) The main aspect of this system is the knowledge economy with which it operates; or, in other words, how little individual participants need to know to be able to make the right decisions. In short, by means of a certain type of symbol, only the most essential information is transmitted forth, and only to those who are interested in it. It would not be just a metaphor if we were to say that the pricing system is like a cash register, or a telecommunications system that allows individual producers to observe only the movement of some factors - just as an engineer can focus only on the consoles of some dials - to adapt their activities to the changes they know only from what is shown by the price movement. Evidently, these adjustments are probably never "perfect" in the sense of perfection that economists use in their analysis of economic equilibrium. However, I fear that our theoretical habit of addressing each problem with the presumption of more or less perfect knowledge on the part of almost everyone involved has almost blinded us to the true function of the price mechanism, leading us to misleadingly apply inadequate standards to judge its efficiency. It is marvelous that in a situation where there is a shortage of a raw material, without any order being given, and perhaps no more than a handful of people know the cause of this scarcity, tens of thousands of people whose identities will never be known , even after months of research, then begin to use this matter or its by-products in a more economical way; that is, they all act in the right direction. This, in itself, is marvelous enough; even if, in a world of constant uncertainties, not everything can organize itself so perfectly that its percentages of profits are constantly kept at the same level as "normal." I deliberately use the word "wonder" to shock the reader and withdraw it from complacency with which we usually take for granted the mechanism. I am convinced that if this were the result of a conscious human project, and that people guided by price changes knew that their decisions are of far greater importance than the realization of their immediate ends, then that mechanism would be praised as one of the greater triumphs of the human mind. Their misfortune is twofold: nor is it the fruit of a human project, nor do the people guided by it tend to understand why they do the things they are driven to do. But those who call for a "conscious direction" - and who can not believe that something that has been created without planning (and, indeed, without even someone understanding it as a whole) can solve problems that we ourselves can not to solve consciously - they must remember the following: the problem is precisely how to expand the extent of resource utilization beyond the extension of the understanding of a single individual; and therefore it is a problem of how to manage the need for conscious control, and how to give incentives for individuals to make the desirable decisions without someone telling them what to do. The problem we are dealing with here is not at all is concerned exclusively with the economy, for it arises along with almost all other real social phenomena, with language and much of our cultural heritage, being in fact the central problem of all social science. As Alfred Whitehead said of something else, "A deeply false truism, repeated by all the manuals and in the speeches of eminent people, says that we should cultivate the habit of thinking about what we are doing. Civilization progresses when we increase the number of important jobs we can do without thinking about them. " This has a profound importance in the social field. We constantly use formulas, symbols, and rules whose meaning we do not understand, but through which we may have access to knowledge that we individually do not possess. We create these practices and institutions on the basis of habits and institutions that have proved successful in their own spheres and which have become the foundation upon which we build civilization. The price system is just one of those creations that man has learned to use (although he is still far from having learned to use it perfectly), after which he came upon him, even before he understood it. Through it not only the division of labor but also the coordinated use of resources based on widely disseminated knowledge become possible. People who like to ridicule any suggestion that things work that way distort our argument by insinuating that we are saying that it is by some miracle that such a system developed spontaneously, becoming the most suitable for modern civilization. It is just the opposite: man can create this division of labor upon which our civilization stands precisely because it has encountered a method that has made it possible. If that had not happened, he might have developed an entirely different kind of civilization, perhaps the "state" of termites, or something totally unimaginable. All we can say is that no one has succeeded in producing an alternative system in which certain features of the existing system - which are respected even by those who violently attack it - can be preserved, especially in relation to the individual's ability to choose his goals and, For several reasons, it is great that the need for the price system for any rational calculation in a complex society is no longer the subject of discussion only between groups with different political opinions. The thesis that without the price system we could not preserve a society based on a division of labor as extensive as ours was met with shouts of laughter when Mises introduced it twenty-five years ago. Today the arguments that some still present to reject this thesis are no longer exclusively political, and this creates an atmosphere much more receptive to thoughtful discussions. When we see Leon Trostky arguing that "economic calculation is unimaginable without market relations"; when Professor Oscar Lange promises Professor von Mises a marble statue in the future Directory of Central Planning, and when Professor Abba P. Lerner rediscovers Adam Smith, emphasizing that the essential utility of the price system is to induce the individual to do that which is of general interest at the moment when it seeks to realize its own interests, then the divergences can no longer be attributed to political prejudices. The remaining dissidents clearly seem to diverge from this position for purely intellectual motives, and more particularly A recent statement by Professor Joseph Schumpeter in his "Capitalism, Socialism and Democracy" provides a perfect example of these methodological differences that I have in mind. The author is one of the most prominent economists among those who analyze the economic phenomenon from some branch of positivism. For him, these phenomena arise as a consequence of the mutual effect exerted by certain objective quantities of goods, almost as if there were no intervention of human minds. Just because of these assumptions, I can understand the following statement - and, to me, astounding. Professor Schumpeter argues that the possibility of rational calculation in the absence of a market for the factors of production is a consequence of the theoretical proposition that "consumers who are evaluating (demanding) consumer goods ipso facto are also evaluating the means of production that enter into the production of those goods. "Taken literally, this statement is simply false. Consumers do nothing of the sort. What Schumpeter's "ipso facto" probably means is that the assessment of the factors of production is implicit, or necessarily follows, from the valuation of consumer goods. But this is also not true. Implication is a logical relationship that can only be asserted safely from assumptions that are for the same individual. It is evident, however, that the values ​​of the factors of production do not depend exclusively on the evaluation of consumer goods, but also on the conditions of supply of the various factors of production. Only a single individual who knew all these factors simultaneously could find a response derived directly from these data. The practical problem arises, however, precisely because such data are never entirely available to a single individual, and because, consequently, it is necessary to solve this problem the use of knowledge that is dispersed by several individuals. would be by no means resolved if we demonstrated that all data, if available for a single mind (as hypothetically would be for the economist observing the problem), would themselves determine the solution; instead, we would need to demonstrate how a solution could be produced by interaction between people who individually have only partial knowledge. To assume that all knowledge can be put at the disposal of a single mind, in the way we assume it may be available to us, as economists dedicated to analyzing an issue, is tantamount to running away from the problem and belittling everything that is important and relevant in the world That an economist of Professor Schumpeter's stature fell into such a trap because of the ambiguity the term "given" has for the unwary could hardly be considered a simple mistake. This suggests, in fact, that there is something fundamentally wrong with an approach that often neglects an essential part of the phenomena we have to deal with: the inevitable imperfection of human knowledge and the necessity arising from a process by which knowledge is constantly acquired and transmitted. Any approach - such as a large part of mathematical economics with its various simultaneous equations - that assumes that the knowledge of people corresponds to the objective facts of each situation, will systematically leave out what is our main task to explain. I am far from denying that in our system the analysis of economic equilibrium has a useful activity to play, but when it reaches the point where it overshadows our leading intellectuals, making them believe that the situation they are describing is of direct relevance to the solution of practical problems, it is more than time to remember that this type of analysis does not deal with the social process at all, and that this is no more than a preliminary step in the investigation of the main problem.