5. The Non-Neutrality of Money

The* monetary economists of the sixteenth and seventeenth centuries succeeded in dissipating the popular fallacies concerning an alleged stability of money. The old error disappeared, but a new one originated, the illusion of money’s neutrality.

Of course, classical economics did its best to dispose of these mistakes. David Hume, the founder of British Political Economy, and John Stuart Mill, the last in the line of classical economists, both dealt with the problem in a masterful way. And then we should not forget Cairnes, who in his essay on the course of depreciation paved the way for a realistic view of the issue involved.1

Notwithstanding these first steps towards a more correct grasp, modern economists incorporated the fallacy of money neutrality into their system of thought.

The reasoning of modern marginal utility economics begins from the assumption of a state of pure barter. The mechanism of exchanging commodities and of market transactions is considered on the supposition that direct exchange alone prevails. The economists depict a purely hypothetical entity, a market without indirect exchange, without a medium of exchange, without money. There is no doubt that this method is the only possible one, that the elimination of money is necessary and that we cannot do without this concept of a market with direct exchange only. But we have to realize that it is a hypothetical concept which has no counterpart in reality. The actual market is necessarily a market of indirect exchange and money transactions.

From this assumption of a market without money, the fallacious idea of neutral money is derived. The economists were so fond of the tool which this hypothetical concept provided that they overestimated the extent of its applicability. They began to believe that all problems of catallactics could be analyzed by means of this fictitious concept. In accordance with this view, they considered that the main work of economic analysis was the study of direct exchange. After that all that was left was to introduce the monetary terms into the formulas obtained. But this was, in their eyes, a work of only secondary importance, because, as they were convinced, the introduction of monetary terms did not affect the substantial operation of the mechanism they had described. The functioning of the market mechanism as demonstrated by the concept of pure barter was not affected by monetary factors.

Of course, the economists knew that the exchange ratio between money and commodities was subject to change. But they believed—and this is exactly the essence of the fallacy of money’s neutrality—that these changes in purchasing power were brought about simultaneously in the whole market and that they affected all commodities to the same extent. The most striking expression of this point of view is to be found in the current metaphorical use of the term “level” in reference to prices. Changes in the supply or demand of money—other things remaining equal—make all prices and wages simultaneously rise or fall. The purchasing power of the monetary unit changes, but the relations among the prices of individual commodities remain the same.

Of course, economists have developed for more than a hundred years the method of index numbers in order to measure changes in purchasing power in a world where the ratios between the prices of individual commodities are in continuous transition. But in doing so, they did not give up the assumption that the consequences of a change in the supply or demand of money were a proportional and simultaneous modification of prices. The method of index numbers was designed to provide them with a means of distinguishing between the consequences of those changes in prices which take their origins from the side of the demand for or supply of individual commodities and those which start from the side of demand for or supply of money.

The erroneous assumption of money neutrality is at the root of all endeavors to establish the formula of a so-called equation of exchange. In dealing with such an equation the mathematical economist assumes that something—one of the elements of the equation—changes and that corresponding changes in the other values must needs follow. These elements of the equation are not items in the individual’s economy, but items of the whole economic system, and consequently the changes occur not with individuals but with the whole economic system, with the Volkswirtschaft as a whole. Proceeding thus, the economists apply unawares for the treatment of monetary problems a method radically different from the modern catallactic method. They revert to the old manner of reasoning which doomed to failure the work of older economists. In those early days philosophers dealt in their speculations with universal concepts, such as mankind and other generic notions. They asked: What is the value of gold or of iron, that is: value in general, for all times and for all people, and again gold or iron in general, all the gold or iron available or even not yet mined. They could not succeed in this way; they discovered only alleged autinomies which were insoluble for them.

All the successful achievements of modern economic theory have to be ascribed to the fact that we have learned to proceed in a different way. We realize that individuals acting in the market are never presented with the choice between all the gold existing and all the iron existing. They do not have to decide whether gold or iron is more useful for mankind as a whole, but they have to choose between two limited quantities both of which they cannot have together. They decide which of these two alternatives is more favorable for them under the conditions and at the moment when they make their decision. These acts of choice performed by individuals faced with alternatives are the ultimate causes of the exchange ratios established in the market. We have to direct our attention to these acts of choice and are not at all interested in the metaphysical and purely academic, nay, vain question of which commodity in general appears more useful in the eyes of a superhuman intelligence surveying earthly conditions from a transcendental point of view.

Monetary problems are economic problems and have to be dealt with in the same way as all other economic problems. The monetary economist does not have to deal with universal entities like volume of trade meaning total volume of trade or quantity of money meaning all the money current in the whole economic system. Still less can he make use of the nebulous metaphor “velocity of circulation.” He has to realize that the demand for money arises from the preferences of individuals within a market society. Because everybody wishes to have a certain amount of cash, sometimes more, sometimes less, there is a demand for money. Money is never simply in the economic system, in the Volkswirtschaft, money is never simply circulating. All the money available is always in the cash holdings of somebody. Every piece of money may one day—sometimes oftener, sometimes more seldom—pass from one man’s cash holding to another man’s. But at every moment it is owned by somebody and is a part of his cash holdings. The decisions of individuals regarding the magnitude of their cash holdings constitute the ultimate factor in the formation of purchasing power.

Changes in the quantity of money and in the demand for money for cash holding do not occur in the economic system as a whole if they do not occur in the households of individuals. These changes in the households of individuals never occur for all individuals at the same time and to the same degree and they therefore never affect their judgments of value to the same extent and at the same time. It is exactly the merit of Hume and Mill that they tried to construct a hypothetical case where the changes in the supply of money could affect all individuals in such a way that the prices of all commodities would rise or fall at the same time and in the same proportion. The failure of their attempts provided a negative proof, and modern economics has added to this the positive proof that the prices of different commodities are not influenced at the same time and to the same extent. The oversimple formula both of the old quantity theory and of contemporary mathematical economists according to which prices, that is all prices, rise or fall in the proportion of the increase or decrease in the quantity of money, is disproved.

To simplify and to shorten our analysis let us look at the case of inflation only. The additional quantity of money does not find its way at first into the pockets of all individuals; not every individual of those benefited first gets the same amount and not every individual reacts to the same additional quantity in the same way. Those first benefited—in the case of gold, the owners of the mines, in the case of government paper money, the treasury—now have greater cash holdings and they are now in a position to offer more money on the market for goods and services they wish to buy. The additional amount of money offered by them on the market makes prices and wages go up. But not all the prices and wages rise, and those which do rise do not rise to the same degree. If the additional money is spent for military purposes, the prices of some commodities only and the wages of only some kinds of labor rise, others remain unchanged or may even temporarily fall. They may fall because there are now on the market some groups of men whose incomes have not risen but who nevertheless are obliged to pay more for some commodities, namely for those asked by the men first benefited by the inflation. Thus, price changes which are the result of the inflation start with some commodities and services only, and are diffused more or less slowly from one group to the others. It takes time till the additional quantity of money has exhausted all its price changing possibilities. But even in the end the different commodities are not affected to the same extent. The process of progressive depreciation has changed the income and the wealth of the different social groups. As long as this depreciation is still going on, as long as the additional quantity of money has not yet exhausted all its possibilities of influencing prices, as long as there are still prices left unchanged at all or not yet changed to the extent that they will be, there are in the community some groups favored and some at a disadvantage. Those selling the commodities or services whose prices rise first are in a position to sell at the new higher prices and to buy what they want to buy at the old still unchanged prices. On the other hand, those who sell commodities or services whose prices remain for some time unchanged are selling at the old prices whereas they already have to buy at the new higher prices. The former are making a specific gain, they are profiteers, the latter are losing, they are the losers, out of whose pockets the extra-gains of the profiteers must come. As long as the inflation is in progress, there is a perpetual shift in income and wealth from some social group, to other social groups. When all price consequences of the inflation are consummated, a transfer of wealth between social groups has taken place. The result is that there is in the economic system a new dispersion of wealth and income and in this new social order the wants of individuals are satisfied to different relative degrees, than formerly. Prices in this new order cannot simply be a multiple of the previous prices.

The social consequences of a change in the purchasing power of money are twofold: first, as money is the standard of deferred payments, the relations between creditors and debtors is changed. Second, as the changes in purchasing power do not affect all prices and wages at the same moment and to the same extent, there is a shift of wealth and income between different social groups. It was one of the errors of all proposals to stabilize purchasing power that they did not take into account this second consequence. We may say that economic theory in general did not pay enough attention to this matter. As far as it did, it principally considered it only in reference to the reaction of a change in a country’s currency on its foreign trade. But this is only a special application of a problem which has a much wider scope.

What is fundamental for economic theory is that there is no constant relation between changes in the quantity of money and in prices. Changes in the supply of money affect individual prices and wages in different ways. The metaphorical use of the term price level is misleading.

The erroneous opinion to the contrary was based on a consideration which may be represented thus: let us think of two absolutely independent systems of static equilibrium A and B. Both are in every respect alike except that to the total quantity of money (M) in A and to every individual cash holding (m) in A there correspond in B a total quantity of Mn and individual cash holdings mn. On these assumptions of course all the prices and wages in B are n times those in A. But they are exactly thus because these are our hypothetical assumptions. But nobody can devise a way by which the system A can be transformed into the system B. Of course it is unpermissible to operate with static equilibrium if we wish to approach a dynamic problem.

Setting aside all qualms about the use of the terms dynamic and static, I wish to say: money is necessarily a dynamic agent and it was a mistake to deal with monetary problems in a static way.

Of course there is no room left for money in a concept of static equilibrium. In forming the concept of a static society we assume that no changes are taking place. Everything is going on in the same old manner. Today is like yesterday and tomorrow will be like today. But under these conditions nobody needs a cash holding. Cash holding is necessary only when the individual does not know what situation he will have to face in an uncertain future. If everybody knows when and what he will have to buy, he does not need a private cash holding and can entrust all his money to the central bank as time deposits due on the dates and in the amounts necessary for his future payments. As everybody would proceed in the same way, the central bank does not need any reserves to meet its obligations. Of course, the total amount which it has to pay out to the buyers every day exactly balances the amount which it receives as deposits from the sellers. If we assume that in this world of static equilibrium once, before the equilibrium was attained, there was metallic currency only, let us say gold, we have to assume that with the gradual approach towards conditions of equilibrium the citizens deposited more and more of their gold and that the bank, which had no need for it, sold the gold to jewelers and others for industrial consumption. With the advent of equilibrium there is no more metallic money, there is in fact no more money at all, but an unsubstantial and immaterial clearing system, which cannot be considered as money in the ordinary sense. It is rather an unrealizable and even unthinkable system of accounting, a numeraire as some economists believed ideal money ought to be. This, if it could be called money, would be neutral money. But we should never forget, that the state of equilibrium is purely hypothetical, that this concept is nothing but a tool for our mental work. Not being able to make experiments, the social sciences have to forge such tools. But we must be very careful in their use. We have to be aware that the state of static equilibrium can never be attained in real life. Still more important is the fact, that in this hypothetical state the individual does not make choices, does not act and does not have to decide between incompatible alternatives. Life in this hypothetical state is therefore robbed of its essential element. In constructing this hypothetical state we want merely to understand the incentives of action, which always implies change, by conceiving conditions, in which no action takes place. But a changeless world would be a dead world. We do not just have to deal with death, but with life, action, and change. In a living world there is no room for neutrality of money.

Money, of course, is a dynamic factor and as such cannot be discussed in terms of static equilibrium.

Let me now briefly point out some of the major conclusions derived from an insight into the non-neutrality of money.

First we have to realize that the abandonment of the fallacious concept of neutral money destroys the last stronghold of the advocates of quantitative economics. For a very long time eminent economists have believed that it will be possible one day to replace qualitative economics by quantitative economics. What renders these hopes vain, is the fact, that in economic quantities we never have any constant ratios among magnitudes. What the economist discovers when he studies relations between demand and prices is not comparable with the work of the natural scientist who determines by experiments in his laboratory constant relations, e.g., the specific gravity of different substances. What the economist determines is of historical value only; he is in his statistical work a historian, but not an experimenter. The work of the late lamented Henry Schultz2 was economic history; what we learn from his research is what happened with some commodities in a limited period of the past in the United States and Canada. It tells us nothing about what happened with the same commodities elsewhere or in another period or what will happen in the future.

But there still has remained the belief that it is different with money. I may cite, for example, Professor Fisher’s book on the Purchasing Power of Money, which is founded on the assumption that the purchasing power of the monetary unit changes in inverse proportion to the quantity of money.3 I think that this assumption is arbitrary and fallacious.

The second conclusion which we have to draw is the futility of all endeavors to make money stable in purchasing power. It is beyond the scope of my short address to explain the advantages of a sound money policy and the disadvantages of both inflation and deflation. But we should not confuse the political concept of sound money with the theoretical concept of stable money. I do not wish to discuss the inner contradictions of this stability concept. From the point of view of the present subject it is more important to emphasize that all proposals for stabilization, apart from other deficiencies, are based on the idea of money’s neutrality. They all suggest methods to undo changes in purchasing power already effected if there has been an inflation they wish to deflate to the same extent and vice versa. They do not realize that by this procedure they do not undo the social consequences of the first change, but simply add to it the social consequences of a new change. If a man has been hurt by being run over by an automobile, it is no remedy to let the car go back over him in the opposition direction.

The popularity of all schemes for stabilization invites us to a philosophical consideration. It is a general weakness of the human mind to regard the state of rest and absence of change as more perfect than the state of motion. The absolute, that old phantom of misguided philosophical speculation, is still with us; its modern name is stability. But stability, e.g., absence of change, is, we have to repeat, absence of life.

The third conclusion which we may draw is the futility of the distinction between statics and dynamics and between short-run and long-run economics. The way in which we have to study monetary changes provides us with the best evidence that every correct economic consideration has to be dynamic and that static concepts are only instrumental. And at the same time we have to realize that all correct economic theorizing is a gradual progress from short-run to long-run effects.

But the most important value of the theory of money’s dynamism is its use for the development of the monetary theory of the trade cycle. The old British Currency-Theory was already in a restricted sense a monetary explanation of the cycle. It studied the consequences of credit expansion on the assumption only that there is credit expansion in one country whereas in the rest of the world things are left unchanged. This seemed to be enough for the explanation of the business cycle in Great Britain in the first half of the nineteenth century. But the explanation of an external drain does not provide an answer to the question what may happen in a completely isolated country or in the case of a simultaneous credit expansion all over the world. But only the answer to this second question could be considered satisfactory under the conditions prevailing in the twentieth century. Only the answer to this second question is important, if we have to consider the proposals for eliminating the cyclical changes either by loosening the international ties of the national economy or by making credit expansion international in the way the Bretton Woods Agreements4 provide. It is the boast of the monetary theory of the trade cycle that it provides us with a satisfactory answer to these and to some other serious problems.

I do not wish to infringe more upon your time and so I wish only to add some remarks on the treatment of the problem by certain younger economists. I myself am not responsible for the term “neutral money.” I have developed a theory of the changes in purchasing power and its social consequences. I have demonstrated that money acts as a dynamic agent and that the assumption that the changes in purchasing power are inversely proportional to the changes in the relation of demand for to the supply of money is fallacious. The term “neutral money” was coined by later authors.5 I do not wish to consider the question of whether it was a happy choice. But in any case I must protest against the belief that it has to be a goal of monetary policy to make money neutral and that it is the duty of the economists to determine a method of doing so. I wish to emphasize that in a living and changing world, in a world of action, there is no room left for a neutral money. Money is non-neutral or it does not exist.

• *[This essay was delivered as a lecture to a group in Paris in 1938 and again to the New York City Economics Club in 1945 and previously unpublished—Ed.]
• 1[David Hume, “On Money,” in Writings on Economics, Eugene Rotwein, ed. (University of Wisconsin Press, Madison, 1970), pp. 33–46; John Stuart Mill, Principles of Political Economy, Sir William Ashley, ed. (1909), bk. 3, chap. 8; John E. Cairnes, Essays in Political Economy (London: MacMillan, 1873), pp. 1–65—Ed.]
• 2[In his treatise Theory and Measurement of Demand (Chicago: University of Chicago Press, 1938) he set forth his crop theory of cycles—Ed.]
• 3[Irving Fisher, The Purchasing Power of Money, 2nd ed. (New York: Macmillan, 1920), p. 157. “there is no possible escape from the conclusion that a change in the quantity of money (M) must normally cause a proportional change in the price level”—Ed.]
• 4[The Bretton Woods agreement in 1945 established an international gold exchange standard that valued the dollar at l/35th of an ounce of gold—Ed.]
• 5[F. A. Hayek, Prices and Production, 2nd ed. (New York: Augustus M. Kelley, 1935), pp. 31 and 129–31—Ed.]