The Stock Market, Credit and Capital Formation

CHAPTER XII

CREDIT CREATION AND THE ATTEMPT TO DETERMINE ITS PROPER LIMITS

75. The view expressed here that credit expansion is liable to end with a crisis is not one that is shared by all economists. It may be emphasized once again that it is not, of course, the credit itself which is “dangerous.” The bank credit created by the note issuing banks and the commercial banks may in the aggregate comprise a very considerable proportion of the total circulation (i.e., the circulation of money including checking deposits) and it would be courting ridicule to claim that this “fiduciary circulation” is dangerous. It is not dangerous any longer. It exerted its effects earlier at the time of its creation, and by now has long been a part of the circulation of means of payment which is entirely harmless and is even necessary in order that the existing price structure may be maintained.

The reason why there are so many misunderstandings and differences of opinion in this sphere of banking theory is that it is a sphere where problems relating to the supply of money and the price level converge with problems relating to the demand for capital and the level of interest rates. The double rôle of bank money as a medium of payment and as money capital has paved the way for a great deal of confusion which many authors seem unable to escape. The muddle is avoided if it is made perfectly clear that bank money functions directly as money capital at the time of its creation by new bank loans and investments. After it has been used as money capital by the entrepreneur who obtained it in the first instance, it becomes part of the general stream of money, and flows in and out of the cash holdings of the various members of the exchange economy, by becoming part of their money income. It can act as money capital a second and third time only when it becomes part of the voluntary savings of an income recipient who forgoes present consumption.1

The double role of money: as circulating medium it affects prices, as money capital it affects interest rates.

It acts as money capital when it is created by loans and investments,—

—and later again only when it is saved out of income.

The creation of bank money (i.e., the granting of new loans or the purchasing of securities by a bank) exerts its effect, like all additions to the supply of money capital, on the rate of interest and on the structure of production. The continued existence of bank money that was created previously (and, of course, has had its short-run effects) is neutral towards the credit market and the structure of production. The creation and continued existence of this money has, however, a lasting effect on the supply of circulating media and the price level. The causal sequence (or rather the sequence of probable tendencies) may be roughly described as follows: The expansion of bank credit will be accompanied by a lowering of interest rates and will lead to a rise in prices and money incomes.2 When the credit expansion ceases and if the volume of bank money can be maintained at the higher level, prices and money incomes may remain at an elevated level, whereas interest rates will rise again. While the price level is dependent on the absolute volume of circulating media, the rate of interest hinges on changes in the volume of circulating media.3

Credit creation has a lasting effect on the price level,—

—but only a temporary one on interest rates.

A cardinal mistake of many writers on monetary theory is that they believe that if the amount of newly created circulating media is only just sufficient to maintain approximate constancy of the price level (which would otherwise have fallen), it will have no effects on the capital market and the structure of production. These writers are the victims of a simple sophism which runs somewhat as follows: “A rate of interest at which no more is invested than is provided for out of voluntary savings leaves the price level unchanged. Therefore, a rate of interest which keeps the price level constant is equivalent to this equilibrium rate of interest.” Unfortunately “leaving the price level unchanged” is not the same thing as “keeping it constant.” The creation of new circulating media so as to keep constant a price level which would otherwise have fallen in response to technical progress, may have the same unstabilizing effect on the supply of money capital that has been described before, and thus be liable to lead to a crisis. In spite of their stabilizing effect on the price level, the emergence of the new circulating media in the form of money capital may cause roundabout processes of production to be undertaken which cannot in the long run be maintained.4

Credit creation may stabilize the price level and yet unstabilize production.

76. Many believers in the ideal of the stable price level, who propose that a fall of prices due to technical progress and falling costs of production should be prevented by means of credit expansion, are fully conscious of the accompanying danger of over-investment. It is, however, open to them to argue that the additional circulating media should be used to finance consumption, and in that case, would not produce the changes in the production structure which eventually lead to a crisis. It would be possible to keep the newly created purchasing power out of investment channels and to pour it exclusively into the hands of consumers either by financing instalment credit, or by subsidizing wage increases, or by financing relief and bonus payments, or by financing other state expenditure. What conclusions does the rough analysis of the previous chapter allow us to draw regarding this programme? The inflationary augmentation of consumer purchasing power would lower the relative share of total purchasing power devoted to investment (reinvestment) purposes. This would mean a relative diminution in the supply of “capital disposal.” If the production of producers’ goods competes with the production of consumers’ goods by using the same productive factors, then a rise in costs may occur, leading to a contraction of production in certain producers’ goods industries and, in turn, to diminished sales in other industries which previously supplied them with materials. Under conditions of full employment of the labour supply, this would probably occur immediately. But even if there is unemployment, the supply of special kinds of labour or of other productive factors may be scarce. The rise in consumer purchasing power and the relative diminution in investment purchasing power will then lead, via a rise in costs, to dislocations in the capital goods industries.

Over-investment by money creation might be prevented, if the new money financed consumption only.

Increased consumers’ demand, however,—

—may raise production costs, reduce real investment,—

—and cause contraction in capital goods industries.

So long as there are factors of production which are scarce, i.e., which rise in price when the demand increases,5 the respective effects of the alternative types of inflation are likely to be these: Credit granted only to producers will lead first to an expansion of the producers’ goods industries (prosperity) and later to a crisis and contraction (depression); credit granted only to consumers may lead directly to a painful contraction of the producers’ goods industries.

Thus both producers’ credit and consumers’ credit can produce disproportionalities.

The question whether it would be possible to distribute the new purchasing power created by bank lending so “ideally” between investors and consumers as to avoid all disturbances at the time when the credit expansion is discontinued, is more than doubtful.

An ideal sharing-out of new credit seems fantastic.

It is hardly necessary to point out that stock exchange credits are usually investment credits. It is only a very much smaller portion of stock exchange loans which is used to finance the purchase of securities from sellers who want to spend the whole or part of their sales proceeds on consumption, that has to be regarded as consumption credit.

Stock exchange credit finances chiefly investment and, only to a smaller extent, consumption.

77. There are many people who admit that there are dangers connected with inflation, but who argue that a “tiny bit” of inflation cannot do any harm. This attitude, which is dictated by an inflationary bias which exists among large sections of the business world, has its counterpart in the scientific treatment of the problem of bank credit. Everywhere the question is asked: What are the limits to which it is possible to go in expanding bank credit without producing harmful effects; what are the limits to a “healthy” credit expansion?

What are the limits to a healthy credit expansion?

It has already been remarked that the answer to this question is commonly influenced by a confusion of the effects on the price level and the effects on the interest rate. There are many proponents of price stabilization who lack all understanding of the connexions between an increase in the quantity of circulating media, the rate of interest, and production. But even first-rate professional economists sometimes fall into this error. Röpke, for instance, said once in his discussion of capital formation: “If no rise in prices occurs this means that the volume of credit is being kept within limits that are necessary for financing the transactions of the economic system and is thus fulfilling a function which is evidently outside the functions of capital.”6 Röpke is here voicing the peculiar view that up to a certain limit, i.e., so long as it does not produce any absolute price rise, a credit expansion does not represent any increase in the supply of money capital. It is not clear why an expansion of bank credit should be sometimes within and sometimes “outside the functions of capital.” Incidentally, Röpke does not always adhere to the argument quoted, especially as in other places he has himself spoken rather disrespectfully of the creed of the price-level stabilizers.7

It has been said that created credit, as long as it does not raise prices, does not function as money capital,—

One author who is an untiring advocate of the universal remedy of the “stable price level” is Gustav Cassel. This is not unimportant since this author has had a great deal of influence on economic writings all over the world. According to Cassel: “Such an increase (in bank lending) is permissible to the extent that the general progress of industry means a greater demand for money,”8 the measure of the demand for money being the constancy of the general price level. In so saying, Cassel is fully conscious of the fact that such creation of an additional supply of circulating media involves a fall in the rate of interest charged by the banks9 and that a fall in the rate of interest charged by the banks gives an artificial stimulus to the production of capital goods.10 One might suppose that Cassel would recognize these two elements as causes of cycles and crises, but all this is forgotten when he comes to the discussion of the stable price level. Indeed, Cassel defines the equilibrium rate of interest1 not as that rate of interest at which the amount of capital investment would be equivalent to the supply of money capital from “natural sources,” but as that rate of interest at which the quantity of circulating media will move in such a way as to give an approximately stable price level.2

—on that an interest rate which stabilizes the price level is the “equilibrium rate”—

The arguments which Cassel adduces in support of his view that “a falsification of the capital market’s situation through too low a rate of interest”3 merely produces a “transition from one position of equilibrium to another” are not sufficient proof that this mechanism functions without producing the phenomena associated with a crisis. Cassel shows that a single reduction of the rate of interest by the banks cannot lead to a lasting inflation since certain factors soon begin operating to counter-balance the stimulus of the lower rate of interest.4 This may be perfectly correct on its own merits, and had already been pointed out by Mises, but it has nothing to do with the question whether the increase in the production of capital can be maintained. Cassel deals with this question in the following sentence, however: “The artificial reduction of the interest rate has, then, led to an artificially reinforced capital production, which is tantamount to a forced increase in the national savings.”5 This sentence, which Cassel does not make the slightest attempt to substantiate, is obviously equivalent to accepting the doctrines of “forced saving” and of the “creative power of bank credit,” which is not quite in harmony with Cassel’s other views.

—and that it leads to forced capital formation without causing dislocations

So far, no satisfactory proof has been given in support of the argument that credit creation, to the extent necessary to prevent a fall in the general price level, will not cause disturbances in the production structure. We are, therefore, constrained to fall back on the results of Hayek’s analysis: “The rate of interest at which, in an expanding economy, the amount of new money entering circulation is just sufficient to keep the price-level stable, is always lower than the rate which would keep the amount of available loan capital equal to the amount simultaneously saved by the public; and thus, despite the stability of the price level, it makes possible a development leading away from the equilibrium position.”6

It seems more probable that price, level stabilization in an expanding economy leads to disequilibrium.

78. We have seen that it is a mistake to attempt to define the limits of a “harmless” expansion of bank credit in terms of price stabilization. One of the few authors who have attempted to define the limits of permissible credit expansion in other terms is Adolf Lampe. “Its limits,” says Lampe, “are determined by: (a) the size of the reserves of the social product, (b) the tempo at which the output of the social product follows the input, (c) whether it is economically possible to put back into the economic process what has been taken out (or its equivalent), so that the social product may be reproduced in time.”7 Commendable as this attempt to determine the limits may be in comparison with the contributions of other students of this problem, it suffers in my opinion from the use of concepts which, if they are not incomprehensible, are at least extremely vague. If the “reserves of social product” are supposed to represent unused productive factors, especially labour which is involuntarily unemployed, then Lampe’s determinant (a) would come to much the same thing as Keynes’ new indicator of credit expansion. It has to be noted, however, that this determinant is not sufficient according to Lampe, and that he refers to two other factors which have to be present if the economic process is to go on smoothly. We shall have occasion to refer again to Keynes’ indicator later in this chapter.

Other criteria for the proper limits of credit expansion have been advanced.

There are certain considerations which can be stated in fairly simple terms, and which seem to me to enable us to find the limits within which it is possible to expand bank credit without incurring the penalties. Newly created credit places money capital at the disposal of the market without any corresponding release of productive factors due to voluntary refraining from consumption. Whereas the normal process of capital formation consists of two steps, saving and investing, newly created bank credit makes it possible for investment to take place in the absence of voluntary saving, and this is what gives rise to the development of disproportionalities in the production process. We are also familiar with the opposite case of onesidedness in the process: intended saving without investment, that is hoarding. In this case, the private saving, as we explained in § 14, fails to produce any saving from the social point of view. In other words, it does not lead to any capital formation. The command over consumers’ goods or over the corresponding factors of production, instead of being made over to an investor in the form of “capital disposal,” is returned to the consumers (and other buyers) in the form of a deflationary fall in prices, or it is even lost if wage rigidities do not permit that the labour force be bought for a reduced amount of money.

Investment without intended saving is inflation;—

—intended saving without investment is deflation;—

This exposition points straight to the answer to our question. No disturbance in the productive structure of the economic system will be caused provided the investment without saving—which is financed by credit expansion—does not exceed the saving without investment—which is sterilized by hoarding. This leads to the conclusion that the limits of a healthy inflation of credit are determined by the extent of the simultaneous deflation due to hoarding.

—hence, inflation of credit is healthy if it compensates for deflation due to hoarding.

High-flown inflationary aspirations do not receive much support from our conclusions. Our rule says that additional purchasing power may be created and lent to investors to the extent that there are funds that have been saved but not invested. In other words, unused purchasing power is substituted for by newly created purchasing power.8 (The interest on the loans or investments accrues to the banks instead of to the savers.) Although this rule may seem to be quite clear-cut, in practice it is very vague owing to the difficulty of estimating the amount and the duration of the hoarding. If we take an increase in idle balances as justification for expanding credit, then we must regard a decrease in idle balances as a reason for contracting credit. If the hoarding and dishoarding approximately balance each other, we arrive at a figure for appropriate credit expansion of exactly nil. Only on condition that the figure for net hoarding is positive can we justify an expansion of credit on these grounds.

It is difficult to estimate the amount of net hoarding.

79. The case of compensating a deflation due to hoarding does not exhaust all possible cases where a credit expansion could take place without leading to a crisis. We may arrive at another case by recalling our theoretical “scheme of the constant supply of money capital” (see above §73). We saw there that it was conceivable that in a period following on a period of credit expansion a rise in voluntary saving might occur which would provide a dose of money capital equivalent to what had initially been provided out of bank credit. In this case the money capital supplied in the first period would be of a purely inflationary character, whereas that supplied in the next period would consist of money capital representing the voluntary giving up of consumption. It might be said that the supply of money capital in this case preceded the intended release of productive factors from the consumption goods industries by one period. But this anticipation of saving would not lead to a crisis so long as the saving and investment activity were in fact maintained on the higher level.9

Another case of healthy inflation is where initial credit creation is “relieved” by corresponding doses of voluntary saving.

The conjuncture of events just described is, however, not likely to occur except by mere chance. The case has nothing to do with the idea of “forced saving” nor with the equality of “ex post savings” with the investment financed out of credit expansion.1 What is meant here is voluntary saving which takes place in the periods succeeding a period of credit expansion, and which serves to continue financing the increased volume of investment called forth previously by the credit expansion. It is not impossible for this to occur, especially as the ability to save may rise in consequence of the previous credit expansion. The increased ability to save cannot, however, last once the rise in prices has caught up with the increased level of money income. Only in the case of unemployed factors being absorbed into the production process in such a way that total real income is increased permanently, can the ability to save be raised permanently. And only then may it be possible for a rise in voluntary saving to keep the supply of money capital at the higher level after the credit injections have ceased.

Increased voluntary savings might be forthcoming because of the “inflated” ability to save,—

—but this would be lasting only if real income were permanently increased.

The saving which is made by entrepreneurs out of the inflated profits accruing from the credit expansion (and which might perhaps be called “secondary saving”) does not suffice to prevent a subsequent reaction. It is more likely to reinforce the tendency to over-investment and to make the subsequent reaction more severe, since the profit inflation will probably reach its end at the same time as the credit inflation. Thus two sources of money capital—credit expansion and corporate saving out of inflated profits—will dry up simultaneously. It seems, therefore, that the setback in investment activity will be inevitably intensified.

The “secondary saving” out of inflated profits would only intensify over-investment and the subsequent reaction.

There is thus little likelihood that investment activity can be maintained at the higher level after the credit expansion has come to an end. The increase in investment which was financed out of bank credit will then turn out to have been nothing more than the upward phase of a trade cycle which is followed by a crisis and depression. We conclude that credit creation may be, so to speak, “money in advance” against the savings of the future; but since the future development of voluntary savings can never be predicted beforehand, there is not much justification for expanding bank credit on the basis of these future savings.

80. Another significant consideration relevant to the determination of the proper limits of credit creation centres around the problems of changes in the volume of money transactions and in cash balances of business firms. The cash balances of the various firms constitute part of their circulating capital. The cash holdings of all firms of a given industry with a given volume of business will be larger or smaller according as there is a lesser or greater degree of vertical integration of production, i.e., unification of different stages of any branch of production in one firm.2 This applies both to their minimum cash reserves and to their average cash balances. The former are the cash reserves which are normally held in a preconceived amount in order to be prepared for unforeseen expenses. The larger the number of firms, the larger will be the aggregate size of these basic reserves, and the smaller the number of firms, the smaller will be the sum of these idle cash holdings. The average cash holdings include active balances which are not determined according to any fixed plan, but are a reflection of the fact that current receipts accumulate in the cash holdings of the firms for some interval (however short) before they are spent. The more firms there are (through which the products have to pass in the course of being processed), the more transactions there are to be settled with money, and, therefore, the higher are the cash balances held by the industry concerned for a given volume of business. The fewer the firms, the fewer the transactions that need to be settled with money, and, therefore, the smaller the relative cash balances held in the industry concerned.

Changes in the degree of vertical integration of industry affect the amounts of cash balances, both idle and active, held for a given volume of business.

Changes in the necessary size of these cash holdings do not only affect the price system. They also represent changes in the demand for working capital and consequently influence the demand for, and supply of, money capital. It might be possible to compensate changes of this kind by increasing or decreasing the volume of credit: in the case of increasing vertical integration of different establishments, it would be necessary to contract credit; and in the case of increasing vertical differentiation of industry, i.e., in the case of the splitting up of firms into separate units controlling separate phases of the productive process, it would be necessary to expand credit. It does not seem to me, however, that such movements could, in actual practice, be taken account of by banking policy.

These changes affect both prices and interest.

Credit creation to offset the effects of reduced integration of industry would be another case of healthly inflation.

It is frequently argued that an increase in population justifies an increase in the volume of money. This argument is, however, not correct in the form in which it is stated. It obviously makes the implicit assumption that the number of separate holders of cash bears a fixed ratio to the size of the population. Only if this were so, could an increase in population be held to justify an increase in the volume of money. An increase in population per se (or in miniature an addition of a baby to the family) does, of course, evoke the desire for “more money” (in reality more real income) with which to feed the additional mouth, but this does not constitute an increased “demand for money” or a reason for increasing the volume of money in the system.

Population growth involves an increased demand for money only if the number of household units is increased;—

On the other hand, the number of separate households, and, consequently, the number of people who want to hold minimum cash reserves, may increase without a growth in population. Let us suppose that after a long period during which the population has been constant, the population figure rises over a number of years in consequence of an increase in the birth-rate, and later becomes stable again. The increase in the number of babies would be no reason for an increase in the volume of money. Later, however, when the population has already stopped increasing, the age distribution of the population will change: There will be a larger number of young people reaching the age when they begin earning their own living and become independent. Quite apart from the probable increase in the supply of labour, the fact that the number of owners of pockets, purses, and bank accounts, increases explains that there will be an increased demand for cash; this will have a deflationary influence and should be compensated by an increase in the volume of circulation.

Thus, it is not the increase in population per se, but the increase in the number of people wanting to hold cash, which gives the expansion of credit its compensatory character.3

—to offset this increase is a fourth case of healthy inflation.

81. Our discussion of the “stable price level” idea and of the proposal of price-stabilizing expansions of credit has already shown that an increase in the volume of production of goods does not prevent an accompanying credit expansion from leading eventually into crisis and depression. One of the views that is most widely held among the public is that a growth in the production of goods requires additional money to finance the increased movement of goods, and that, if the additional money is not provided, disturbances, leading to a crisis and depression, are certain to result. But it has been sufficiently established that a fall in prices which is due to increased productivity need not give rise to economic depression.4

Increased production does not warrant credit creation.

More production without more money need not cause depressions.

The most extreme form of the argument that there should be an expansion in the volume of money every time there is an increase in the output of goods is to be found in the so-called “classic plan of money creation” of Bendixen.5 He proposed that every good produced should be accompanied by an increase in the amount of money (through the discounting of commercial bills) corresponding to the value of the newly produced good. This doctrine found an enthusiastic response6 in certain circles in Germany; to-day it no longer has any following.

The idea that all commercial bills could be discounted without harm because they would merely bring the amount of money into equilibrium with the “needs of trade” was the chief mistake of the banking school (Tooke,7 Fullarton,8 &c.). They failed to see —and this is a blindness which continues to afflict many contemporary writers of textbooks—that the demand for loans from the banks (which was confused with the demand for money) is dependent also on the conditions on which such loans are obtainable.9 It has long been realized that in normal times the banks can cause, fairly quickly, a substantial rise in the total volume of bills discounted by lowering their interest rate and following a more liberal policy in respect to selection and rationing. It was because of this, i.e., because the volume of bills brought forward for discount is dependent not only on the “physical turnover of goods,” but also on banking policy, that the whole question with which we are concerned here arose: the question as to what are the limits to which a credit expansion can go without giving rise to the danger of a subsequent depression.

The theory that purely commercial bank credit matches the needs of trade—

—is of no use.

The doctrine of pure commercial credit has been elaborated in modern times by reference to the kind of goods against which the loans are made. Thus, while it is agreed that credit expansion which goes to finance the production of producers’ goods and durable consumers’ goods may lead to a crisis, it is held that an expansion of credit is harmless so long as it is “properly used.” It is supposed to be “properly used” when the credits are applied to the financing of increases in the production of consumers’ goods which do not require new fixed investments.1

It has been argued that any self-liquidating credit for consumers’ goods industries is “healthy.”

This theory has its roots in the liquidity rules formerly preached by many banking theorists. (These rules related, however, to the lending of genuine short-term savings and not to new credit created by the banks.) According to Lederer “the granting of created credit as working capital for the production of consumers’ goods”2 is harmless and is in no way inflationary, because in this case “the credit would serve to cause income streams to be produced simultaneously with the goods, and these income streams serve to buy the goods. In so far as an immediate supply of saleable goods is forthcoming as the counterpart of the credits, and in so far as the credits immediately give rise to income streams, the consumption of the goods provides the means for paying back the credits.”3

The view that the expansion of credit for financing the production of consumers’ goods will not lead to disproportionalities of the kind associated with inflation can be disproved by the following argument. Either the consumers’ goods industries would have borrowed on the money market, or the capital market, in the absence of any expansion of bank credit, in which case the satisfaction of their demand for funds by means of the credit expansion obviously implies that there is so much less pressure on the credit market, and that some producers’ goods industry, which would not otherwise have obtained credit to finance an expansion, will be enabled to do so by this means. As a consequence the eventual results of expansion (boom and depression), which Lederer also admits in general, will appear. Or the consumers’ goods industries would not have had any incentive to extend production in the absence of the credit expansion; in this case the fact that they now enter the market for producers’ goods with relatively increased buying power as against all other industries (which are supposed not to obtain credit directly or indirectly) may lead to a change in the distribution of productive factors involving a shift from the stages far from consumption to the stages near to consumption. In this case disturbances in the producers’ goods industries, as described above, may occur without being preceded by the boom phase of a trade cycle. In a later section (§ 98) we shall show that the second alternative is not to be taken very seriously and that the first possibility is by far the more probable.

This view neglects probable repercussions on capital goods industries,—

—via the capital market,—

—or via shifts in demand.

Thus, it seems that the likelihood that a credit expansion will be crash-proof is not increased by the fact that loans are made to selected industries on the basis of certain rules about liquidity.

82. In recent years4 there has been a rapid growth in the literature on the subject of what kind of credit policy is least likely to produce crises. The discussion of the criteria and “guides” of credit policy shifted to an entirely different plane once it was realized that “stable money” and “neutral money” imply different monetary policies. To-day, all the better grade textbooks contain quite a lengthy catalogue of possible guides to credit policy: stable cost of living, stable wholesale price level, stable factor prices, stable “general prices” (including services, rents, and securities), constant volume of money (including checking deposits), constant “effective” circulation, constant total money income, stable level of employment, are the main items in the list. To discuss and compare these various guides would take us far away from the main topics of this book. It is, however, not irrelevant to make a few observations on one index which is receiving increasing emphasis as a criterion for credit expansion, namely, the existence of unemployment.

Many guides for crash-proof monetary policies have been developed.

It is practically indisputable that unemployment can be diminished by credit expansion provided simultaneous increases in money wages are prevented, or provided at least that such increases lag behind the tempo of the credit expansion. What is open to doubt, apart from the question of whether such a wage policy is likely to be pursued, is only whether the expansion of credit which is undertaken in face of unemployment contains the seed of a reaction or not. (To answer this question in the affirmative does not imply that one disapproves of credit expansion unconditionally.)

Credit expansion reduces unemployment if money wage rates do not rise or if they lag behind.

But is such expansion not liable to reactions?

Credit expansion for the purpose of financing private investment will have slim chances so long as the prospective rates of return continue to be negative. If it is considered too long to wait until the anticipated rates of return become positive, all that is practicable is an expansion of credit for financing public works.5 But whether it is private or public investment that is concerned, the credit expansion which is undertaken in order to finance it, will necessarily produce changes in relative prices and changes in the structure of production. The question we have to ask is: Do these changes lead eventually to an “untenable situation,” and consequently to a reaction, in spite of the fact that unemployed labour was available for investment?

In its original formulation the Mises-Hayek theory started out from a state of full employment and on this basis it was possible to argue that an investment inflation will draw productive factors away from the stages of production near to the consumers’ goods end, and that this situation is not tenable in the long run and is bound to lead to a reaction. It was easy to challenge this thesis of the “distorted structure of production” by arguing that it becomes inapplicable if there is a supply of unemployed factors.

The dislocation arguments were based on the assumption of full employment.

This argument finally led up to Mr. Keynes’ proposal that the complete disappearance of involuntary unemployment should be regarded as the proper limit of credit expansion. Up till the time when “full employment” has been reached, Keynes sees no particular dangers in the financing of increased investment by means of credit creation. “When full employment is reached, any attempt to increase investment still further will set up a tendency in money-prices to rise without limit, . . .; i.e., we shall have reached a state of true inflation. Up to this point, however, rising prices will be associated with an increasing aggregate real income.”6

Disappearance of involuntary unemployment has been suggested as marking the limit of “healthy” credit expansion.

The mere fact that Keynes confines the term “true inflation” to that increase in the circulation which is not accompanied by any increase in production is merely a change of name. A definition does nothing to alter the substance of the matter. The question whether an expansion of investment financed through a credit expansion is likely to produce an unstable situation long before full employment has been reached is not decided merely by refusing to call such an expansion “true inflation.” Keynes, however, really believes that it is possible to perpetuate the boom,7 so long as private investment is continuously stimulated by means of cheap money and is supplemented and, if necessary, even replaced by public investment.

Easy money for private investment, and, if necessary, its replacement by public investment, are postulated in order to secure perpetual prosperity.

Before considering whether such a policy is practicable in the long run, let us see what its implications would be. Given the willingness to offset every deficiency of private investment activity by public investment, the sense of the concept “crisis” is of course changed. After all, it is the contraction of private investment activity from a relatively high level to a much lower level, and the causes and consequences of this contraction, which form the main part of the subject-matter of cycle theory. The main point which required to be explained was why the upswing, however initiated, should necessarily lead to a situation in which it became impossible or unprofitable to keep private investment going on the previous scale. A scarcity of money capital, a contraction of demand, a rise in costs, an increase in the risk estimates, were only a few of the factors that were adduced in explanation of this point. Naturally, one may say that there would not be any decline in aggregate investment if any gap that arose were always filled by public investment without regard to the profitability of that investment, that is to say, if the demand, cost, and risk elements in investment could be neglected and if the necessary money capital were provided by the commercial banks and the central banks in unlimited quantities. But this only shelves the problem: it does not solve it.

The real problems of cycle theory: demand, cost, risk, capital supply,—

—are all whisked away by the public investment “solution.”

To aim at correcting a situation through a temporary dose of public works is one thing. It is another thing to aim at guaranteeing full employment all the time by undertaking public works at whatever level may be necessary to maintain investment at a given level. In the first case, the authorities concerned hope by their intervention to correct the situation in such a way as to create more opportunities for investments that are profitable on the basis of cost-price relationships. In the second case, the significance of cost-price relationships for the functioning of the existing economic system, and for the determination of capitalistic production plans, is cast to the winds. To judge the direction and extent of investment and production according to whether it will pay, that is, whether the undertaking is justified by the relations between costs and prices is, however, something more than a mere liberal-conservative prejudice.

It has been shown (§ 78) that an inflation by public investment can be justified as a compensatory measure for a deflation due to private hoarding. If it can be demonstrated that deflation due to hoarding is on the increase, it will appear appropriate to speed up the public investment inflation. (The public investment will in this case be designed to prevent the decline in money income which Keynes fears will arise from too small a propensity to consume.) Inflation of public investment which exceeds the deflation due to private hoarding and which therefore causes money incomes and costs to rise is quite another matter. It is very probable that the private investment activity, which is first stimulated by the artificial increase in effective demand, will collapse as soon as the public investment inflation is checked. It is even likely that private investment activity will decline under the influence of unavoidable increases in costs, if the public investment inflation is kept going for a long period at a constant rate. And this makes it very probable that the attempt to reach and maintain full employment by means of a public investment inflation would involve increasing rates of inflation which would almost certainly lead to an eventual collapse.

Inflation by public investment to compensate for deflation by private hoarding may be sound.

Public investment inflation in excess of private hoarding, raising incomes and costs, is likely to lead to a reaction before full employment is reached.

All leading economists (with almost no exceptions) are of the opinion that, in general and under given conditions, an increase in employment is only possible if there is a (temporary) fall in real wage rates. (A wage policy based on the immediate adjustment of wage rates to the rise in prices would make every net expansion of credit a “true inflation” in Keynes’ sense.) Keynes believes that a fall in money wages can lead to an increase in employment only through a concomitant fall in interest rates, and that the fall in interest rates would have the same effect without a fall in money wage rates.8 Does Keynes really believe that with an interest rate reduced to 1 per cent, and a wage rate of 60 cents per hour, the same things will be produced as with the 1 per cent interest rate and a wage rate of 50 cents per hour? I do not believe it. And I believe that the decision as to what will be produced is of decisive importance in determining whether or not the production can be maintained in the long run.

Different goods are produced if only interest rates, instead of both interest and wage rates, are lowered;—

—and this may have a bearing on the question whether production can be maintained without a setback.

If labour were the only factor of production necessary for the forced investment, then, given unemployment, investments could be carried out without affecting other branches and stages of production. This is not the case. There are always things whose supply is so scarce that the forced investment must have some adverse effects on, that is to say, withdraw factors from, other lines of production: this is bound to contribute later towards a more or less painful reversion in the alterations of the production structure.

Labour is not the only factor of production: other factors may be drawn away from other uses.

83. We may summarize the conclusions of the last sections by saying that movements of the price level, the volume of commercial paper, the production of goods in general or of consumers’ goods in particular, and also the existence of unemployment, give us no measure of the extent to which credit expansion may proceed without resulting in a crisis. We found that the limits to which a credit expansion can go without producing a crisis may be (theoretically fairly clearly) defined as follows: The expansion may go just so far as is sufficient to offset deflation due to spontaneous hoarding or to the increase in the number of holders of cash and in the number of pockets and accounts through which payments have to pass, and under certain circumstances it may go so far as to give “money in advance” against the voluntary savings of the immediate future. Only if and in so far as we are able to point to some practical indices of these factors which set the proper limits of a “healthy” credit inflation, can we say that there is unqualified justification for a policy of expanding credit.

If the credit expansion exceeds the limits mentioned, we have to allow for the probability of a set-back. Probably there will be many politicians who will estimate the danger of the future reversal as a lesser evil so long as the credit expansion helps us to surmount other present economic or political difficulties. This is not what we are discussing. We have confined ourselves to the question of the limits to which a credit expansion can go without causing those “cumulative-reversive” movements which form the essence of cyclical fluctuations.9

The danger of a set-back inherent in an expansion beyond the limits discussed is often considered a lesser evil than prolonged slackness.

Our conclusions make it easier for us to find answers to two questions which are relevant to our main discussion. (1) Are the “proper” limits of inflation (that is, the limits of credit creation which if passed will cause disturbances in the economic process) moved farther up or lower down, and (2) are the consequences of inflation likely to be milder or more severe, according as the created credit takes the form of loans to industry on the one side or loans to the stock exchange on the other?

Has the quality of credit any bearing on the “proper” limits of inflation,—

—or on the penalties resulting from excesses?

The first question meets with two diametrically opposed views. The one, according to which the stock exchange “ties up” credit, implies that the proper limits of inflation are reached later in the case of loans to the stock exchange than in the case of loans to industry; the other, which argues that stock exchange credit is used “more intensively,”1 or that it is more likely to be invested in fixed capital, implies that the limits are reached earlier in the case of stock exchange credit than in the case of loans to industry.

Are the limits narrower or wider for stock exchange credit than for other kinds of credit?

In reality, however—as will be demonstrated—it makes no difference, from the point of view of the “proper limits” of inflation, what kind of credit is given. Neither the duration of the loan, nor the purpose for which it is visibly (i.e., apparently) used, can deprive a credit expansion which goes beyond a certain point of its inflationary character. (It is, of course, another thing if strict provisions and conditions relating to the duration and the purpose for which the credit may be used by the borrower, have the incidental effect that they restrict the volume of borrowing. Here we are considering whether the duration and purpose of a given quantity of credit has much to do with its effects.)

A part from the quantitative credit control which is incidental to qualitative discrimination,—

There are probably a good many optimists who, on the basis of the maxims of practical banking, believe that loans granted for investment in working capital are less dangerous than loans granted for investment in fixed capital, and that a larger dose of the first may be risked than of the second. In the next chapter we shall call attention to certain serious misunderstandings as to the nature of working capital and fixed capital, and shall later show also that the effect of an increase in credit is hardly ever dependent on its form. Anticipating these findings, we may for the moment repeat that the “proper” limits of credit expansion are not affected by the nature and quality of the credit.

—our questions require examination of the nature of long-term and short-term capital,—

The second question relating to the effects of inflation is less significant once we have answered the first. One can never know whether an increase in credit granted for purposes of financing working capital, even if it is used according to the conditions prescribed, does not mean an increase in the amount of credit used for financing fixed capital in the system as a whole, for every increase in the supply of credit permits the fuller satisfaction of the aggregate demand for credit and therefore makes it possible for some borrowers who were previously excluded from the market by the competition of others, to satisfy their demand for credit. Who can tell what kind of demand for capital will be exerted by those entrepreneurs who were previously unable to borrow on the credit market, but who now become the “marginal borrowers”2 in consequence of the easing of the market?

—and of the possibilities of controlling the uses to which credit is put.

This scepticism as to whether we can tell exactly what is the eventual “final” use to which new credit is put, shall not prevent us from analysing the results of the new credit if it is used in one way rather than another. For this purpose, however, we must first examine critically the existing views as to the “fundamental” difference between investment in working capital and investment in fixed capital.

____________________________

The tricky qualifications which are necessary (but so easily overlooked), if the interest rate is explained as a function of the stock of money, are a serious disadvantage of this approach. The traditional explanation of the interest rate in terms of a flow of loanable funds is preferable. It is easy to see that an increase in the stock of bank money constitutes an addition to the flow of loanable funds in the period in which the increase takes place. The same is true for a release of hitherto idle funds or the use of temporary surplus cash balances.

For an able discussion of the “stock v. flow” analysis, see E. S. Shaw, “False Issues in the Interest-Theory Controversy,” Journal of Political Economy, Vol. XLVI (1938), pp. 838-856.

J. M. Keynes in his Treatise on Money, Vol. I, defined the natural rate of interest without reference to the price level, solely on the basis of the equilibrium in the capital market: “Thus the natural rate of interest is the rate at which saving and the value of investment are exactly balanced” (p. 155). Nevertheless, the connexion with the stable price level is implicit in Keynes’s fundamental equations.