The Stock Market, Credit and Capital Formation

CHAPTER VI

THE DEMAND FOR MONEY BY THE STOCK MARKET

36. In this chapter we shall carry out the promise made on several occasions in previous chapters to analyse our problem more closely from the standpoint of monetary theory.

First of all, we must examine the argument that the stock exchange takes money, or circulating media, away from other markets. This argument is advanced even by authors who disagree with the thesis that capital is tied up on the stock exchange. It goes without saying, of course, that those who defend the theory of the tie-up of capital implicitly hold that purchasing power is tied up.

It is held that the stock exchange takes circulating media away from other markets,—

According to Reisch, there is “no doubt whatever”1 that circulating media are tied up by stock exchange transactions and are released when the stock exchange boom comes to an end.2 His view has been very neatly put by H. von Beckerath in the following sentences: “The money which is withdrawn from expenditure on the markets for goods and labour, and used as unit of account for business on the stock exchange, leads to a temporary reduction in the demand for goods and for labour. This is to say that the money is held up on its way and for the time being can neither be spent nor lent in the economic process proper.”3

The idea, then, is that the demand for other economic goods is reduced in favour of the demand for securities. It is a fairly generally held opinion that by exerting a “demand for circulating media,” the securities market comes into competition with other markets. Balogh, for instance, says that “circulating media move from one market to another but are ‘held up’ on each of them for some short or long interval of time.”4

—and that the demand for goods falls as the demand for securities rises—

He speaks of a “circulationary tie-up”5 to indicate that circulating media are held up for a particularly long time on a rising stock market. Palyi, famous for his sharp wit and tongue, also finds, in an analysis of American conditions, that “the remainder of the circulating media . . . were used to purchase securities and real estate and were until recently tied up in these uses”6; he thinks it necessary to add somewhat scornfully in parentheses: “There is a new-found theory which holds that the stock exchange never ties up capital even in the short run, but that the money paid in the morning flows out into the ‘economic system’ in the evening in order to return to the stock exchange the following morning: no account will be taken of this ingenious theory here.” Nor will any account be taken here of this ingenious method of criticizing the caricature of a theory.

In so far as the argument concerns not the provision of capital but the provision of circulating media, Cassel (at whom Palyi’s ironic comments were aimed) also inclines to the view that the securities market competes for circulating media with the rest of the economic system. This is apparent from Cassel’s remark that when the demand for money by the stock exchange rises the commodity price level can be kept stable only by the creation of new bank money. “We must therefore,” he says, “come to the conclusion that, if the Stock Exchange should require an increase in the amount of money in circulation, . . . the increase can and should be made by the creation of new means of payment in proper adjustment to the aim and view: this money will doubtless consist mainly of bank credits on cheque account. In this case the amount of money available for industrial and commercial purposes will remain unchanged, and the general level of commodity prices can thus be kept constant. Hence, providing that the bank policy is as rational as has been assumed, the Stock Exchange cannot, in this case either, have a disturbing effect on the amount of money available for industry and trade.”7 If this proposition is correct, then, in the absence of a “rational” banking policy and under the assumption of “other things being unchanged,” the demand for money by the stock exchange will “have a disturbing effect on the amount of money available for industry and trade.”8 By “disturbing effect” is meant, of course, a fall in the commodity price level or, more generally, a decline in demand on the markets for commodities. Despite the astonishing unanimity among the various authors on this point, the correctness of the assumed causal nexus “rising stock market: falling commodity markets” may still be questioned.

—and that new bank money should be created as the stock market demands more money.

37. The conception of a demand for circulating media by particular markets is, in my opinion, not always a very fortunate one. The same can be said of Balogh’s assumption of the “temporary deflationary effect of the stock exchange boom.”9 If higher prices and an increased turnover on one market tie up more purchasing power, there is, according to such reasoning, a consequent deflationary effect on other markets. If we use this form of expression, we have therefore to say that an increase in prices or sales on the fruit market causes a deflation on the fish market.

Increased demand on one market is said to mean “deflation” for other markets.

If the demand for fish falls off in favour of the demand for fruit, then, if other circumstances remain the same, fruit prices will certainly rise and fish prices fall. But it is no explanation of this price shift to say that “an increased demand for circulating media on the fruit market has a deflationary effect on the fish market.” It is self-evident that if there is a shift of demand from one commodity to another, and purchasing power is used to buy another commodity in place of the one previously preferred, the price of the one for which the demand increases will rise at the expense of the price of the one for which the demand declines. Can it be inferred from this that increased interest in securities must raise the prices of the latter at the expense of the prices of commodities?

Do security prices rise at the expense of commodity prices ?

When someone wishes to acquire securities and obtains the funds necessary for the purchase by refraining from buying things that he previously used to consume, the consequent shift in demand is called “saving.” The savings process represents (if no hoarding is involved) a shift in demand from present goods to future goods, and leads, by way of the corresponding shifts in prices, to shifts in production. It is usually assumed that a significant price shift takes place not only between consumers’ goods and securities but also between consumers’ goods and producers’ goods. It may seem strange that the price fall in consumers’ goods should correspond on the other side to price rises in two categories of things at the same time. But there is nothing complicated about this, for the rise in price of titles to capital goods may actually involve the rise in the prices of the capital goods themselves.

A shift of demand from consumers’ goods to securities is “saving.”

Consumers goods fall in price: can both securities and producers’ goods rise correspondingly ?

Those who are accustomed to think in terms of a constant velocity of circulation of money will probably not find this explanation easy to accept. According to their view money (under which we include bank deposits on current account) performs a fixed number of transactions in a given period of time, and the price level is determined by the turnover of goods, the quantity of money, and the fixed velocity of circulation. The velocity of circulation is accordingly not conceived of as a dependent variable. Our reference to the shift in demand from consumers’ goods to capital goods, however, implied “additional transactions”—the purchase of securities—and those who adhere to the theory of a “constant” velocity of circulation will reject the argument according to which the titles to capital goods (securities) as well as the actual capital goods rise in price.

No—if the transactions velocity of circulation is constant;—

There are other authors, however, who assume not a constant transactions velocity of money but a constant income velocity or circuit velocity. These authors will find no difficulty in accepting the proposition that security prices and producers’ goods prices rise together, and that the supposed causal nexus “rise in security prices: fall in commodity prices” does not hold.

—yes—if the income velocity of circulation is constant.

But any assumptions of this kind concerning the velocity of circulation of money are quite arbitrary. It is necessary to ask whether the turnover of securities lengthens the circuit round which money has to flow, and if this is the case, whether the transactions velocity may not rise correspondingly. If the first question could be answered in the negative, or the second in the affirmative, then the argument that the commodity price level is independent of the volume of transactions on the securities market would be substantiated.

A priori assumptions about velocity are useless.

38. The more plausible argument may seem to be that a rise in security prices and an increase in the turnover of securities must lead to a fall in the so-called price level, simply because the demand for money by the securities market and therefore by the economic system as a whole will have risen. Given an increased demand for money or circulating media1 and an unchanged supply of money, it would be difficult to imagine anything else than an inevitable fall in prices. The question, however, is whether an increase in turnover of securities, which may be an increase in the number of securities traded or a rise in their prices or both, involves an increased demand for money or circulating media. This question is suggested by the fact that the turnover on the security exchange is effected, for the most part, not with the use of circulating media, but by a system of reciprocal cancellation, or, that is, by a clearing process.

It is question whether increased security turnover involves an increased demand for money.

The introduction of the clearing mechanism into our analysis at this point is essential. There is no other sphere of the modern economy besides the stock exchange for which one is justified in arguing that the clearing mechanism may take care of an increased turnover and avoid an increase in the demand for money. It was made clear by Mises that we are seldom justified in supposing that an increase in the demand for money due to additional business will be automatically “compensated” by an extension of the clearing mechanism.2 Mises said explicitly: “An extension of the clearing system . . . can never be called forth automatically by an increase in the demand for money.”3 Nevertheless it seems to me that stock exchange business, when there is an increased turnover of securities between members of the stock exchange, is a special case which falls outside of this proposition.

The stock exchange clearing may obviate an increase in the demand for money.

39. The clearing procedure is an almost indispensable part of the technique of operating on the stock exchange. With transactions within a more or less closed circle of people, most of the claims can be settled by balancing with counter-claims without the use of money. Such reciprocal cancellation will be possible for a major part of all claims even when such procedure is confined to the transactions of a single day. The possibility of using this off-setting procedure is greatly extended when the business of several days is brought together in a settlement period. This practice is followed on many leading stock exchanges even where there is a legal prohibition against forward dealings and only cash business is allowed.

The members of a clearing system settle their mutual claims by reciprocal cancellation.

The gross value of the securities traded on the stock exchange never has to be paid either in cash or by cheque; only the differences have to be paid. Very few people fully realize what an important part is played by this process of off-setting claims against each other and paying the differences. In a book published as long ago as 1905 it was estimated, by one who was well acquainted with the facts,4 that on the English stock exchange 90% of the obligations were settled by off-setting and only 10% were paid by means of bank cheques.

A passing reference may be made here to Albert Hahn’s treatment of the whole problem of stock exchange credit. In Hahn’s view, the main problem is whether or not the stock exchange absorbs money in the narrower sense, i.e., cash. “The purchase of securities, the so-called stock exchange turnover, as such takes place almost exclusively without the use of cash and therefore . . . exerts scarcely any effect on the credit market.”5 The transaction of stock exchange business without the use of cash is an essential element in Hahn’s theory of credit. As he regards the demand for cash as a decisive factor in the determination of the rate of interest,6 he attributes more importance to the absence of the use of cash in stock exchange operations than most other authors. When we say here that the turnover of securities need not involve any increase in the demand for money or circulating media, we mean not that it requires only bank money and no cash, but that the major part of the stock exchange turnover requires neither the one nor the other. The off-setting mechanism makes it possible very largely to dispense with both cash and bank money.

They use no cash—

—and very little bank money.

40. These considerations do not, however, exclude altogether the possibility of a rise in the demand for money during a stock exchange boom. Even if only a fraction, let us say 10%, of the turnover on the stock exchange makes use of the monetary circulation (bank deposits of course), then, assuming a constant ratio between the volume of transactions effected by the clearing mechanism and the volume of transactions effected with the use of cheques, an increase in stock exchange turnover would still cause an increase in the absorption of bank money in absolute figures. It is, however, not correct to assume that the proportion of the transactions which can be settled by clearing remains constant. It will be immediately apparent that when the volume of transactions increases, the possibilities of off-setting are augmented not only absolutely but relatively, and that the balance of the differences which have to be paid is not proportional to the level of transactions.

The ratio of cheque payments to total turnover falls as turnover increases.

The notion that a rising stock market requires a larger volume of circulating media than a falling market is, so far as concerns the narrower circle of operations, i.e., those which come under the settlement procedure, not valid, since falling prices are just as conducive to “differences” as rising prices. Since it is only the differences which have to be settled by payment, and differences are of equal frequency on a rising market as on a falling one, the a priori assumption that in a boom an increased circulation of money is needed for the purpose of “in-and-out trading” of securities on the stock exchange is unfounded.

Rising prices are no more cause for increasing clearing differences than are falling prices,—

The stock exchange turnover may increase by “quantity” or by “value,” i.e., more securities may be traded at unchanged prices or the same number of securities may be traded at higher prices, and there are of course any number of possible combinations of these factors. The proposition that there is no logical necessity for the differences for settlement to rise with an increase in the turnover figures, holds equally well for a “quantity” or a “value” increase in turnover. If a group of speculators undertake a large number of transactions among themselves, the balance which remains to be paid after off-setting need not be greater than it would be if the turnover had been smaller.

—and a higher turnover need not be cause for higher clearing differences either.

An increase in turnover will tend to bring with it an increase in differences to be settled by cheque payments only if the increased business is not evenly distributed among the various clearing-house members, or, more correctly, if the unevenness in the distribution of business among brokers is increased by the increased turnover. The probability that this will happen in the course of a stock market boom is fairly high for the following reasons: (1) brokers are often specialized as to the type of customers they serve, and an increase in trading may find market opinions divided as between these types of customers; (2) different brokers may have different opinions as to expected market developments, and may advise their customers accordingly, so that selling and buying orders are unevenly distributed over the brokers.

Whether balances to be settled by check rise with an increase in turnover,—

depends on the distribution of selling and buying orders over the brokers.

The sales must, of course, equal the purchases. If one half of the brokers served the customers who did the selling, and the other half of the brokers served the customers who did the buying, then any increase in turnover would involve an equal absolute increase in payments for settlement. If, however, each broker served both selling and buying customers, the absolute increase in cheque payments would fall short of the absolute increase in turnover. If the increase in turnover were such that the distribution of sellers and buyers among brokers remained unchanged, total turnover and payments for settlement would rise in the same proportion. If the increase in turnover were such as to make for a more even distribution of buying and selling orders among the various brokers, the amount of cheque payments would fall relatively, and indeed, might fall absolutely. All these developments are possible; an inspection of clearing-house statistics7 shows that in the past the amount of payments for settlement usually rose absolutely but fell relatively with an increase in turnover.

Check payments might conceivably fall with increased business;—

—actually they rose for the most part.

The absolute increase in cheque payments which may thus accompany rising stock market transactions can be taken care of out of unchanged totals of brokers’ cash balances. In other words, there is no logical necessity for a rise in clearing balances requiring settlement to cause a rise in the bank balances held by brokers at the close of the day. There is almost no statistical evidence available which might show whether, in point of fact, brokers carried larger balances when transactions were larger. The reason why they easily could do more business without higher bank balances will become obvious from an analysis of the mechanism of stock exchange loans.8

Increased clearing balances, however, can be settled without increased bank balances.

41. The settlement procedure is open only to actual members of the stock exchange, i.e., the jobbers (dealers) and brokers. Securities are traded, however, not only between members of the stock exchange but also between brokers and the public. It would be a serious error to confine our investigation of the problems connected with the stock exchange to the activities of the professional dealers, since the activity of these dealers is directed by the willingness to buy and sell on the part of the public. There is an old stock exchange joke which says that no inn can keep going in the long run if the bar-tenders have nothing else to do except play billiards with each other. It can exist only if there are customers to serve. In the same way the members of the stock exchange live not from “playing with each other,” but from the operations which they undertake on behalf of the public. A rise on the securities market cannot last any length of time unless the public is both willing and able to make increased purchases. But all that was said about the clearing mechanism dispensing with the use of money does not apply to transactions between the public and the stock exchange, and we must therefore continue our investigation in this direction.

No special cleaning exists for payments between brokers and the public,—

The fact that “inside business” on the stock exchange, as I have tried to show, need not have the effect of tying up more circulating media in times of boom does not mean that “outside business” carried on between brokers and the public may not have this effect.

—hence, money might be tied up for these payments.

42. The attempt is sometimes made to dispose of the argument that the stock exchange or the speculating public cause a tie-up of circulating media by comparing the process to a “sieve with wide holes.” The securities market is supposed to be analogous to a sieve because the sellers of securities obtain the money just as soon as it is invested by the buyers, and the circulating media to a certain extent merely “run through”; they remain at the disposal of the whole market without any “tie-up” or “absorption.” But this formula could be applied just as well to any other market with the result that money would never be “held up” or “absorbed” anywhere. The seller of a commodity also obtains the money spent by the buyer, and it would be impossible to explain any price shift if we were to argue that the seller might spend the money he received for his commodity to purchase the goods which the person who bought from him had to forgo.

The “sieve” analogy explains little:—

—on every market the seller gets what the buyer pays.

Similar views are often to be found elsewhere, as, for example, among those who advocate a policy of subsidizing certain groups of producers for the purpose of giving them more purchasing power to spend on other products. It should be obvious to anyone after a little thought that, with a given speed of transactions and a given quantity of monetary media, an increased total expenditure on one product can take place only at the expense of a diminished total expenditure on another product.

It should be noted that this proposition relates not merely to increased prices but to increased “outlays,” i.e., the product of price times quantity. Emphasis might also be laid on the word “product” in another sense, for it might be possible for money to circulate at a different speed in respect of payments for products than in respect of other payments. The theory of the so-called “cession payments,” as it was, for example, developed by “Wieser,9 seems implicitly to assume that payments which do not relate to purchases of goods take place, so to speak, “in no time,” or, more precisely, that they edge their way in between the payments for purchases of goods without causing any postponement of the latter.

The theory of “cession payments—

When such a “cession payment” or “transfer payment” takes place, the payer makes over his buying power to the payee without, according to Wieser’s theory, necessarily causing any changes in the direction of production. As examples of transfer payments, Wieser referred to loans, investments, insurance premiums, gifts, charity, tax payments. These payments are, in themselves, not supposed to have any effect on the disposition over goods and on the production of goods; it is only as the recipients come on to the market for goods and services that they can, through their purchases and the respective “price payments,” cause changes in the direction of production in so far as they use their buying power in a different manner from that in which those who previously held command over the funds had used them. If, for example, a borrower, or a recipient of charity, buys the same things as the lender, or the benefactor, would have bought, then in a stationary economy the transfer payments would have caused no change. According to Wieser’s theory, the transfer payment itself may be regarded as directly indifferent from the standpoint of the price system.10 It is only the subsequent price payments by the recipient that can lead to price shifts. Thus the demand of the borrower will, for instance, raise the prices of certain means of production while the decline in the demand of the lender lowers the prices of certain consumers’ goods; or the demand of the recipient of relief will raise the prices of certain consumers’ goods while the decline in the demand of the benefactor or the taxpayer causes the prices of certain producers’ goods to fall.

—or “transfer payments”—

—holds that there are certain payments which do not influence prices.

Wieser did not explain how the mechanism of payment differs in the case of transfer payments from the case of price payments, or how the time sequence of payments should make the direct effect of transfer payments neutral towards the price system. He was obviously concerned exclusively with the system of mutual interdependence of commodity prices, and he made certain simple assumptions which avoided complicated questions connected with the circuit flow of money. The assumptions he made are essentially the same as those which are implicit in the concept of “neutral money.” Under the assumption of neutral money, disturbances of the circuit flow of money cannot occur, or must somehow be compensated.

Apparently “netural money” is tacitly assumed.

43. Wieser should not, however, have stopped his analysis where he did. Given neutral money, not only the transfer payments which he enumerated would be indifferent from the standpoint of the price system, but certain price payments would be equally “indifferent.” Payments for goods which cannot be produced or reproduced or of which the production cannot be may increased, would have to be regarded as “indifferent” in the described sense—indifferent because the prices paid for these goods cannot exert any influence on their production or on the disposition of the productive factors.1 Let us assume a stationary state and suppose that a certain individual A possesses a highly prized picture by a celebrated painter. The picture comes under the category of non-reproducible goods. Now if B wants to acquire this picture and obtains it at a high price, then B’s payment to A need not result in any shift in the interdependent price structure of the economy, providing A uses the purchasing power he acquires in the same way as B would have used it had he not bought the picture. Here we have an example of a price payment which is—all this still under the neutral money supposition—of the same “indifferent” character as a transfer payment. The high price fetched by the picture would leave all other prices unaffected. The same thing might be true in the case of any good which is the object of exchange, so long as, whether for technical, legal, or economic reasons, its production cannot be increased despite the rise in its price. The prices of such goods may rise without necessitating any changes in other prices.2

Certain “price payments” may have the character of “transfer payments”—

—if they are for goods whose production cannot be affected.

The conclusions of the previous paragraphs might also be relevant to the case of a rise in security prices. The payment of the price of the securities is in the nature of a transfer payment. The purchase of securities is neither more nor less of a transfer payment than every loan; it is a transfer payment acknowledged by a special kind of certificate or receipt. And if the seller of the securities uses the purchasing power he received in order to buy the same goods as those of which the buyer of the securities relinquished the purchase, and if the purchase takes place at the same time as it would have been made by the buyer of the securities, then the rise in security prices will leave commodity prices unchanged. But if the seller of the securities buys producers’ goods, as may happen especially in the case of new issues, then the prices of these goods will rise at the expense of those goods which the buyer of the securities had to give up. The so-called general commodity price level, exclusive of securities, would not, however, be affected: security prices could rise without there being any consequent fall in other prices in general.

Payments for securities are undoubtedly of the nature of transfer payments.

44. This theory of transfer payments is, however, of no value in explaining reality unless it can be plausibly shown that the recipient makes use of his purchasing power without delay. “Without delay” means: at the same time that it would have been used if the transfer payment had not been made. Suppose, under conditions of a stationary circuit flow of money, N had to make a price payment to M, the person next to him in the circuit; instead of doing this he made first a transfer payment to N’ which enabled the latter to take over the goods from M; if a time interval, however small, elapsed between the receipt of the transfer payment and the use of these funds for a price payment, then a postponement of the demand for the goods and a consequent tendency to a price fall would be unavoidable.

The theory of transfer payments is of no value unless it proves absence of delays.

When transfer payments delay the purchase of products, product prices tend to fall.

The assumption of a loss of time resulting from the transfer payment can be avoided only under one condition. If N makes the transfer payment to N’ earlier than he would have made the price payment to M, then the payment by N’ can reach M still without delay. It is possible to think of a number of institutions, or habits, which make it probable that many transfer payments do take place more quickly after the receipt of income than would expenditures on the market for goods. The income-recipient who hands over purchasing power to his wife or his housekeeper does it in such a way that the purchases take place no later than if he had had to go to the market himself. The debtor who intends to devote part of his income to debt payment will usually make the necessary transfer immediately after he receives his income, whereas he will make his purchases of commodities only gradually over the income-period.

It may well be that transfer payments are so fast that purchases are not delayed.

Can we say the same of loans or of the acquisition of shares by savers? In a schematic picture of the circuit flow of money, we might assume quite arbitrarily that incomes were paid out regularly on Thursday; that the loan market functioned on Friday; and the commodity market on Saturday of each week. In this case transfer payments, however large, would not delay the purchase of goods; or, to use another terminology, the demand for money by the economic system as a whole would be independent of the turnover on the credit market; or, to use still another formulation, the increase in the “money work to be done,” i.e., the increase in money transactions, would be “automatically” compensated by a rise in the transactions velocity.3

Various institutions—

This institution of the Friday loan market and the Saturday commodity market is far from existing in reality. Nevertheless it is still possible that in reality something does take place which allows the results of this imaginary institution to be approximately achieved. Budgeting in advance by the majority of income-recipients, for example, would tend to have the effect indicated. If the individual budgets to save a fixed proportion of his income, and decides to use his savings to purchase securities, it is very probable that he will do this right at the beginning of the income-period, so that his average cash balance will be lower than it would have been if he had spent all his income on consumption. Thus, if he buys the securities from someone who wants to use the proceeds in the commodity market, it is not unreal to assume that they will be used there no later than would have been the case if no transfer payment had intervened.

—and habits—

—may help to avoid delays arising from transfers of money capital.

A fairly plausible case can thus be made out for the hypothesis of neutral transfer payments. But the strange thing is that very few authors have bothered their heads about the loss of time caused by transfer payments when they have been dealing with ordinary loans, relief payments, tax payments and the like.4 It has been thought necessary to emphasize the lapse of time only in the case of transfer payments connected with the securities market. In so far as it is simply a matter of the flow of purchasing power through the stock exchange, i.e., the transfer of purchasing power from the purchaser of the shares to a seller who intends to use it to purchase goods or services, it is difficult to see why the lapse of time should have been thought a greater evil here than in the case of ordinary loan transactions and other transfer payments. As the argument usually runs in terms of whether stock exchange credit has harmful consequences which other kinds of credit have not, it is unnecessary to try to prove that stock exchange credit finds its way onto the commodity market in “no time”; the question is only whether the purchasing power transferred is likely to take “more time” before it becomes demand for goods and services in the case of stock exchange credit than in the case of other kinds of credit.

Not whether funds flow through the stock exchange in “no time” but whether it takes “more time” than through other channels, is the question.

45. The reasons which may permit the neglect of the lapse of time associated with transfer payments and certain analogous cases of price payments, particularly security transactions, are not sufficient to allow us to disregard the lapse of time which takes place when there is a continual repetition of the same event. If the recipient of the transfer payment again transfers the purchasing power to someone else, and the next recipient does the same thing so that no demand for products is exercised during this time, then the interval which elapses before such a demand arises cannot be disregarded.

The delay caused by a chain of transfer payments must not he neglected,—

A case where a long chain of transfer payments may occur is perhaps to be found in connexion with the conversion and funding of credits: loans that have just been raised may be used to pay back old loans, and the sums repaid may be re-lent in order to be used again for paying back other loans and so on. However, no authors have regarded this problem as an important one. Other cases, too, are conceivable where a series of three or four successive transfers may take place before the purchasing power is employed on the market for goods and services.5 It has already been noticed that successions of purchases and sales can take place on the securities market, and that there is consequently a strong possibility that purchasing power may change hands many times without being used on other markets.

—and repeated security transactions might be a case in point.

We have already emphasized, perhaps more than enough, that the purchase and sale of securities within the actual stock exchange, i.e., between the members of the stock exchange, by reason of its clearing organization requires hardly any circulating media and that an increase in turnover scarcely requires more circulating media. Here, however, we are concerned solely with payments between brokers and the public which usually take place through ordinary circulating media, mostly bank deposits. If purchaser A gives a cheque to his broker and seller B asks for a cheque from his broker, if B then hands his cheque back to his broker for the purpose of buying other securities and C the seller asks for a cheque for his sales proceeds, and this process goes on repeating itself, then for the time that it lasts cheque accounts will be tied up in security speculation.

If speculators who sell ask for cheques from their brokers, and send cheques for their new purchases, bank deposits are tied up.

The picture just drawn does not, however, represent the situation on all stock exchanges. In the United States, for example, it is by no means usual for the seller of securities who is contemplating buying other securities to request his broker to give him a cheque for the amount due. It is more usual for him to leave it on account with his broker until his new order to buy has been given and executed. The broker, however, will not maintain idle bank deposits to the amount corresponding to the total of all the deposits his customers keep with him. He is more likely to use his bank deposits, once they exceed a certain minimum, to cover his debts, or if he has no debts, to grant loans. Customers who keep on selling, buying, selling, buying, do not therefore use ordinary circulating media such as bank deposits for these transactions: their accounts with brokers perform the function of purchasing power between these customers. There is thus a separate money, so to speak, in the form of brokerage deposits, which serves to effect security transactions between the regular customers of the brokers.6 The buyer of shares draws on his broker-age deposit and the seller acquires a brokerage deposit, which, the next time he buys securities, is transferred to a third speculator and then to a fourth and so on. In short, speculation by the public can also proceed without the use of bank deposits, or, that is, without the use of ordinary circulating media, so long as the seller does not require his broker to pay out what is due to him. (See Appendix B for a description of the circulation of brokerage deposits.)

But in the United States speculators customarily keep their deposits with brokers, if they plan further trading,—

—and the brokers’ bank deposits may be only a fraction of the customers’ deposits with them.

Hence the customers’ “brokerage deposits” are a peculiar type of money—

—for the speculating public.

On some exchanges, however, it is usual for the broker to send a cheque to the seller of the securities “automatically,” or, that is, without being especially requested to do so. And even on exchanges where this is not the general rule, there are customers who request payment by cheque. If customers after having taken their funds away from their brokers continue to speculate, a chain of transfer payments is carried on with bank deposits. Such a chain also occurs when customers withdraw their funds from their broker in order to lend them (not on the same day) to other speculators, who perhaps again buy shares from people who demand immediate payment by cheque and do not decide until later either to speculate further or to lend their funds at call. In short, when sellers keep their sales proceeds, not with their brokers but on account with their bank, until they decide to use them on the stock exchange again, there is undeniably a tie-up of the deposits in question.7

Where, however, speculators switch back and forth from brokerage deposits to bank deposits, bank deposits are tied up.

46. All that remains to ask is: when are these chains of transfer payments between the bank accounts of security speculators likely to arise? From all that we have seen up to now it would appear that in order that this shall happen it is necessary for the “speculative fever” to infect a very wide circle; this means, in the case of most countries, that it must extend to circles which are not regular customers of the brokers and which usually have little or nothing to do with security transactions. Moreover, the chain of transfer payments comes to an end as soon as a seller uses the sale proceeds to purchase goods or services (which will most often occur if the seller is a firm issuing new securities). For a continued chain of stock transactions it would therefore be necessary that the incentive to new issuing activity, or, more specifically, the incentive to real investment be smaller than the incentive to security speculation.

This condition arises only when the speculative fever affects the general public,—

—and when real investment is much less attractive than stock speculation.

Let us therefore consider the possible causes of security speculation by the public and ask whether the conditions formulated above are likely to prevail. The motive for security speculation by the public lies in the expectation of further increases in security prices. These expectations may be based in the first place on prospects of increased dividend payments by the corporations. If the prospective profits of the enterprises actually do rise, there will be a corresponding desire to expand and an increased demand for capital on the part of these enterprises. This demand for capital for industrial activity will induce firms to sell out their holdings of securities (securities held in portfolio) and to float new issues: thus the same motive which invites purchases of securities will also lead the sellers to employ the sales proceeds in production.

Good prospects for business profits invite both demand for stocks and real investment.

It may happen, however, that the expectations of a rise in the prices of securities have no such material justification. There may be a feeling of optimism which calls forth a supply of liquid cash balances (dishoarding) sufficient to turn it into an effective demand for securities. This is especially likely to occur if the movement is supported by additional bank credit. We shall return to this point presently. Here we want to inquire what happens in the case where there is no increase in the supply of money capital and no rise in profits of the enterprises to form the basis for the sudden development of boom sentiment. The most probable result in this case is a quick recession of security prices. For higher stock prices will invite a new supply of securities, and the corporations, which want to take advantage of the higher prices in order to draw funds from the stock exchange and use them for real investment, will find that there are no additional funds to be had. Chains of speculative security transactions are, therefore, hardly likely to develop in these circumstances.

No boom can develop unless optimism, is supported by increased supply of money capital.

It is impossible for the profits of all or of the majority of enterprises to rise without an increase in the effective monetary circulation (through the creation of new credit or dishoarding) unless industry is presented with a general fall in wages or a reduction of taxation. Under these circumstances the improved profit prospects will, it is true, cause security prices to rise, but this rise will take place almost at one stroke and not by way of a gradual upward movement in the stock market. Chains of speculation can develop only as the result of continual price rises over a longer period. A single rise in the level of profits cannot produce a continuous rise in capital values and cannot, therefore, lead to extensive speculation by the public.

Continual rise in stock prices cannot be explained by better business.

47. A factor which is capable of evoking expectations of a rise in security prices is a reduction of the interest rate. In so far as this reduction occurs merely as the result of an increased supply of intended new savings,8 the likelihood of a long-lasting upward movement of the market is rather meagre. It is easy to see that if dividend prospects are unchanged and the rate of interest is reduced, security prices will rise,9 and it is more than probable that a sufficient amount of security sales from “final sellers” (unloading by temporary holders and new issues) will be quickly forthcoming: comparatively small offerings of securities will suffice to absorb the increased supply of new savings and to drain them off to other markets. For no matter how the supply of money capital derived from current new savings may fluctuate, it is scarcely conceivable that the total supply of money capital can ever rise to unexpected dimensions as the result of an increased flow from this source. If the public devotes only its new savings to the securities market, and the new demand at once causes some groups of securities to become “firmer,” it will not be necessary for the purchasing power of the public to be withdrawn from the commodity market until it has “run through” all the securities quoted on the exchange and has adjusted the prices of securities, one after the other, to the new market conditions. In reality this task is performed by quickly reacting professional speculators who require no money to carry out their transactions. The public’s money is not “held up” because the professional speculators, who discount the public demand, will already have raised the level of security prices and thus called into being a new supply of securities from the producers.

Continual rise cannot be caused by increased voluntary saving either.

A new supply of securities would quickly stop soaring stook prices,—

—if the demand came only from intended new savings.

If it were not for the elasticity of bank credit, which has often been regarded as such a good thing, a boom in security values could not last for any length of time.10 In the absence of inflationary credit the funds available for lending to the public for security purchases would soon be exhausted, since even a large supply is ultimately limited. The supply of funds derived solely from current new savings and current amortization allowances is fairly inelastic, and optimism about the development of security prices would promptly lead to a “tightening” on the credit market, and the cessation of speculation “for the rise.” There would thus be no chains of speculative transactions and the limited amount of credit available would pass into production without delay.

The supply of intended savings and amortization allowances is fairly inelastic.

Only if the credit organization of the banks (by means of inflationary credit) or large-scale dishoarding by the public make the supply of loanable funds highly elastic, can a lasting boom develop. The demand for credit by optimistic speculators rises as the borrowed funds are used for stock purchases from “final sellers.” The reason why this increased demand does not lead quickly to the exhaustion of the supply is that the supply of credit is not restricted to the scarce supply of current new savings: If the demand rises the banks are able to grant additional credit on unchanged or practically unchanged terms. The professional speculators cannot anticipate the entire development at one stroke because they do not know to what limits the credit expansion will go. The upward movement of security values which is kept going by this means is capable of producing a chain of speculative operations, and it is then possible for the money derived from credit expansion to remain “tied up” for a time in a succession of transfer payments connected with stock exchange transactions.

A lasting boom can result only from inflationary credit supply.

Since credit inflation is a condition for chains of speculative transactions, it is only inflationary funds which are in danger of becoming tied up.

It does not, of course, depend on the origin of each particular dollar coming onto the stock exchange whether it will be drained off to other markets immediately or only after some delay. This is not what was meant when we said that it is the money derived from credit expansion that is likely to be tied up in stock exchange transactions. It is of course possible for funds which come out of real savings to “get stuck” in the way described, but this is only probable if a particularly abundant credit supply has been produced by the emergence of inflationary credit. It is not the origin, but the excessive dimensions of the supply of credit, which is the decisive factor. The supply can, however, reach these dimensions only if it comes from an inflationary source.

48. We have shown that it is possible for bank deposits to be temporarily tied up in stock exchange operations and so not to flow immediately into “productive channels.” Does this lend support to the view that there can be an “absorption of the country’s credit in speculative security operations to an alarming extent”1; or the view that stock exchange operations rob industry and “legitimate” business of the use of the available supply of capital?

The possibility of absorption of funds by stock transactions—

These views have not been substantiated so far. We have given sufficient proof that only a very small fraction of all stock exchange transactions are capable of tying up bank money. Moreover, it should be remembered that:

—is limited to smallest dimensions,—

(1) our conclusions related to inflationary credit or rather to periods when inflationary credit was being created;

—and limited to inflationary credit,—

(2) the money which flows onto the stock exchange and is tied up in a series of operations, need not come directly from stock exchange credits (brokers’ loans) but that any “inflationary” credit, no matter in what form it was created, may find its way onto the stock exchange;

—but not limited to stock exchange credit.

(3) an important distinction has to be drawn between a delay in the productive employment of funds derived from intended savings and a delay in the productive employment of funds from inflationary sources.

The fact that stock exchange speculation by the public may tie up inflationary credit will probably not be judged an evil once the effects of this inflationary credit on production are realized. If inflationary funds were held up for the time being on the stock exchange, there would be a temporary “localization of the inflation.”2 The vague notion which many people have had of funds being “held up”3 may in this case be not so far from the truth. Here we have attempted to make this vague notion more precise by describing the conditions necessary for a long series of transfer payments. The results of our analysis prevent us from making the mistake of speaking generally about the “deflationary effect of the stock exchange boom” where, at best, the effect is one of temporarily absorbing part of the inflation.

The absorption may be considered a temporary “localization of the inflation,”—

The newly created funds may make their way in the first instance to that section of the public which is interested in the securities market; the boom sentiment of these people raises the prices of the shares of various classes of enterprises and some sellers may hold their funds on bank account for intervals between transactions. As various sellers “get out of the market” and as new issues are floated, the inflationary credit is drained off into production.4

—before the funds are drained off into production.

The phenomenon of the temporary tie-up of inflationary credit in security speculation would be very useful in assisting the monetary authorities to frame their credit policy. If the volume of credit of all the banks and the movements in the securities market were carefully watched, it might be possible to put an early brake on the boom and thus succeed in avoiding a more violent reaction. A restrictive credit policy applied at the right moment would check the progressive watering of the capital supply through the expansiveness of bank lending. A measure of bank policy of this kind should not, however, be associated with any such foolish slogans as “Down with stock exchange credit and let industry have it!” because the very purpose of the measure would be to stop the expansion before the credits had given an excessive stimulus to industrial activity. If the stock exchange really had the power to absorb inflationary credit for good and for all, it would probably be a very healthy arrangement from the point of view of industrial production, because the misdirection of investment which is caused by the “artificially” easy facilities for procuring capital would be avoided. In reality, however, stock exchange transactions tie up only a relatively trivial amount of the inflationary credit and do so merely for a short time. The stock exchange credits begin to “work” only too quickly on production. If the authorities are aiming at a rational banking policy, they should not complain of the stock exchange withdrawing money from industry, but should take advantage of a temporary localization of the inflation to try as far as possible to neutralize the overflow of the latent part of the inflation into production by putting a brake on the credit expansion.5

If there were absorption to any significant extent, it might help in the control of credit,—

—not through qualitative discrimination but through quantitative restriction.

____________________

I agree entirely with Professor Ellis that the inflation absorbing effect does not begin to act early enough. The main point, however, is that the “absorption” affects only a trivial fraction of the newly created credits. In practice, therefore, there can be no question of stock exchange speculation depriving industry of all or even of a considerable part of the funds created by inflation. The problem might be put in this way: Assume that a credit expansion is taking place at the rate of 100 units of monetary media per unit of time. A large part of this new money passes through the stock exchange. In response to the rise in share prices wide circles of the public begin after some time to get interested in stock speculation. The resulting transactions “tie up” some money, let us say 5 or 10 units, so that in each unit of time, instead of 100 units, only 95 or 90 units of the newly created money flow to the industrial markets. Thus, given a constant rate of credit expansion, there would be a somewhat smaller rate of flow to industry. Those who are of the opinion that the rate of growth of industrial expansion should never decline, even if it could be kept up only by credit expansion, will, in this case, advocate still easier credit conditions. Those who are of the opinion that the credit expansion should in any case be checked (and the sooner the better), will not be worried by the (small) possibility that part of the inflation will be absorbed by speculation in the way described.

Incidentally, Professor Ellis sees the causes of possible “absorption” less in stock exchange transactions than in induced hoarding activity. We shall deal with this in Chapter VIII.