13. Anachronism of the Liquidity Preference Concept *

It is the contention of this chapter that the concept of liquidity preference is not applicable under present conditions. In part, it has always been superfluous; moreover, the aspects to which it was once appropriate have lost practical importance through institutional changes that have taken place in the last decade.

THE PRESENT CONCEPT OF LIQUIDITY PREFERENCE

As is well known, liquidity preference was conceived within the framework of the theory of interest. At some times it has been considered more important than at others.

a. In classical theory the liquidity preference concept, or something equivalent to it, is, generally speaking, not mentioned. The interest rate keeps the supply of money saved in balance with the demand for money to invest. The concept of hoarding or the possibility of hoarding plays no major role. The theory of interest was monistic, “pure.” ¹

b. However, it had always been observed that the credit market supply was at times influenced by demand for credit that did not originate in the normal desire for capital, i.e., purchasing power for productive purposes; that a special “money or cash demand” existed.

As a result of this observation, neoclassical theorists developed, in the so-called “loanable fund” theories of interest, “dualistic” theories of credit demand: money was demanded not only to be spent but also to be hoarded, and supplied not only because saved but also because dishoarded ²—or incidentally, of course, because created by banks. These theories thus achieved a synthesis of what can be called “purely monetary” and of “pure” theories of credit demand and supply.³

c. This dualistic theory prevailed in continental European and Anglo-Saxon literature until Keynes replaced it by a monistic, purely “money demand” theory.

According to Keynes’ liquidity preference theory, interest is paid for the “desire to hold wealth in the form of cash” and received as “the reward for parting with cash” against some instrument of saving.⁴ It is “the reward for not hoarding,” not “the reward for not spending.” ⁵ In other words, the owner of a savings account receives interest because he does not hoard, not because he does not spend money; and he loses interest because he hoards money, not because he spends it. The demand for cash comes from the desire to transform savings accounts into cash. It is denied that money is demanded in order to be spent, not to be hoarded. In short, the switching from savings accounts into cash for hoarding, not the withdrawal of savings accounts for spending, is considered the only possibility. At least on the surface, this is a “monistic” theory of cash demand for hoarding purposes.

Keynes’ liquidity preference theory has often been criticized with much acuteness. It has been demonstrated that a monistic theory, neglecting the influence of saving and borrowing for consumption or production purposes on the interest rate, must necessarily be one-sided and unrealistic.⁶

My own main objection is that his theory is not consistent even as a “hoarding theory of interest,” and does not explain what it purports to explain, namely, that interest is paid for keeping liquidity, and interest is received for parting with it.

1. Cash is defined not only as actual cash but may include “money-time deposits with banks and, occasionally, even such instruments as treasury bills, and it is as a rule co-extensive with bank deposits.” ⁷ As one can earn interest on all such investments, one receives interest for holding liquidity, not for parting with it.⁸ This interest may, it is true, be lower than interest paid on longer-term investments; but in times of money stringency short-term interest rates can be, and have been, higher than long-term; so that for holding liquidity, one not only receives interest, but even higher interest than illiquid investments would yield.

2. Keynes gives four motives for wanting cash: the Income motive, the Business motive, the Precautionary motive, and the Speculative motive.⁹ All are clearly taken from the arsenal of the classical “pure” interest theory: anyone who holds cash for the purposes mentioned obviously holds it with the intention of spending it, either at regular intervals or speculatively sooner or later than is usual.

If individuals obtain cash or checking accounts for these purposes they have to pay interest, not for hoarding the money but for keeping it to be spent sooner or later. So in reality, interest is paid for spending, not for hoarding money.¹⁰

3. The liquidity preference concept is so widened that it is supposed to be effective not only when somebody hoards or does not spend money because he wishes to retain cash, but also when he does not spend money for quite other reasons. It has thus become a negative rather than a positive concept. It suggests that hoarding, for instance because of low profits on an investment, is due to a special demand for liquidity, not to the simple fact that profits for the time being are expected to be lower than lost interest.

The general objection to Keynes’ theory, however, is that it is monistic, not dualistic. “Such loose phrases as that interest is not the reward of non-spending but the reward for not hoarding seem to argue a curious inhibition against visualizing more than two margins at once.” ¹¹

d. Since Keynes issued his General Theory, the neoclassical dualistic theory seems to have made a comeback—again, very generally speaking. It is recognized that money is wanted for spending as well as for hoarding. As a consequence of this dualistic approach, the interplay of the credit demand for productive purposes and for holding liquidity—which is considered to decline as interest rates rise—has a major role in contemporary writing on interest rates.

However, the concept of liquidity preference is much broader than the neoclassical concept of money demand. The demand is directed towards bank accounts and short-term investments as well as cash.

e. We shall try to prove that the concept of liquidity preference is superfluous and confusing when it goes beyond the neoclassical concept of money or cash demand. Money demand, too, has become an obsolete concept because of certain changes in currency systems. For all practical purposes, the classical monistic “pure” theory of interest is necessary and also adequate to explain interest rates in a modern economy.

THE CRITERION OF LIQUIDITY PREFERENCE

Within the Keynesian system the liquidity preference concept is introduced to explain why, under certain conditions, the supply of credit available for ordinary demand is curtailed by an extraordinary demand. This special demand is supposed to absorb a part of the available cash; therefore some other demand is left unsatisfied. Liquidity preference is, as has been correctly stated,¹² a sort of “death trap” for savings in that it raises (by withholding the hoarded money from the credit markets) the interest rate above the level that assures (within a given demand schedule for credits) a given level of purchasing power. In other words, liquidity preference is supposed to explain what before Keynes was simply called “deflationary pressure from the credit supply side.”

Deflationary pressure from the credit supply side is present—

1. If, ceteris paribus, effective demand, MV, declines, i.e., when there is deflation;

2. If the decline is due to the withholding of purchasing power from the credit markets by creditors who keep their funds in cash or relatively liquid rather than illiquid forms, thereby curtailing the supply of credit and raising interest rates above those that would otherwise prevail.

Let us now examine to what degree the liquidity preference concept, as generally used nowadays, fulfills these conditions and to what degree it belongs to quite different theoretical categories. The concept covers four kinds of liquidity preference, each of which coincides with a different situation in the money and capital markets: interest rates are low on short- and long-term, on first- and lower-grade investments; or they are low on short- and long-term investments but not on lower-grade investments; or they are low only on short-term money; or finally, they may be high on short-term money too.

LIQUIDITY PREFERENCE OF THE ENTREPRENEUR: LIQUIDITY PREFERENCE AS THE INVERSE OF INVESTMENT PREFERENCE

The liquidity preference of the entrepreneur is, directly at least, not at all due to a curtailment of the credit supply. If MV dwindles, it is because demand for credit is weak.

If an entrepreneur holds cash or checking accounts for speculative or precautionary reasons and does not spend them within a normal period, he undoubtedly exercises a deflationary influence on MV. However, what causes the money to become idle is not a special liquidity preference of the entrepreneur. It is the simple consequence of the fact that for the time being profits on productive investments are deemed smaller than interest received for holding bank accounts, the zero interest on cash, or incidentally, the interest that would have to be paid for borrowing money. When the entrepreneur borrows from the bank, the marginal efficiency of capital is acknowledged to be solely responsible for the credit demand schedule at various interest rate levels.¹³ But there is not the slightest difference between borrowing from a bank and from oneself. The liquidity preference is fully accounted for by the demand schedule for capital, which compares the utility of various amounts of capital with the utility of corresponding amounts of cash or bank accounts. To speak of a special liquidity preference as a motive for refraining from productive investments is double counting, and as illogical as it is to speak of a special hunger preference of somebody who does not want to eat. Liquidity preference is merely the inverse of the preference to invest. The accumulation of cash or of bank accounts is just an expression of a weakening demand for credit—a downward shifting of the credit demand curve. It is not caused by a special liquidity preference, nor does it cause deflation. It is the consequence and reflection of what used to be called self-deflation.

It could be argued that the entrepreneur who accumulates cash or bank accounts curtails the supply of money that could serve as a source of credit for others and thus exercises indirectly a deflationary pressure. We shall revert below to the question whether and under what conditions that might happen.

Liquidity preference of entrepreneurs is usually accompanied by low short- and long-term interest rates, high bond prices, and ample offering of bank credits. This combination appears at the very end of a cyclical depression when inventories have been liquidated under the impact of deflation. Banks have become liquid and are ready to grant new credit, whereas entrepreneurs are reluctant to apply for credit because they fear losses from further declining prices for their products. Credits are low and the economy remains in a state of underactivity.

To explain this situation, the concept of a “low marginal efficiency of capital,” or its expectation, is fully adequate, and the liquidity preference concept is entirely unnecessary.

LIQUIDITY PREFERENCE OF CREDITORS: LIQUIDITY PREFERENCE AS AN EXPRESSION OF THE RISK FACTOR

Just before the cyclical phase mentioned above is reached, short -and long-term interest rates are usually low, government and AAA bonds high, but venture capital, either directly or by means of the stock market and bank credits, is scarce; and what little is available is very costly. There is a margin between the “pure” interest rate, which is low, and the interest rate that comprises the risk premium, which is high. Though clearly arising because the flow of investible funds reaches only first risk investments and is prevented from reaching investments with higher risks, this situation too is often considered to be caused by liquidity preferences, this time of creditors or banks.

If banks are reluctant to grant new credits, it is because they fear their funds will be newly “frozen” and lost in case of enforced liquidation, not because they have “no money” or the money has disappeared in a “death trap.” They have all the money they need either in their safes or as reserves in the central banks. What they desire to retain is not cash—in fact they try to get rid of it whenever they see a half-way safe or profitable investment opportunity—but their own bank liquidity in the private economic sense of the word. In this limited sense, I myself, twenty-six years before Keynes, considered interest as the reward for parting with liquidity.¹⁴

Incidentally it should be noted that the stringency of bank credit at this juncture is caused not only by a curtailing of supply but also by an increase of demand for credit. A new demand for credit, called in German “Durchhaltekredite,” appears. In a cyclical crisis, goods become unsalable at the prevailing price level. Producers are at first reluctant to incur losses by liquidating their accumulating inventories at lower prices. Instead they allow their inventories to pile up. Lacking money to pay current production expenditures, they have to apply to the banks for additional credit. Although this credit demand gets its impulse from the fact that buyers withhold their purchasing power from the markets for fear of falling prices, the ensuing credit stringency is caused not only by a dwindling supply of bank credit but also by the increased credit demand for the purpose of postponing the liquidation of abnormally high inventories. There is a genuine new credit demand.

The behavior of creditors that creates the situation characterized by low interest rates for first-class investments and high interest rates for lower-grade investments, as just described, is really not due to a liquidity preference at all. The situation is caused by the blocking of the flow between the pure money and capital markets and the investment markets, not by a deflationary pressure from the money side. Therefore the liquidity preference concept has no raison d’être as a special category; what it tries to explain is entirely covered by the concept of the risk of the lender.¹⁵ It can also be interpreted as the lowering of the demand schedules of professional lenders who consider that only the interest rates at which they borrow themselves, not the risk premium, are covered by the high interest rates on the investment markets.

This margin between the pure and what has been called the gross interest rate, caused principally by the nonfunctioning of the banking system, undoubtedly tends to narrow. Such a nonfunctioning is not likely to be tolerated in the future. Organizations—many of which are already in existence—will grant the credit banks will be reluctant to give, or the governments will guarantee such credit so that the banks will no longer be concerned about their liquidity.¹⁶ It is a matter of opinion how far this development can be considered to have gone already.

LIQUIDITY PREFERENCE OF THE ARBITRAGEUR WHO ARBITRAGES BETWEEN MONEY AND CAPITAL MARKETS

Having considered the case where the pure interest rate is low on money as well as on capital markets, we now examine the case where short-term interests are low, but long-term interests, even yields on government bonds, are high.

Arising in a cyclical phase somewhat before the one previously described, this is the situation that plays such an important role in Keynes’ liquidity preference theory.¹⁷ He considers it a case of the liquidity preference of owners of bank accounts or of banks that are bullish on interest rates because they fear an aggravation of the crisis.¹⁸

Again there seems to be no special reason to retain a special concept of liquidity preference of the owner of bank accounts or cash. What really causes the margin between long- and short-term money is that the arbitrageurs between money and capital markets, who usually borrow from the banks on short-term and buy long-term bonds (the banks themselves can also be arbitrageurs in this sense), hesitate when they consider that they would lose more on the price of the bonds than they would gain from the interest rate margin. The banks in particular hesitate to use their money-creating power to buy long-term bonds. Therefore, what is called liquidity preference is again merely the inverse expression of a low investment preference of the professional investor in long-term securities. Incidentally, this margin, too, clearly tends to disappear so that the importance of the phenomenon in practice, and therefore also in theory, must be considered as decreasing. The reasons are two-fold:

First, the prevailing easy-money policy has greatly reduced the risk that short-term interest rates will be allowed to rise. Secondly, the power the Federal Reserve System now has to buy long-term government bonds—which it will use, if for no other reasons than fiscal—puts the long-term interest rate, formerly solely dependent on the arbitrage between money and capital markets, almost as directly under its control as the short-term interest rate.

LIQUIDITY PREFERENCE THAT CREATES HIGH SHORT-TERM INTEREST RATES: LIQUIDITY PREFERENCE IDENTICAL WITH CASH PREFERENCE

Undoubtedly at times the money markets were very tight and short-term rates very high. Then there was real deflationary pressure from the money side. We contend, however, that these situations cannot be explained by liquidity preference in the usual way if bank accounts and short-term investments as well as cash are considered objects of the preference. We must distinguish between liquidity preference for cash, in the narrowest sense, on the one hand, and for bank accounts, short-term investments, etc., on the other.

The liquidity preference that could influence the supply for money and exercise a deflationary pressure from the money side on the basic short-term money rates can only be a preference for cash, never for other investments. To prove this, we have only to examine the demand and supply for purchasing power in an economy without currency. The procedure is the one I followed in my first article on money market problems, quoted above.¹⁹

A. Liquidity Preference in a Money-free Economy. In an economy in which there can be no demand for or supply of cash, the demand for loans is obviously identical with the demand for and supply of sight deposits (checking accounts) in banks. As the ability of banks to create such accounts autonomously by granting credit furnishes the marginal supply of such credit, we can say that the supply of credit is dependent upon the ability of the banks to create credit through creating debtor and creditor accounts.

In such a situation money cannot be kept from the credit market because no money exists; there is no “death trap” for money. And as far as sight deposits are concerned, they are never withdrawn. The fact that they are “held” means that they have been loaned—through the bank as intermediary—to a debtor of the bank. The persons who hold them for liquidity reasons thereby satisfy their liquidity preference.²⁰ No hoarding can raise the interest on short-term money markets. The money might not spill over to capital markets, it is true, but liquidity preference can never create high interest rates on the money markets as long as bank accounts or money market instruments, not cash, are hoarded.

Again it could be argued that deflation would ensue if banks were unable or unwilling to grant new credit which would compensate a decreasing V through an increasing M; also that legal reserve requirements on the one hand and liquidity fears on the other could set a limit to such granting of new credit.

However, as far as liquidity fears (as described on p. 152) are concerned, they could never prevent the banks from buying first-class securities in any amount they wished, and thus from lowering the supply price of credit to any desired level. Moreover, reference to legal reserve requirements is clearly out of place in a “money-free” economy. If a certain percentage of bank accounts have to be covered by balances with central banks, holding the former amounts to holding, indirectly, the latter or the cash that can be freely obtained against the accounts.²¹ Therefore, if bank accounts for which a coverage is required are hoarded instead of spent, this is really the case of hoarding real cash discussed below under B.

For continental Europe, where reserve requirements were unknown and private banks were not dependent upon the central banks for credit expansion, as long as the public did not convert its deposits into cash, reference to reserve requirements would, of course, be entirely out of place. Even in countries such as the United States, where coverage requirements exist, they could be of practical importance only at the beginning of a depression when, as described above, credit demand increases. In the later stages of a depression, after bank balances have contracted, the excess of reserves is so large that every credit demanded can be granted.

B. Cash Preference in the Past. What has led and theoretically in the future could lead to a deflationary pressure from the credit supply side is merely a demand for cash—a cash preference, we may call it. It is the pre-Keynesian concept of money demand to which we thus return and which, in the opinion of this author, should never have been given up.

What are the reasons for such a money demand? Obviously it could never appear in a “money-free” economy, where the demands for purchasing power in the form of bank accounts and for loanable funds are always identical. Only when two kinds of purchasing power—bank deposits and cash—co-exist and are to a certain extent interchangeable can there be a demand for money outside of and not satisfied by bank deposits.

The customary assumption seems to be that to hold cash, instead of either holding a bank account or spending the cash, gives a certain pleasure or convenience that declines with the amounts involved. All sorts of supply and demand curves attempt to show the interaction of the demands for money to hoard and for money to spend at various levels of interest rates. However, this type of analysis, while logically perfect, cannot describe realistically how the demand for money affects the supply of credit. It fails also to clarify how far the liquidity preference concept, even in the narrowest sense of “money demand,” is still of actual importance at present and will be so in the future. To do this, we must recognize that there are several varieties of money demand and that they are governed on the supply as well as on the demand side by quite different laws. The following varieties must be distinguished: ²²

a. Extraordinary Money Demand.²³ The most outstanding examples of extraordinary money demand occurred during the famous bank crises of 1847, 1857, and 1866 in England. The public, fearing for the safety of deposits, sought to turn them into legal tender. Because there was not enough money to transform the billions of bank deposits into real money, a terrific deflationary process ensued. Each crisis led to the suspension of the Peel Act. As soon as the Bank of England was permitted to issue as many bills as it wanted, the demand for cash subsided. Similar crises have occurred in the United States, the most recent in 1933.

During such bank crises the money demand curve runs very high, almost horizontally, i.e., it is virtually independent of the interest rate. The public is prepared to pay very high interest rates for practically unlimited amounts. The supply curve, on the other hand, after a certain point runs nearly vertically. As more money is not available at higher rates, there is a real and terrific deflationary pressure from the money side.

b. A special case of the “extraordinary money demand” is the extraordinary demand for gold or foreign exchange,²⁴ which arises when not only “bank money” but also legal tender is suspect; so that a flight from the domestic currency ensues. The demand curve runs from the right to the left at very high levels, but not horizontally. The higher the interest rates, the more they are considered to offset the losses from currency depreciation. The supply curve, too, moves up to higher levels.

Under modern conditions, an extraordinary gold or foreign exchange demand usually leads to either the devaluation of the currency or a system of currency restrictions—witness the money disturbances of the ’thirties. Confidence in money at the former parity is not restored because readjustments of the basic disequilibria are not to be expected in a world of rigid costs.

c. Ordinary Money Demand.²⁵ We distinguish five varieties:

1. The demand arising because people are paying cash rather than by check.

2. The demand arising from a concentration of needs for cash at certain times.

In addition to these two, which, for obvious reasons, I have called technical,²⁶ there are three cases of nontechnical, “economic,” money demand:

3. The demand arising when bank accounts grow with prices and/or output. Obviously, the percentage that has to be paid in cash must keep pace with the growth of bank accounts.

4. The demand arising from the higher turnover of bank accounts. More money is demanded because the cash usually brought to the banks in between payments has to remain longer outside the banks, thereby curtailing the money supply.

5. The demand arising when banks grant more credit to compensate for a decreasing V—as described above. It is partly satisfied by the influx of money to the banks when the velocity of turnover of bank accounts slows down and people keep their cash reserves in banks. However, to the extent that people keep their reserves at home, thus really hoarding, the banks need new cash to grant new credit, of which a certain percentage must be paid out in real cash. If money demands 3 and 4 can be called inflationary, the money demand described here can be called antideflationary. This demand for additional money arises because people “hoard” rather than “save.” It is the only case where the “death trap” really works and the hoarding theory of interest is justified.

The main reason for differentiating these five varieties of money demand is that the satisfaction of each has quite different results. Satisfying money demands 1 and 2 is economically neutral, leading to neither credit expansion nor money inflation. Satisfying money demands 3 and 4 leads to inflation by enabling either credits to expand or the velocity of purchasing power to accelerate. By meeting or not meeting this inflationary money demand, monetary authorities can endeavor to stabilize the price level. Satisfying money demand 5 leads to an increase of credit, but not to inflation, because the increase of M offsets a previous decrease of V. Satisfying this kind of antideflationary money demand also is economically neutral.

It has become customary to consider all sorts of money demand as inversely dependent upon interest rates. The higher the interest rates, it is assumed, the less the propensity to pay cash and to hoard instead of to save, because the “quasi profit” of cash is seen to decline gradually.²⁷ However, the effect of the interest rate has been in the past and certainly is for the present and future very much overrated. To be realistic, one must consider the demand curve for cash as running at a high level almost horizontally from the left to the right at first, then very soon almost vertically downwards. In the past when money was offered only at relatively high rates, the horizontal branch may perhaps have been cut by the supply curve, but today when money is offered at very low rates, only the vertical part, which reflects very inelastic demand, is cut by the supply curve. Therefore, even if the supply curve moves further down—in other words, if interest rates decline further—no more money is actually used.

Without doubt we are now experiencing the development of a new sort of technical money demand: black market operations and the possibility of tax evasion have tremendously increased the desire to pay large amounts in cash instead of by check. The demand curve for money therefore runs a long distance almost horizontally, i.e., for the additional demand not lower but nearly the same interest rates are offered. This demand, however, is satisfied at low interest rates because of the prevailing elasticity of the money supply.

EFFECTS OF MONEY DEMAND ON INTEREST RATES IN THE FUTURE

From the above I think it is evident that the future influence of liquidity preference, even in the denatured form of cash preference, on interest rates will be very limited, if it does not disappear entirely. Indeed, its influence may well be so small that theory could afford to ignore the existence of liquidity or cash preference and revert to the classical monistic “pure” interest theories.

a. Extraordinary Money Demand. There is no doubt that if a bank crisis occurs in the future the central banks of every country will immediately supply the banking system with money to meet the extraordinary money demand.

b. Extraordinary Demand for Gold and Foreign Exchange. Despite some illusions nourished during the Bretton Woods discussions, but fading more and more, interest rates will not be allowed to rise because of an external drain on foreign exchange, especially not if it is caused by capital flight. Devaluation and currency restrictions will always be preferred to the traditional play of the gold standard. Deflations from so-called external drains will not be tolerated, and the so-called external discount policy will be a thing of the past.

c. Technical Money Demand. Technical money demand, likewise, will no longer be able to bring about deflation. People may use or hoard as much money as they want. The supply will always be adequate. It is inconceivable that money scarcity due to changing habits of payment will be tolerated in the future.

d. Economic Money Demand. We must distinguish between inflationary money demands 3 and 4 and the antideflationary money demand 5.

1. The antideflationary money demand represents, as shown above, the real and only “death trap” situation. However, the “death trap” cannot cause money stringency in the future. Central banks will always offer enough cash to the banking system to enable “compensating” new credit to be created. As has been repeatedly stated,²⁸ any demand for money arising from increased liquidity preference can easily be met by issuing new money. It not only can be, but actually will be. And if this is true for a cyclical increase in liquidity preference, it must be even more true for a secular increase,²⁹ which Keynes seems to have had in mind in the first place. Governmental liquidity production will always outrun the liquidity preference of the public. There is no longer room for the assumption of the inverse dependence of the size of hoards on interest rates that Keynesians emphasize so much. Their assumption ignores the disappearance of the gold standard and its replacement by an entirely elastic currency throughout the world. The quantity of money issued is no longer governed by considerations of legal limitations; rather these legal limitations are always modified as soon as other considerations seem to suggest increases in the quantity of money.

2. By controlling the inflationary money demand, central banks can, at least theoretically, control inflation and practice the so-called internal discount policy.

From all we have said it is clear that this part of “economic money demand” is identical with the “pure” credit demand. Cash, in a certain proportion to noncash forms of purchasing power, is demanded if and when funds for consumption or production purposes are wanted.

The problem of whether and at what interest rate central banks will be prepared to satisfy the economic money demand is the same as the problem of whether the easy-money policy which stabilizes interest rates for all practical purposes will be maintained. The relative merits of interest rate flexibility and stability are still being discussed. If any general statement can be ventured at this moment, it is that the overwhelming opinion seems inclined to sacrifice interest rate flexibility for interest rate stability. The internal discount policy, designed to combat inflation of domestic prices, will likewise hardly be applied in the future. Price fixing, rationing, and other measures will be preferred to the natural means, namely, a restrictive discount policy.³⁰ Among the many reasons the chief one, aside from fiscal considerations, is that easy money is thought to incite and perpetuate investments. In the writer’s opinion, it does so only under very special conditions.

SOME CONSEQUENCES OF CHANGES IN MONEY DEMAND

Keynes’ General Theory of Employment as presented in his General Theory, Chapter 18, rests essentially upon the choice of the independent and dependent variables of his system. Among his independent variables the most important is the rate of interest,³¹ which is dependent upon the state of liquidity preference and on the quantity of money;³² his dependent variables are the volume of employment and national income.³³ We have tried to show that interest rates are stabilized, liquidity preferences frustrated, and the quantity of money always created in accordance with prices and output which, in turn, are dependent upon quite different independent variables. Consequently, Keynes’ choice of independent variables seems so unrealistic that his Employment Theory is deprived of its usefulness as a tool of analysis.

Even more important than these theoretical effects of the changes in money demand are the implifications for the economy itself:

Deflationary pressure from the money side is not likely in the future. Neither the extraordinary money nor the extraordinary foreign exchange demand of money crises will be allowed to repeat themselves. Nor will deflations from an increase in “technical money” demand or from an “antideflationary” money demand, i.e., from liquidity preference proper, be allowed—undoubtedly a favorable development.

On the other hand, the adoption of an apparently permanent easy money policy has not only caused the disappearance of a “death trap” for credits, but has eliminated the possibility of stabilizing business activity through interest rate manipulation. Since very low interest rates cannot be reduced further, the stabilizing effect of interest rate reduction in a depression is destroyed; and if a ceiling is put upon interest rates, the stabilizing effect of high interest rates during a boom cannot be counted upon either.

This means that our modern credit system, while protected against autonomous deflations from the credit supply side, has become very sensitive to changes on the demand side. Low demand and deflationary tendencies and high demand and inflationary tendencies will no longer be mitigated by compensating interest rate movements.³⁴

If our analysis is correct, the pattern of future business cycles will differ greatly from that of past. Without trying to predict this future pattern, the following brief remarks may be ventured:

1. The beginning of a depression will no longer be characterized by high interest rates. The specific features of a monetary and banking crisis will be absent. The market for long-term government bonds may remain strong. There need not be anything like a general crisis of credit and confidence. Such symptoms were virtually absent already from the recession of 1938.

The unnecessary and exaggerated liquidations under the impact of deflations will not be repeated. On the other hand, too high inventories built up during the boom will not be immediately liquidated; the liquidation process will be protracted.

2. The depression, while less severe in the absence of real deflationary pressure, may be prolonged because the stimulating effect of declining interest rates is absent.

3. During prosperity the rising demand for credit will no longer be counteracted by the increasing cost of the supply. Easy money will postpone the end of the boom, but by no means indefinitely. If not curtailed by other changes detrimental to investment, e.g., too high wages, taxes, etc., overspeculation will go further than usual.

The paradoxical result of all this is that the business cycle will not be more stabilized in the future than it has in the past, in spite of all endeavors. There will be no more deflations from the credit supply side; however, this advantage will be offset by the stabilization of interest rates, which has destroyed a potentially strong anti-cyclical measure.³⁵

* Appeared first in Kyklos, International Review for Social Sciences, 1947, 3.

¹Cf. Gottfried Haberler, Prosperity and Depression, 1941, p. 195.

²Ibid., p. 196.

³ My own work on the subject goes back to 1918, when I published an article entitled “Der Gegenstand des Geld- und Kapitalmarktes in der modernen Wirtschaft: ein Beitrag zur Theorie des Bankgeschäfts,” in Archiv für Sozialwissenschaft und Sozialpolitik, 1925, vol. 51, p. 289; and “Zur Frage des sogenannten Vertrauens in die Währung,” in Archiv für Sozialwissenschaft und Sozialpolitik, 1925, vol. 52, p. 289. My “dualistic” theory was finally formulated in my Volkswirtschaftliche Theorie des Bankkredits, 3rd edition, 1930, p. 53 ff.

⁴ J. M. Keynes, The General Theory of Employment, Interest, and Money, New York, 1936, p. 167.

⁵Ibid., p. 174.

⁶Cf. Jacob Viner, “Mr. Keynes and the Causes of Unemployment,” in Quarterly Journal of Economics, 1936-37, p. 152. D. H. Robertson, “Alternative Theories of the Rate of Interest,” in Economic Journal, vol. 47, 1937, p. 431, and “Mr. Keynes and the Rate of Interest,” in Essays in Monetary Theory, 1940, p. 1 ff. For a good summarization of the criticism of Keynes’ liquidity preference theory see George Halm, Monetary Theory, 1942, pp. 72-73 and 220-23.

⁷ Keynes, op. cit., p. 167, footnote 1.

⁸Cf. Halm, op. cit., p. 221.

⁹ Keynes, op. cit., pp. 195-96.

¹⁰Cf. Halm, op. cit., p. 72.

¹¹ D. H. Robertson, loc. cit., Economic Journal.

¹² Robertson, “Some Notes on Mr. Keynes’ General Theory of Employment,” in Quarterly Journal of Economics, vol. 5, 1937, p. 173.

¹³ Keynes, op. cit., p. 135.

¹⁴ Hahn, Volkswirtschaftliche Theorie des Bankkredits, 1st edition, p. 102: “If the amount of credits given by the banks is dependent on their private liquidity, the interest rate, i.e., the price that has to be paid for the credit, is merely the reward for the loss of liquidity caused by the granting of the credit.” Keynes uses almost exactly the same words (General Theory, p. 167): “The mere definition of the rate of interest tells us in so many words that the rate of interest is the reward for parting with liquidity for a specified period.” Keynes stressed later the importance of the “liquidity of banks” (Economic Journal, vol. 47, 1937, p. 660).

¹⁵ Mentioned by Keynes, General Theory, p. 144.

¹⁶ This is why I cannot place the same emphasis on the risk factor as H. C. Wallich does in his interesting article, “Changing Significance of the Interest Rate,” American Economic Review, 1946, p. 76, where, as far as pure interest rates are concerned, conclusions similar to mine are drawn.

¹⁷ Keynes, op. cit., pp. 168 ff. and pp. 202 ff.

¹⁸ That rising interest rates during a boom have nothing to do with liquidity preference in any sense but are due to higher profit expectations has been correctly observed by Jacob Viner, loc. cit.

¹⁹ Mr. Hawtrey followed the same procedure in Chapter I, “Credit Without Money,” in his Currency and Credit, 1919. Incidentally, the procedure leads inevitably to the statement that investments and savings are necessarily equal. In a “world with only credit,” as soon as a credit has been granted and the credited amount spent for productive purposes, new deposit accounts are created. I formulated the theorem of the equality of investment and saving as early as 1920 with the statement that savings are either always invested or nonexistent (Volkswirtschaftliche Theorie des Bankkredits, 1st ed., p. 153). Keynes expressed the same idea in his General Theory: “No one can save without acquiring an asset, whether it be cash or a debt or capital goods” (p. 81), and “in the new situation someone does choose to hold the additional money” (p. 83).

²⁰ As clearly recognized by Viner, loc. cit., p. 155.

²¹ Accounts with the Reichsbank were therefore called “Giralgeld” by German writers and considered as money rather than as bank accounts.

²² The distinctions are those made in my Volkswirtschaftliche Theorie des Bankkredits, 3d ed., p. 54 ff.

²³Ibid., p. 66.

²⁴Ibid., p. 70.

²⁵Ibid., p. 74.

²⁶Ibid., p. 78 ff.

²⁷ I contrasted the “Quasi-Zinsgenuss” of the cash-holder to the receipt of interest by the holder of bank accounts in my article cited above (Archiv für Sozialwissenschaft, 1925, Vol. 52, p. 304).

²⁸Cf. Howard S. Ellis, “Monetary Policy and Investment,” American Economic Review, Vol. XXX, No. 1, March 1940, Supplement, p. 29; and Gottfried Haberler, “The Interest Rate and Capital Formation,” Capital Formation and its Elements, National Industrial Conference Board, New York, 1939, pp. 126-27.

²⁹ As quite correctly stated by Ellis, loc. cit., and Jakob Viner, loc. cit., pp. 152 to 160.

³⁰ See Chapter 7.

³¹ Keynes, op. cit., p. 245.

³²Ibid., p. 246.

³³Ibid., p. 245.

³⁴ To what extent the elasticity of money supply is now considered a matter of course is shown by the introduction of the so-called Acceleration Principle as an explanation of business cycles. The requirement of large capital expenditures within a relatively short period, for an increase in the production of consumer goods, has always been recognized. It is one of the cornerstones of Spiethoff’s Overproduction Theory. Later monetary business-cycle theorists did well to reject Spiethoff’s theory. They argued that what led to the over-proportional capital expenditure was not so much the extraordinary demand as a too low discount rate which failed to curtail the credit supply sufficiently to guarantee an equal distribution of capital expenditures over time. If the old Spiethoff theory can today be presented in a new form without arousing objections, it is only because interest rates are practically stabilized and the credit supply perfectly elastic, so that a concentrated increase of demand causes a concentrated increase of expenditures for the production of capital goods.

³⁵ See Chapter 7.