XXVI. Factors Affecting the Rate of Interest in the Short Run

Although a full discussion of the monetary problems to which the existence of the" real" rate of interest gives rise lies outside the scope of the present book, it would 353 354 The Money Rate of Interest PT. IV hardly be appropriate to leave our subject without giving a somewhat more definite indication of how the rate of interest we have been discussing and the money rate of interest are related. At this point we can Limited scope of present discussion of give no more than an outline of the answers nloney rate 01 Interest t th . bl A f 11 d· ·on o e maIn pro ems. u ISCUSSI of the whole complex of problems involved would require another book of about the same size as this one - even supposing that, in the present state of our knowledge, any such systematic and exhaustive treatment of these as yet imperfectly explored problems could be attempted successfully. As has been explained earlier inthis volume, its task is to lay the foundations for the treatment of these problems, not to discuss them in any detail. And we shall confine ourselves in this final Part to the task of showing how these theoretical foundations can be used for the elucidation of certain salient points in the discussion of these more complex problems. We shall not attempt to follow all the possible complications or to explore the consequences of the different possible assumptions with any microscopic accuracy.

For the purposes of this discussion it will be necessary to alter the terminology used in the earlier Parts of this book. As the traditional terminology which we have Use of the term" rate followed up to this point clearly creates 01 Interest" the danger of some confusion if it is retained in the discussion of monetary problems, it will probably be best if, for the purposes of this final Part, we reserve the term "rate of interest" exclusively for the money rate of interest, that is, the price paid for loans of money, and describe the real rate of return as the rate of profit. Our main problem, then, is to explain how the existence of a system of rates of profit, which in terms of anyone commodity will tend to correspond to a uniform time rate,! 1 For the exact meaning of the concept of a uniform time rate of return (measured in terms of anyone commodity) compare above, p. 167.

CR. XXVI Short-run Influences 355 will affect the terms on which money will be lent and borrowed. There can be no doubt that the existence of such a rate of profit on investments is the main source of the demand for loans of money, Relation between the since command over present money is com-rate 01 profit and the rate 01 Interest In mand over present resources which can be equilibrium turned into future commodities at a profit. And there can also be little doubt that the existence of such a rate of profit is at least one of the reasons why people who might themselves employ the money profitably, will not be willing to lend it without special remuneration, and that therefore the rate of profit will also affect the supply of loanable money funds. If the rate of money expendi ture always remained constant, so that the moneyexpendi ture during any period was always equal to the amount of money spent during the preceding period of equal length, and if consequently we could assume that all the money received during any period would be respent, after an (on the average) constant interval, either on consumers' goods or on some income-bearing assets, it would clearly be justifiable to assume that the rate of interest would be directly determined by the rate of profit. To every increase in the demand for one com modity (or other type of asset) there would correspond an exactly equal decrease in the demand for another kind of commodity. That is, prices would be determined in th~ same way as in the imaginary barter economy. And, in particular, the demand for investment goods would be exactly equal to that part of their assets which people did not want to have in the form of consumers' goods.

The supply of funds not spent on consumers' goods would become equal to the demand for such funds at a rate of interest corresponding to the rate of profit as deter mined by the given prices. There would be differences between the rates of profit people exp.ected to earn in their own businesses and the rates of interest at which they would be willing to lend and to borrow, correspond356 The Money Rate of Interest PT. IV ing to the different degrees of risk. But the net rate of interest would tend to be equal to the net rate of profit. And the relative prices of the various types of goods, and therefore the price differences, would depend solely on the relation of the proportions in which people distributed their money expenditure between consumers' goods and capital goods to the proportions in which these two types of goods were available. While this would undoubtedly be the position once equilibrium had been established, it is one of the oldest facts known to economic theory that changes in the quantity of money, or changes in its Influence of monetary changes on rate of "velocity of circulation" (or the " demand Interest for money"), will deflect the rate of interest from this equilibrium position and may keep it for considerable periods above or below the figure deter mined by the real factors. This fact has scarcely ever been denied by economists, and since the time of Richard Cantillon and David Hume 1 it has been the subject of theoretical analysis which has been further developed in more recent times, particularly by Knut Wicksell and his followers.2 But it has also given rise to a recurrent scientific fashion, from John Law down to L. A. Hahn 3 and J. M.

Keynes, of regarding the rate of interest as being solely de pendent on the quantity of money and the varying desires of people to keep certain balances of money in hand. 1 Cf. R. Cantillon, Essai sur la nature du commerce en general (1754), Part III, chaps. 7 and 8; and D. Hume, Essays MOTal, Political,and Literary (1752), Part II, Essay IV, " On Interest ". 2 In view of the apparently widespread impression that the influence of liquidity considerations on the rate of interest is a new discovery, the present author may perhaps be excused for pointing out that more than ten years ago he described cyclical fluctuations as largely due to the fact that the rate of interest is in the short run "determined by considerations oj banking liquidity" (Geldtheorie und Konjunkturtheorie, Vienna, 1929, p. 103; English edition, Monetary Theory and the Trade Cycle, London, 1933, p. 180). a See L. A. Hahn, Volk8Wirtschajtliche TheOTie des Bankkredits, Tiibingen, 1920, pp. 102 et seq., chapter headed" Der Zins als Preis des Liquiditatsverlustes."

CR. XXVI Short-run I njluences 357 We are here not primarily concerned with the transitory or purely dynamic effects of monetary changes on the rate of interest. But if it is true - as we must assume in the light of all evidence - that changes in the quantity of money affect the rate of interest, there must exist, even in equilibrium conditions, some relationship between the quantity of money people want to hold and the rate of interest. It is this relationship which we must first try to elucidate. Now, as has been pointed out in an earlier chapter,! the desire of people to hold money cannot readily be fitted into the rigid definition of equilibrium we have used up to this point. At least, in an Extension of concepl economy in which people were absolutely of equilibrium used certain about the future, there would be no need to hold any money beyond the comparatively small quantities necessitated by the discontinuity of transactions and the inconvenience and cost of investing such small amounts for very short periods. But the assumption of certainty about the more distant future, although we have so far based our argument on it, is not really essential for our concept of equilibrium. The plans of the various indi viduals may be compatible with the extent to which they are definite,2 and yet the individuals may at the same time be uncertain about what will happen after a certain date and may wish to keep some general reserve against whatever may happen in that more uncertain future. In this way our system can be made to include the desire of the individuals to hold money as a general reserve of command over resources.

It is clear that to the individual the holding of money 1 Cf. above, Chapter III. 2 The interesting problem of how far, despite the fact that the plans of the different individuals are somewhat vague, the "law of large numbers" may yet create sufficient regularity so that the vagaries of the individual decisions will not disappoint expectations, cannot be considered here.

358 The Money Rate of Interest PT. IV is one form of holding his asSets 1 and will compete with other forms of investment for the resources at his com mand. Although holding money yields no direct return, To the Individual the it may, by enabling the holder to take holding of money Is d t f C •• b one form of invest-a van age 0 unloreseen opportunities, e ment as much a means of reaping a return as any factor of production. And changes in the relative attractiveness of holding money or holding other resources will induce him to keep at different times different pro portions of his total assets in the one form or the other. Just as his endeavour to maximise his income will make him distribute his resources between the various forms of investment in the narrower sense in such a way as to equalise their returns, so he will also distribute his assets between investment in goods, in money loans, and in cash balances in such a way as to equalise the· advantage he expects to derive from these kinds of assets. If we assume, as we shall do to begin with, that only money is regarded as really liquid, and that all investments in goods and loans of money are considered equally illiquid, the equilibrium position between investments in goods and investments in loans of money will be reached when the net returns, i.e. the expected physical returns less compensation for risk and incident trouble or effort, are equal. The return from the holding of cash, being by its nature not so much an expectation of a definite return as an expectation of various uncertain possibilities, is less easily measured. We might perhaps speak of a mean return expected to be derived from the holding of a certain amount of money. But it is probably more convenient not to concentrate on this somewhat intangible return, but to relate the quantity of money a person is willing to hold to the amount of profit or interest which he could 1 Henceforth we shall use the term" assets" whenever we want to describe the aggregate of real capital, money, and securities, which from the point of view of any individual would be regarded as his " capital".

CR. XXVI Short-run I njluences 359 expect to earn if he invested that money now in goods or loans, and which he consequently sacrifices in order to keep himself in a position to take advantage of more uncertain possibilities. Any change in the relative attractiveness of holding money and holding investments respectively and any change in the supply of money and investments is there fore likely to change the way in which any Changes In Ihe dlstrl person will distribute his assets between buUon 01 assets will ailed Ihe rale of In these two outlets. It is not difficult to see teresl and the rale 01 that any tendency toward such a change in profll the distribution of assets is bound to affect the rate of interest and the rate of profit. And it follows that changes in these rates may occur even when the factors which we have so far treated as their sole determinants, i.e. the profitability of investment and the willingness to save, remain unchanged, and that the affect of any changes in these latter factors may be modified by a new element, the changes in the demand for the different kinds of assets, to which they may give rise.

In a general manner this effect of "liquidity preference" and the quantity of money on the rate of interest may be described by saying that the rate of interest must be such that people in general will be induced to The shorl-run deter keep as liquidity reserves just that part mlnallon 01 the rate 01 Interest: ,assump of the existing amount of money which is lions on which connot required to transact current business. sldered It is undoubtedly true that in this sense the quantity of money and liquidity preference will influence the rate of interest. However, this is very far from saying, as Mr. Keynes and his school do, that, even in the short run, the rate of interest is determined solely by the quantity of money and people's liquidity preferences, and still less that in the long run the rate of interest is primarily deter mined by these monetary factors. In the present chapter we shall be concerned merely with the effects which these monetary factors will have 360 The Money Rate of Interest PT. IV on the determination of the rate of interest in the very short run, postponing the discussion of the slower reper cussions of any change till the next chapter. This means in particular that we shall here consider only the tem porary equilibrium which will be established after a change has taken place in the market for money loans, and before the consequent changes in income and final demand have had time to affect expected returns. In order to simplify the argument in this first stage, we shall assume that there is only one homogeneous kind of money in existence, the quantity of which is fixed, and which is clearly demarcated in respect to its liquidity from all other assets, whether money loans or real assets, these other assets being regarded for present purposes as all equally illiquid. The most important consequence of this assumption is that in the present context we can disregard any differences between the different rates of return on funds that are in any sense invested, and particularly between the rate of interest and the rate of profit. So we can confine our selves to the relation between the purely psychical return from holding money and all other physical returns from investments of every kind.

We may begin by considering the argument which is at the back of the assertion that, at least in the short run, the rate of interest is determined solely by the quantity of money and liquidity prefer-Cause of erroneous belief that rate of ence. It can be shown without great Interest Is determined difficulty that this view is due to the solely by quantity of money and IIquld-treatment of one source of the demand Ity preference for money as if it were the sole determinant of its price, an error which is rather similar to the older belief that, since the industrial demand for gold has some influence on the value of gold, the value of monetary gold depends solely on its industrial uses. The case of the relationship between the demand for money, liquidity preference, and the rate of interest, appears only superficially more plausible because it is, CR. XXVI Short-run Influences 361 of course, true that the whole demand for money is derived from a desire for holding money. But not all the desire to hold money is due to liquidity preference, nor can it be assumed that the demand for money due to other circumstances can in the short run be considered as constant. Only if one or the other of these two con ditions were satisfied could it be said that the price of money depended solely on liquidity preference and the quantity of money in existence. In fact, of course, we hold money not only because we do not know what to do with it, but also, and in normal times probably to a much greater extent, because we intend to use it for particular purposes some time later and cannot con veniently invest it in the meantime. And while a rise in the expected rate of return on investments will make it relatively less attractive to hold money merely in the hope that it will prove more useful at some uncertain later date, it will at the same time increase the amount of money that will be needed to transact the business promising any given rate of return.

The misleading impression that the rate of interest is determined solely by the quantity of money and liquidity preference is based on the wrong suggestion, implied in this type of analysis, that the demand for The Inlluence of pro money is dependent solely on liquidity ductlvlty concealed In .. liquidity prefereu.ce preference. But the description of the function" demand for money in terms of a curve or function, which is called a liquidity preference curve or function, simply means that under this name all sorts of influences, in cluding in particular the productivity of investment, have been lumped together. It is, of course, always possible so to define the terms used in an argument as to make the conclusions purely analytibal and necessarily true. And this is exactly what is being done when liquidity prefer ence is so defined as to include all factors which determine the demand for money, and it is then concluded that the price of money loans depends exclusively on the quantity 362 The Money Rate of Interest PT. IV of money and liquidity preference. But in this form the statement amounts to no more than saying that the rate of interest depends on the demand for and sup'ply of money without telling us anything as to which factor on the demand side is of most importance.

In order to show that in fact the liquidity preference curve cannot be regarded as independent of the pro ductivity of investment but merely conceals or rather includes the productivity element, and consequently that the demonstration that the rate of interest is completely determined by this curve and the quantity of money does not prove that it is independent of the productivity of investment, we need ask only one question: Are the amounts which people are assumed to be willing to hold for reasons of liquidity at any given rate of interest supposed to be independent of the amounts which th~y can invest at that rate 1 Only if this question could reasonably be answered in the affirmative could the liquidity preference curve be regarded as independent df the productivity of investment. But even if this were the case, it could surely apply only to the amounts of money held as liquidity reserves and would therefore not enable us to derive the rate of interest from the total supply of money. If, however, as is almost certainly always the case, the answer to our question is in the negative, that is, if the amount of money people are willing to hold as liquidity reserves depends inter alia on ' how much they can invest at a given rate of return, this means that the whole productivity element has been smuggled into the so-called liquidity preference curve.

In this way the assertion that the rate of interest depends solely on liquidity preference and not on the productivity of investment is deprived of all foundation. 1 1 There has probably been some confusion with the idea that under perfect competition the investment demand schedule which any individual faces must be horizontal, that is that the amount which any individual can invest at a given rate of interest is unlimited. But, OK. XXVI Short-run Influences 363 The real position can be illustrated by slightly modi fying a diagram used by Professor Hicks in this con nection. 1 It is based on the assumption that the quantity of money is fixed, and it has two curves, Diagrammatic lIlus one showing the amounts of money people tratlon of relation between productivity will be willing to spend during any given and liquidity preferperiod at various rates of interest out of ence their given cash holdings, and the other showing the rates i o M FIG. 27 Nm of return which people expect to get on given amounts of expenditure. Thus in Fig. 27 the curve marked a shows that with a rise in the rate of interest (measured along the ordinate Oi) people will be willing to spend increasing amounts out of their given money holdings on investments (the amounts spent are measured along the abscissa Om),while the curve marked b shows that as this expenditure increases the expected rate of return falls. Since for the moment we are concerned merely quite apart, from the fact that perfect competition merely requires that no person counts on his action affecting prices, we have to deal here, not with the investment opportunities open to an individual, but with the investment demand schedule of society as a whole.

1 Cf. J. R. Hicks, 1937, p. 153.

364 The Money Rate of Interest PT. IV with the very short-term effects, this expenditure induced by changes in the expected returns will refer only to direct investments and will not include the further changes in expenditure on the part of the people whose receipts are increased by the investments. In this respect our diagram differs from the similar diagram used by Professor Hicks, who includes all the indirect effects on income. Our curve represents simply what Mr. Keynes calls the investment demand schedule, or the schedule of the marginal efficiency of capital, at the moment concerned. This method of representation means that instead of asking, as Mr. Keynes does, what quantity of money people will want to hold at various rates of interest and profit and with a given money income, we ask what amounts of money people will be willing to spend on investment with given money balances but at various expected rates of return. This means that we treat income as a dependent variable.

It follows from the definition of our curves that the rate of interest and profit will at any moment be fixed at the point of intersection of the two curves (the point P in the diagram). Suppose now that the investment demand schedule (our curve b) is raised, say by an invention. If cash balances were rigidly fixed so that the increase of expected returns would not induce people to release any money out of their balances (i.e. if the a-curve were a vertical line), the rate of interest would rise by the full amount of the rise in the investment demand schedule. Or, in other words, if the demand for money were perfectly inelastic with respect to the rate of interest, the rate of interest would depend solely on the productivity of investment (and the rate of saving, which, however, for our present short-term analysis we can treat as constant) and would closely follow any change in the productivity of investment. This is, of course, the case mentioned at the beginning of this chapter and the CH. XXVI Short-run Influences 365 one that was traditionally discussed in the pure (as dis tinguished from the monetary) theory of interest.

If, however, as is more likely, a rise in the expected rate of return will induce people to release some money from their cash balances (so that our a-curve slopes upwards to the right), a rise of the b-curve will not raise the point of intersection of the two curves or therefore the rate of interest by the full amount by which the b-curve has risen, but only by somewhat less; how much less, will depend on the slope of the a-curve. This means that the release of money from the liquidity reserves will increase the supply of funds at the same time as the demand for funds is increased, and the rate of interest will therefore rise less than if the supply were fixed. We might even theoretically conceive of an extreme case where within certain limits the desire to hold cash is perfectly elastic (i.e. our a-curve horizontal), so that in consequence of a rise of the b-curve, just enough cash will be released from idle balances to keep the rate of interest at its former level. In this case it might indeed be said that the rate of interest was determined solely by liquidity preference, i.e. the desire to hold money. For so long as our b-curve intersects the a-curve in its horizontal part, shifts of the b-curve will in the short run have-no influence on the height of the rate of interest.

It is instructive to consider somewhat more closely the conditions under which this may be true. This will show how extremely limited an application this theoretically possible case has. A hori zontal a-curve, as we have seen, would mean that people would in all circum stances invest just as much as could be invested at a fixed rate of return, no matter Conditions under which Uquldlly pre ference CQuld be re garded as sole short run determinant 01 rale of Interest how large the amounts were. The only condition under which this would appear at all likely is that the rate of return should already have fallen so low as only just to 366 The Money Rate of Interest !'T. IV compensate for the extra risk of lending or investing in real assets compared with holding money; In such a situation people would indeed invest just as much as they could invest at that minimum rate, and would hoard all the rest. But since we know that people actually do lend even at a fraction of one per cent, this minimum is evidently very low. All that the contention would there fore appear to imply is that there is some minimum figure for the rate of return below which it would never fall, and that if that figure has been reached all changes in the amount of investment will be financed by exactly equal changes in idle money balances. But even this would be strictly true in actual life only if we could regard the holding of money, as we do here, as subject to no risk either of loss or of depreciation. In fact we do know that this is not so and that in some circumstances people will even pay something for having their money kept in some form other than cash (or even bank balances), i.e.

that they will sometimes prefer to invest even at a negative rate of return. It seems therefore that we must assume that the amounts of money people are willing to hold will decrease with every rise in the expected rate of return, from zero or even below zero upwards, and that therefore our a-curve will throughout be upward-sloping in greater or lesser degree. We can therefore dismiss from our mind the case of an a-curve which is absolutely horizontal even in parts, and may confine our attention to the case where it is more pr~bable Ibape of or less upward-sloping. It still remains a-elllYe probable, however, that with a very low rate of return its slope will be slight. But it will clearly rise with rising rates of return, since the greater the reduction which has already taken place in idle balances the greater will the further rise of the rate of return have to be in order to induce the release of a further amount of given magnitude from those balances. And since there is clearly a maximum beyond which, for technical reasons, CR. XXVI Short-run Influence8 367 the velocity of circulation. cannot be increased (because there are no "idle" balances left), the curve must tend to become vertical for very high rates of return.

The significance for the determination of the rate of interest of this conclusion (that our a-curve will slope upwards to the right) is that a shifting of the curve of return upwards or downwards will lead to the release or absorption of varying amounts of money from idle balances, and that the immediate effect of a change of the curve of return on the rate of interest will be modified to that extent. The nature of this process can be further illustrated by means of our diagram if we introduce one further simplifying assumption which enables us to interpret it in a second way. The assumption which we Theiwosouroesofihe have to make for this purpose is that the demand for mODey amount of money that will be held by entrepreneurs in connection with investments for which they have definite plans, and by the recipients of the income created by all investments (the total of "transaction balances "), will change in exact proportion to the total of these payments.

On this assumption the distance OM in our diagram, which expresses expenditure on investment, can also be interpreted as representing the amount of transaction balances held; and since the total quantity of money in the hands of all concerned is assumed to be constant, and ON represents the case where idle balances are zero and all the money is held in transaction balances, MN measures the amount of idle balances held at any moment. Thus interpreted the diagram shows how the given supply of money will in various circumstances be distri buted between active balances and idle balances. It shows us how the two competing uses of money will in the short run jointly determine the rate of return on all kinds of investment and how misleading any assertion must be that the rate of interest will always depend either on liquidity preference only or on productivity only.

368 The Money Rate of Interest PT. IV The theory that the rate of interest depends solely on liquidity preference is an inference from the implicit assumption that the whole demand for money is due to liquidity preference - or at least that the demand for money for other purposes may be treated as constant. It will now also be clear that it is not sufficient, as Pro fessors Hicks and Lange have done,! merely to add the volume of money income to liquidity preference as a second determinant. For before incomes can rise with a given quantity of money a rise in returns must occur in order to induce somebody to reduce his idle balances. And it is only in consequence of a previous increase in investments, which, unless our a-curve is horizontal, will mean a higher rate of interest, that incomes and final demand will increase and in turn affect the investment demand schedule. This is, however, already outside the very short-term effects which we are considering in the present chapter. It cannot· be denied, therefore, that even in the shortest of short runs the investment demand schedule has it direct influence on the rate of interest and that any change in the former will directly lead to a change in the latter.

1 O. Lange, 1938, pp. 16 et seq.