XXII. The Adjustment of the Capital Structure to Foreseen Changes

CHAPTER XXII THE ADJUSTMENT OF THE CAPITAL STRUCTURE TO FORESEEN CHANGES THE term "capital" is a convenient description of the aggregate of nonpermanent resources and one difficult to dispense with entirely. Its use is, however, as we have Dangers of conceiving observed before, beset with dangers. Not capital as a .. fund" hI· f h d h' h of quanlltallvely de-t e east serIOus 0 t ese, an one to W lC 'ermined magnilude the majority of economists have succumbed at one time or another, is the temptation to regard the stock of " capital" as a quasi-homogeneous, quanti tatively determined magnitude which can, like the supply of any other factor of production, be treated as a datum of economic analysis. One of the main conclusions of the whole of the preceding discussion is that the supply of capital can not be treated as a single quantity in this sense. Nevertheless, in view of the established position which this idea occupies in economic theory, it seems advisable to examine it further.

The idea that capital is a quantitatively determined fund which has some existence apart from and beyond the con crete nonpermanent resources in which it is" embodied", that it is "an entity capable of maintaining its quantity while altering its form '',1 pervades in greater or lesser degree almost all the literature on the subject. Indeed it may be said that in the usual treatment of dynamic problems, the idea of a given stock, or a given supply, of capital occupies a central position. It is usually taken for granted that capitalists will normally aim at keeping their capital stock constant in some quantitative sense, and that they will succeed in doing so. Any serious 1 A. C. Pigou, 1935, p. 239. 294 cn. XXII The Adjustment to Foreseen Changes 295 attempt to analyse what is meant by "maintaining capital intact" ought soon to have shown that behind the specious simplicity of the idea there lurked a host of confusions and illegitimate assumptions. But most economists seem to have been una ware of these diffi culties and have made no such attempt. 1 The term " capital" was applied to the stock of nonpermanent resources, because it was felt that all the items in that stock had certain common qualities. This led to the idea that they might be treated as a homogeneous mass, that in the heterogeneous components of this aggregate was embodied some common substance which could be preserved irrespective of changes in its composition.

The consequence of the introduction of this fictitious magnitude becomes evident as soon as it is used as a supposed " datum" in the analysis of the effects of any sort of change. In the mere description of The quantity of capital t t · t t h' h d d cannot be treated as a s a lonary s a e w IC was regar e as given in the analysis being already in existence it mattered little. of dynamic changes If everything had always been correctly foreseen, and conditions always turned out to be what they were expected to be when the capital goods were created, there would never arise any problem of deciding how to use the existing capital goods. 2 They would simply be put to the use for which they were made. It is only in con nection with adjustment to unforeseen changes that the stock of capital goods has to be treated as a datum. But clearly it is precisely in this connection that it is not per missible to describe the stock as a single magnitude, or 1 The only systematic attempt to clear up the meaning of this concept, that made by Professor Pigou, I have discussed in some detail on another occasion (cf. 1935b). Parts of this as well as of a later article (1936a) on a related topic have been incorporated in the present chapter.

2 This is strictly true only of capital goods in the narrower (or in some respects wider) traditional sense of "produced means of pro duction ", but not of the wasting natural resources which are included in our definition of capital. On this point see my article just quoted (1936a, p. 226).

296 Oapitalistic Production under Oompetition PT. III in terms of some common unit of measurement. It is essential to go back to its real components and to de scribe it by a full enumeration and description of all its constituent parts. 1 It also becomes necessary to make a systematic study of how those in command of capital goods will behave when they find that events are turning out differently from what they expected. Here we should Nor Is there a clearly f defined neutral atti-begin with a consideration 0 the concrete tude o( entrepreneurs which can be said to represent the normal, Involving neither ad ditions to, nor sub tractions (rom, their capital stock opportunities which will be open to them under the new conditions, and of their pre ferences for the income streams of different shapes between which they can now choose. In discussions of this problem it is usually implied that there is a clear line between the normal process of maintaining and replacing the existing capital, and making a net addition to it. It is assumed that it is always possible to decide in an unequivocal way whether the amount of capital remains constant, increases, or decreases, and that there are typical phenomena con nected with each of these processes which, at least con ceptually, can be clearly separated. But as soon as one tries to apply these categories to a world where things are changing, all these alluringly simple concepts become dependent in more than one way on the answer to the following question: Exactly what is meant by a constant stock of capital ~ It is impossible to define net income (or "earnings"), and therefore savings, before one has separated from the gross produce those quantities which are required for capital maintenance. It is equally impossible to say what are additions to the stock of real capital before one knows what capital goods are required to make up for current depreciation. All this is one 1 The only author who, to my knowledge, has clearly seen that the traditional way of treating a quantitatively determined stock of capital as a datum is illegitimate, was Knut Wicksell. Cf. his statement quot~d above, Chapter I, p. 8.

CH. xxrr The Adjustment to Foreseen Changes 297 of those problems where it becomes particularly clear that the real problems of capital arise only when we go beyond the limits of the construction of a stationary state. As has been pointed out before, the very existence of nonpermanent resources which will not or ca.nnot be reproduced in an identical form is incompatible with the idea of a strictly stationary, repetitive process. It will always cause a process of continuous change in which each step is determined by the historical accident of the existence of a certain collection of non -permanent resources. Before we can proceed, however, to discuss the reac tions of the capitalist entrepreneur 1 to unforeseen changes, we must clear away certain widespread con fusions concerning the appropriate atti-I h The reaction ole tude towards changes which can be fore-capitalists 10 loreseen seen. The points to be brought out here changes are all implicit in the discussions of the preceding chapters, but they will be a useful introduction to the problems of the next chapter if we restate them systematically. There are two main sources of confusion, both connected with prac tices widely adopted by entrepreneurs in their deprecia tion and amortisation policies. The one is the practice of considering the capital stock mainly as representing a certain money value and of aiming at keeping this money value constant. The other is the habit of regarding depreciation in the narrower sense, as caused by wear and tear, as being something more fundamental and important than a " mere" loss of value through" obsolescence".

The first of these two practices represents a sort of abbreviated or simplified method, a kind of mechanical rule-of-thumb for obtaining an approximate solution to problems which are - as we shall see - sometimes ex tremely intricate. The historical explanation of this practice of capital accounting in terms of money is to be 1 In this connection the functions of the capitalist and of the entre preneur cannot be clearly distinguished and the two terms will there fore be used interchangeably.

298 Capitalistic Production under Competition PT. In found· in the technique of double-entry book-keeping. And the practice derives a certain justification from the fact that a large part of the capital of many firms is proM vided in the form of loans of money, and alntalnlng the money value of that one of the chief purposes of bookcapital constant k·· k t· th t th " 1· eeplng IS to ma e cer aln a ese labilities" are covered by " assets". This is also, of course, the chief source of the conception of capital as a "fund". That such is, in rough outline, the actual practice of entrepreneurs has important consequences which would require careful consideration in a more realistic investiga The rationale of maln-tion. As monetary problems, however, tainlng capital Intact they largely fall outside the scope of the present study. But there is a preliminary task which is germane to the present investigation, and that is to determine what will be the most appropriate action of entrepreneurs who wish to preserve their stock of non permanent resources, given the ultimate ends which this stock serves. It is only against the background of some such set of principles deduced from the rationale of " maintaining capital intact" that it will be possible, in more realistic studies, to judge the significance and consequences of the practices actually followed.

The "maintenance of capital intact" is, of course, not an aim in itself. It is desired only because of certain consequences which are known to follow from the failure to maintain capital. We shall see, moreover, that once it is applied to a changing world the phrase has no definite meaning independently of the reason why entre preneurs want. to maintain their capital. We are not interested in the magnitude of capital because there is any inherent advantage in any of its conceivable absolute measurements. We are interested in it because, ceteris paribus, a change in it will cause a change in the income to be expected from it, and because in consequence every change in it may be regarded as a symptom of such a change in the really relevant magnitude: income.

OH. un The Adju8tment to Fore8een Change8 299 What, then, are the reasons why we wish capital to behave in a particular way ~ The main reason is evidently that the persons who draw an income from capital want to avoid using up unintentionally parts of The significant mag the sources of this income. They believe nUude Is Ihe lime shape of the Income that these must be preserved if income is to to be obtained be kept at the present level. They want to avoid uninten tional "splashing" or " stinting" (to borrow Professor D. H. Robertson's terms) which would have the effect of later reducing income below (or raising it above) the level at which they aim. Capital accounting in this sense is simply a shorthand device for preventing involuntary encroachments upon future income. Whatever the time shape of the future income stream, derived from the capital in his possession, at which an individual aims, there still remains the problem of deciding what is the appropriate disposition over the individual assets in his possession. And although we certainly have no right to assume that every person will normally aim at a permanent constant stream of income from his capital, there is prob ably some justification for regarding this case as one of special interest. Moreover, even when a capitalist aims at some other shape of income stream, the problem remains essentially the same, and the case of the constant income stream may simply be regarded as a standard with which the others may be compared.

Will this aim always be achieved if the immediate goal followed from moment to moment is to maintain the money value of the capital stock constant 1 The answer is no. Quite apart from the possibility of general changes in prices, due to monetary causes which do not concern us here, it is obvious that, under changing conditions, the money value of the stock of non-per Xe.plng Ibe compo,l lion or Ihe money valu. 01 Ih. stock 01 capital constant will not secure a constant Income stream manent resources which is required at successive moments in order to secure a constant stream of real income, need not be constant even if all changes have been foreseen.

300 Capitalistic Production under Competition PT. III If it were known beforehand that from a certain date onward a stock consisting of the same capital goods would bring a smaller return than before, a policy which aimed merely at maintaining a stock of goods of the same kind would fail to yield a constant income. The con sumption of the larger income during the earlier period would be at the expense of a later reduction to a lower figure. A policy aiming at a constant income stream would have to use parts of-the higher" return" during the earlier period to build up a larger stock of non permanent resources such that the return obtained after the change would be equal to the income actually con sumed before the change. And the same is true, mutatis mutandis, of an expected increase in the return from a given stock of capital goods. It will be seen that a series of such changes in either direction, if correctly foreseen from the beginning, will make it necessary, if the income obtained is to be kept constant, for the stock of non permanent resources sometimes to increase and some times to decrease, no matter in what terms we measure that stock. And no conceivable sort of monetary policy could make prices behave in such a way that a policy which aimed merely at keeping the money value of the capital stock constant would simultaneously secure a constant real income from that stock.

The main point, as will become clearer when we pro ceed to consider in detail the reactions of capitalists to unforeseen changes, is that as soon as we go back to the Changes In the meas urable dimension of the capital stook it sell play no essential rtile In the complete economic calculus rationale of maintaining capital intact, the quantity of capital drops right out of the picture as a directly relevant magnitude. Its place is taken by a direct consideration of the size of the income streams that may be expected at different dates. What is relevant is whether a person maintains a stock of nonpermanent resources which will secure him an increasing, constant, or decreas ing income stream, not whether this stock itself increases, • CR. XXII The Adjustment to Foreseen Changes 301 remains constant, or decreases in any of its directly measurable dimensions. And - as will also be discussed in some detail later - it is these changes of income at different dates, and not changes in some absolute size of the capital stock, which must be made the basis of the common distinction between current consumption and saving on the one hand, and between current production and investment on the other hand, if these distinctions are to have a definite and useful meaning.

There is also a second point with respect to which the materialist conception of capital as a measurable sub stance leads to erroneous views. It relates to the distinc tion between depreciation in the narrower •. Obsolescence sense, or physlCal depletIOn through wear and tear, and" mere" obsolescence, or the losses in value which occur without a corresponding change in physical substance. In the light of our whole approach it will be obvious that any suggestion that the former is in any sense more fundamental than the latter is baseless, and we may deal with this point more briefly. We need only remind ourselves that the entrepreneurs are quite as likely to foresee that a capital good will become useless long before it is physically worn out or decayed, as they are to know that its useful life will be terminated only by its physical breakdown. l We then see immediately that in the one case just as in the other, if a constant income 1 There can be no doubt that in actual life many investments are made with complete awareness of the fact that the period during which the instrument concerned will be useful will be much shorter than its possible physical duration. In the case of most very durable constructions, like the permanent way of'a railroad, the prospective " economic life" ought to be regarded as much shorter than the possible " physical life ". In many cases it lies in the very nature of a product that it must be made almost infinitely durable, although it is needed only for a very transient purpose. It is impossible to adjust the dura bility of a machine to the short period during which it may be needed, and in many other cases the strength needed from a construction while it is used necessitates its being made in a form which will last much longer than the period during which it will be needed.

302 Capitalistic Production under Competition PT. III stream is to be maintained, only such parts of the expected services may be regarded as net income as are compatible with the reproduction of new nonpermanent resources which will secure an equal income in the future. This means in particular that investment in capital goods liable to obsolescence must be so restricted that the prices of their services, or their gross return, will include an amortisation quota sufficient to replace them ultimately by new capital goods which will yield a future income equal in value to the net return of the old ones which are being consumed. 1 But it should be borne in mind that the concept of amortisation itself suggests the misleading idea that a certain quantity of capital has to be recovered and reinvested. The essential point, of course, is not that the new capital good (by which the one that has become 1 The significance of such a decision on the part of the capitalist becomes particularly clear if we consider the following case. Suppose that the capitalist has to choose between two investments of equal cost, both represented by instruments of equal physical durability, but one of which is expected to remain useful so long as it lasts physically, and the other of which serves only a very transient pur pose. Under what conditions will he regard the two investments as equally attractive? The first answer is, of course, if they promise him the same permanent income. But under what conditions do they promise him the same permanent income? Suppose that the gross receipts from the investments while they yield a return are equal in both cases, and are just sufficient to provide the same income plus an allowance for depreciation proportional to the physical deterioration.

The effect will then clearly be that in the case where the instrument ceases to be useful long before it is worn out physically only a fraction of the sum originally invested will have been recovered, which of course will bring only a much lower income in the future. To decide in favour of this alternative would mean treating an income stream which starts at a given magnitude but decreases later, as equal to an income stream which is permanently kept at the initial magnitude of the former. In order that investment in the instrument of only transitory usefulness may appear equally attractive to that in the instrument of more lasting usefulness, it will be necessary for the former, while it remains in use, to produce gross returns sufficiently large to allow for full replacement of its original income-earning power. In other words, in order for the two investments to be regarded as equally attractive, the expected return must be sufficient to cover, in addition to the same income, not only depreciation in the narrower sense, but also obsolescence.

CR. XXII The Adjustment to Foreseen Changes 303 obsolete is replaced) should be of the same magnitude, but only that it should promise to provide the same income. There is still a further question of consi9.erable import ance which we must consider explicitly: What income is it that the capitalist is supposed to consider when he decides about the use of his nonpermanent DilIeren •• s accordIng resources 1 Is it solely the income which as only Income from capItal or all Income he derives from capital or is it his total Isregardedasrelevant income, derived from all his resources, permanent and nonpermanent? The answer to this question assumes special significance when we go on to apply these considerations to the case, which will be the rule in a competitive society, where the nonpermanent resources and the permanent resources with which they co-operate belong to different groups of people. In every society it is, of course, only a fraction of the people who deliberately regulate - or are in a position deliberately to regulate - the shape of their future income streams, and by so doing become capitalists.

But, as we have seen, any decision about capital will affect not merely the income from capital. It will also affect the income from the permanent resources and, therefore, in a competitive society, the income of people other than the capitalists making the decision. So far as concerns the isolated capitalist there can be no doubt that the dominant consideration influencing his decision will be the income to be derived from all his resources. But are we to regard him as maintaining capital intact if he keeps his total income constant, or if he keeps only his income from capital constant? The difficulty arises from the fact that in view of the limited life of any individual all his resources are in a sense nonpermanent from his point of view, and any rigid distinction between permanent and nonpermanent resources becomes impos sible. If we applied our definition of capital strictly and included all nonpermanent resources, including human labour, the difficulty would of course disappear. Yet 304 Capitalistic Production 'under Competition PT. III the fact that the total supply of labour is not provided merely from economic motives makes this practice inadvis able. From the point of view of society the distinction between the nonpermanent resources and resources which are either really permanent or which, like human labour, are replaced irrespective of their productive capacity, remains important. It is therefore convenient in this connection to follow traditional usage and to look upon the capitalist, qua capitalist, as if he drew income from capital only, so that in his case aiming at a constant income stream from capital and aiming at a constant total income stream become one and the same thing.

Then any consideration of the probable fate of any income which the capitalist of the real world may draw from sources other than his capital, and which will influence his decisions concerning this capital, will, for theoretical purposes, and in conformity with traditional usage, have to be classed among the motives which will lead him to save or dissave, that is, to aim at an increas ing or decreasing income stream from his capital. 1 This must conclude this admittedly sketchy and in complete discussion of the attitude of the capitalist to changes which are completely foreseen. But this assump tion of complete foresight is in any case so unrealistic 1 The interesting fact about such a policy on the part of capitalists is that it tends to accentuate the fluctuations in the income of other resources. At least this is so in what is usually considered as the " normal" case, where the capitalists keep the quantity of capital in some sense constant. If we take, for instance, the case where a foreseen change tends to increase the income of the permanent resources at the expense of the income from capital, the attempts of the capitalists to provide against a decrease of their income by building up a larger stock of capital will further increase the marginal productivity of the per· manent resources. And, vice versa, if capitalists foresee that an im· pending change will tend to increase their income at the expense of that of the permanent resources, and consequently start to decumulate capital, this will have the effect of reducing the marginal productivity, and consequently the income, of the permanent resources, even below the figure at which it would have been kept if the stock of capital had been held constant in some quantitative sense.

CH. XXII The Adjustment to Foreseen Changes 305 that the only justification for any detailed consideration of it is that it illustrates certain general principles. The more important question is the reaction of the capitalists to unforeseen changes. T his will be considered in the following chapters. 21 CHAPTER XXIII THE EFFECTS OF UNFORESEEN CHANGES AND IN PARTICULAR OF INVENTIONS IN considering the alterations which capitalists will have to make in their amortisation policy in face of a change in circumstances which they did not foresee at the time R II f h I of making some particular investment eac ODS 0 t e cap talisls to unforeseen which is affected by this change, it makes changes little difference whether we assume that the unforeseen event occurs quite unexpectedly or whether we assume that its imminence becomes known some time after the investment has been made. In both cases the capitalists learn that from a given (present or future) date onwards gross returns from their investment will be gmaller or larger than what they expected when the investments were made, and that consequently the mnortigation and reinvestment plan on which they had originally decided will not secure them a constant income gtream for the future. To go on consuming as much as they had planned has become incompatible with main taining consumption permanently at the present level.