Capital and Its Structure
By Ludwig M. Lachmann
For a long time now the theory of capital has been under a cloud. Twenty years ago, when Professor Knight launched his attack on the capital theories of Boehm-Bawerk and Wicksell, there opened a controversy which continued for years on both sides of the Atlantic. Today very little is heard of all this. The centre of interest has shifted to other fields.In practice of course problems concerning capital have by no means lost their interest. There can be few economists who do not use the word ‘capital’ almost every day of their working lives. But apart from some notable exceptions, economists have ceased to ask fundamental questions about capital. It is pertinent to enquire why this has happened. It would seem that there are three major reasons to account for this curious neglect…. [From the text]
First Pub. Date
Kansas City: Sheed Andrews and McMeel, Inc.
First published 1956 by Bell and Sons, Ltd., on behalf of the London School of Economics and Political Science 2nd edition.
The text of this edition is copyright ©1977, The Institute for Humane Studies.
In what follows we shall make use of our newly acquired knowledge of structural relationships between assets in general, and capital goods in particular, in order to elucidate some problems of the Trade Cycle. By ‘Trade Cycle’ we shall mean nothing more precise than the periodic ups and downs of output, incomes, and employment to which modern industrial economies seem to be prone. We do not assume a high degree of uniformity between successive fluctuations, but just enough similarity to make comparison possible. As Professor Hicks has said, ‘We ought not to expect that actual cycles will repeat each other at all closely. Certainly the cycles of reality do not repeat each other; they have, at the most, a family likeness.’
The task of trade cycle theory is therefore not confined, as it has been so often in the past, to explaining the similarities of successive fluctuations. The dissimilarities also have to be accounted for. It is certainly our task to indicate causes for downturn and upturn, and to analyse the cumulative processes of expansion and contraction. But on the evidence we have no right to believe that these causes will always be the same, nor to doubt that their relative force will vary from case to case. Similar causes will of course produce similar results. The dissimilarites we observe have then to be explained by the large number of potential causes not all of which become actual in each instance. The similarities are too many for the group of possible causes to be very large, but the dissimilarities are too many for it to be very small.
The Trade Cycle cannot be appropriately described by means of one theoretical model. We need a number of models each showing what happens when certain potential causes become operative. The many models that have been constructed by economists in the past are therefore not necessarily incompatible with each other. Overinvestment and
underconsumption theories, for instance, are not mutually exclusive. None of them of course is the true theory of the Trade Cycle; each is probably an unduly broad generalization of certain historical facts. Once we admit the dissimilarity of different historical fluctuations we can no longer look for an identical explanation. In dealing with industrial and financial fluctuations eclecticism is the proper attitude to take. There is little reason to believe that the causes of the crisis of 1929 were the same as those of the crisis of 1873.
Of late this has come to be more widely recognized. Professor Hicks, for instance, distinguishes between ‘weak booms’ which ‘die by working themselves out’ and which lend themselves to an underconsumptionist explanation, and ‘strong booms’ which end by ‘hitting the ceiling’ and in the explanation of which scarcity of productive services must play a part. And a feeling of the immense complexity of trade cycle problems is fairly noticeable in many quarters.
There can of course be no question of our traversing the whole immense field in this chapter, nor even of our reviewing the literature of the past twenty years. Our aim is to elucidate some trade cycle problems, it is not to set forth a new trade cycle theory. The day of ‘comprehensive’ trade cycles theories is long past. For our part, our interest in the matter is largely, though not exclusively, confined to the exploration of the part played by
structural maladjustment. We see no reason to believe that its influence is ubiquitous, but even less to doubt that in many cases it is pronounced. In this connection the ‘strong boom’ is of particular interest to us. Where the capital resources available for investment prove to be inadequate, their composition cannot be a matter of indifference. We have here evidently an instance of
inconsistent capital change.
The Hicksian theory of the trade cycle is a theory of the strong boom. We shall therefore in this chapter start with a critical examination of certain of its features. In doing so we shall encounter certain difficulties which arise from the assumption, implicit in most of Professor Hicks’ theory, that all capital is homogeneous. We shall then find that some of these difficulties can be met by introducing assumptions about the capital structure and its distortion in a strong boom which
appear to follow from the main argument of this book. Finally we shall survey the situation on the morrow of the downturn and study the measures necessary for readjustment, in particular the forms of capital regrouping which, if undertaken in time, would prevent a further deterioration of the situation.
In the Hicksian model investment plays the most prominent part in generating the cycle. Investment means the creation of new capital goods. Yet Professor Hicks, while he has much to say about investment, has little to say about capital. A theory of investment without a theory of capital, however, is very much like Hamlet without the Prince. Now, Professor Hicks regards the cyclical vicissitudes he describes as a concomitant of an ‘expanding economy’. It might be held that in dealing with economic expansion we are only interested in rates of expansion, and not in the expanding magnitudes themselves. In fact, the whole purpose of Professor Hicks’ model is to show how unequal rates of expansion engender cyclical fluctuations. But if the cause of the unequal rates of expansion lies in the expanding magnitudes themselves, the latter cannot simply be ignored. Where these magnitudes change their composition in the process, a theoretical model which neglects such change must be regarded as inadequate. Ultimately, this consequence of the heterogeneity of resources must produce ‘structural stress’ and thus cause the rate of capital expansion to slow down. The neglect of this fact in the Hicksian model leads to certain difficulties. Of these we shall give three examples.
First, Professor Hicks, following Mr. Harrod, makes much of the ‘background of economic expansion’ against which his model is set; but about the forces engendering this expansion remarkably little is said. Several times
*77 we are told that expansion may be due either to growth of population or ‘due to many of the various causes which can be grouped together as increasing productivity’.
*78 This of course may simply mean
technical progress. But what about the division of labour and capital as forces of progress, and their implications?
We saw in Chapter V that the division of capital entails a change in the composition of capital in the direction of more complex complementarity, and that such change usually takes the form of increasing prevalence of fixed capital. How is this fact accounted for in the Hicksian model? It would appear to be covered by the ‘long-range’ investment (p. 59) much of which (but how much?) is included in the notion of
autonomous investment. Many critics of Professor Hicks’ work have pointed out what a vague and unsatisfactory notion Autonomous Investment is, and that instead of being formally defined its meaning is merely illustrated by means of enumeration of a few examples. Be that as it may, for our purposes the distinction between Autonomous and Induced Investment proves particularly unfortunate in that it separates conceptually what economically is inseparable because complementary. Economic progress, where it is not due to changes in technical knowledge, is largely the result of a changing combination of an increasing number of specific capital resources, some of them indivisible. If some of these are the product of autonomous, some of induced investment, as all working capital evidently is, we can gain no clear and comprehensive picture of their
This is of course due to the fact that Professor Hicks’ interest is confined to the relationship between changes in output and quantitative change in (some) capital. Professor Hicks does not discuss the effect of capital change on output, or rather, the latter effect is partly
subsumed in the general results of autonomous investment, and partly in the magnitude of the Accelerator. The changes in the composition of C without which Y cannot grow are disregarded, and this makes a correct understanding of the vicissitudes which befall the economic system at the ‘ceiling’ all but impossible. We shall later on return to the subject of autonomous and induced investment. Meanwhile we shall note that the source of the trouble evidently lies in the implicit assumption that all capital is homogeneous.
Secondly, in the Hicksian model the acceleration coefficient ‘
v‘ plays a vital part. It is assumed to remain constant throughout the upswing of the cycle, as well as between cycles.
Critics like Professor Lundberg
*79 have pointed out that this is a highly unrealistic assumption to make. In shipping ‘
v‘ must be higher than in the grocery trade. As
varies in different industries, so does
It is readily seen that, again, the difficulty is due to the implied assumption that all capital is homogeneous. In fact, it might almost be said that Professor Hicks has forced the ‘constant
v‘ hypothesis upon himself. Of course, even if all capital were homogeneous, the ratio of capital to output might still be different in different industries, and the cause of such variations would still have to be explained. But there can be little doubt that in reality qualitative heterogeneity of capital is the most frequent cause of quantitative variations in the Accelerator, for instance the different ratio of fixed to working capital in different industries. As long as we remain oblivious of this fact we cannot account for these variations and shall tend to disregard them; hence the assumption of a constant accelerator.
Third, when Professor Hicks comes to grapple with the problem of the ‘ceiling’ he has to drop the homogeneity assumption. It is easy to see why: Were he to follow Keynes in assuming complete homogeneity of all resources, there would be no ceiling at all. There would only be a point of full employment, and beyond it the realm of inflation. Multiplier and accelerator being what they are, there is no more reason why our economic system, once set in motion, should stop at this particular point rather than at any other. Evidently the reason for the existence of the ceiling has to be sought in unequal expansibility in different sectors which cannot be overcome by intersectional transfer of resources. But Professor Hicks employs only two ‘sectional ceilings’, and hence only one heterogeneity. ‘Let us therefore suppose (as is realistic) that different sorts of resources are specialized to the production of investment goods and consumption goods respectively; and consider what happens if the production of investment goods reaches its ceiling at a time when the production of consumption goods is still capable of further expansion.’
It is true that later on he admits, ‘We could easily have made a further advance by splitting up these ceilings, and allowing a sectional ceiling for every product…but I do not think that it would make much difference to the argument.’
*81 In some ways, however, it can be shown that it would make a difference.
The reality of the ceiling has been doubted by one of Professor Hicks’ critics.
*82 It is perhaps not surprising that a generation brought up on an intellectual diet of Keynesianism and memories of 1929 should no longer be able to grasp the meaning of a strong boom. There can be little doubt that in history strong booms have ‘hit the ceiling’, i.e. been checked by a scarcity of resources. But where are we to look for the manifestations of scarcity? We suggest that, historically speaking, they are primarily to be found in the sphere of industrial raw materials, that in the past the
raw material ceiling has been the sectional ceiling of crucial importance.
In his interesting study of
World Production, Prices and Trade,1870-1960*83 Professor W. A. Lewis has computed ‘terms of trade’ for industrial raw materials (primary products other than food), viz. the ratio of their prices to those of manufactured goods. His statistics show that between 1870 and 1913 all the years in which the index (1913 = 100) reaches or passes the 100 mark were ‘boom top years’.
These, then, were years in which industrial raw materials became relatively scarce.
The argument gains further support from the cyclical record of the divergence of the index number of the quantity of international trade in primary products, C
T, from that of the world production of manufactures (excluding the U.S.A. and the U.S.S.R.), M
N. Professor Lewis finds that over the period 1881 to 1929 ‘a 1 per cent increase in world manufacturing is associated with an 0.87 per cent increase in world trade in primary products’ (p. 113).
The annual divergencies from this ratio bear an obvious relation to the trade cycle.
|‘PERCENTAGE DIVERGENCE BETWEEN ACTUAL AND CALCULATED C
The minuses, it will be seen, appear at the top and bottom of each column, the boom years, while the pluses are in the middle'(p. 114).
Professor Lewis’ interpretation of these figures is: ‘Stocks of primary products are accumulated by importing countries during the slump, and are used up during the boom.’ But this interpretation is both, factually and analytically, open to doubt. By no stretch of the imagination can years like 1895, 1905, and 1912, all years of strong positive deviation, be called slump years. In these years world production of manufactures (excluding Russia) increased by 9.8, 10.6, and 8.7 per cent respectively while their secular trend rate of annual increase was 3.6 per cent (p. 126). In these years, then, the expansion of raw material production not merely kept in step with, but actually exceeded the rate of industrial expansion. There was no raw material ceiling and general expansion continued unhampered.
Moreover, Professor Lewis’ interpretation is pre-Hicksian, ignores the ceiling, and is based on underconsumptionist premises. The true explanation appears to be that in 1872-3, 1890, 1900 and 1907 manufacturing industries ran into a raw material ceiling. We conclude that the raw material ceiling has often been of fairly decisive importance.
Professor Hicks speaks of an investment goods ceiling. Investment goods may be broadly, though not inadequately, divided into industrial raw materials and fixed capital goods. Can we treat the two together in the way Professor Hicks does? Can we assume that the two sub-ceilings are hit at the same point of time? We may doubt it, since if it were so, why do the intersectional terms of trade fluctuate so much? If the two sub-ceilings are not hit at the same time, but the raw material ceiling reached first, there will be fixed capital goods which cannot come into full operation, at least not in the way such operation was planned
ex ante, owing to lack of their working capital complements. They will provide a peculiar kind of unplanned ‘excess capacity’ and constitute a phenomenon which appears to be of crucial importance. Unless the various ‘investment goods sub-ceilings’ are all encountered at the same moment (and why should they?) the emergence of this ‘dynamic excess capacity’ is almost inevitable.
It might be said that raw material prices being more flexible than fixed capital goods prices, relative price figures tell us little about relative scarcity. It is true that a fixed capital goods ceiling will manifest itself, at least at first, in delayed delivery rather than in higher prices, so that absence of higher prices does not necessarily mean absence of excess demand. But the delay in delivery can only postpone, and not prevent, the emergence of excess capacity, unless of course the raw material shortage is merely temporary, not a ‘
ceiling‘ but a ‘bottleneck’. The mere fact that after both sub-ceilings have been reached the output of both, raw materials and fixed capital goods, will slow down, is irrelevant. It is relative scarcity of complementary factors which here causes excess capacity and upsets plans. For no factor can be used in isolation, complementarity is of the essence of all plans, and withdrawal of a factor, or its failure to turn up at the appointed time, will equally endanger the success of the production plans.
We have just dealt with a phenomenon which occurs
before the new capital combinations can be taken into use. We must now ask what happens afterwards.
The effect of capital investment on output raises a number of questions, not merely of ‘effective demand’, of consumption keeping in step with output. If we were dealing with a weak
boom, these would be the most relevant questions to ask. But the effect of investment on output also depends, and certainly in a strong boom, on the degree of complementarity of the different capital resources employed, in other words on the degree of consistency of such capital change. This raises a number of questions which, to our knowledge, have rarely, if ever, been asked, at least in the field of trade cycle theory.
We said above that the distinction between autonomous and induced investment is, for our purposes, unfortunate since it tends to separate conceptually what economically is inseparable because complementary. But at the same time its critical examination will afford us a welcome opportunity for a study of forms of consistent and inconsistent capital change accompanying industrial fluctuations. By elucidating the nature of those economic forces which make the various forms of capital change inseparable we may hope to learn a good deal about the direct and indirect effects of investment on output.
The relationship between autonomous and induced investment may be viewed from three different angles.
First, the two in general are complementary as they jointly determine total incomes and employment. This is the aspect in which Professor Hicks is chiefly interested.
Secondly, at or near the ceiling they begin to compete for resources. In fact, here a ‘dip’ in autonomous investment at the right moment would give induced investment a ‘breathing space’ before the ceiling is hit. This is implicit in the whole argument.
Third, there is the much larger issue which concerns us here: the complementarity of the products of the two types of investment, the actual capital resources, after they have taken shape. Professor Hicks does not deal with this question; if it has any place in his model it is subsumed in the slope of the ceiling.
If any structural complementarity exists between the capital resources in an economic system, this clearly is an important problem. Variations in the rate of the two types of investment must have some effect on the productivity of the capital resources produced. Some of these variations at least must fall into the class of inconsistent capital changes. This is clearly seen in the case of induced investment which depends on the rate of increase of output, hence is not independent of the
productivity of earlier investment, both autonomous and induced.
Structural complementarity of course does not mean fixed coefficients of production. We saw above (p. 42) that in an uncertain world the need for reserves sets a limit to the fixity of the coefficients of production within the framework of a plan. What is true for the complementarity of the production plan is equally true for structural complementarity, only that the place of the reserve assets is here taken by various forms of ‘excess capacity’. There must be some flexibility in the overall capital structure. Transport and power resources, for instance, must be such as to permit some growth and regrouping of secondary industries, and the same applies to all raw material production. It would be wrong to think that consistent capital change in the growth of autonomous and induced capital requires a one-to-one, or any other fixed relationship. On the contrary, a certain excess capacity, for instance in transport and power production, is necessary if a position is to be avoided in which any increase in capital in one industry requires a corresponding decline in another. It is just such excess capacity that makes a large number of capital changes in secondary industries consistent with each other. But all this means is that there is, in an industrial economy, as a rule a fairly wide range over which variations in the different rates of investment would be consistent changes. It does not mean that inconsistent change cannot exist.
There certainly can be too much autonomous investment relatively to induced, so that we have too much autonomous capital and too little induced. Or it may be the other way round: induced investment pushing against the ceiling which is the result of too little autonomous investment in the past, for instance a raw material capacity ceiling as indicated by Professor Lewis’ tables. Neither of course could happen in a Keynesian world, a world without ceilings and scarce resources, or at least there it would mean at worst a temporary hitch, a ‘bottleneck’, until ‘the other’ kind of investment has caught up. But in an economy moving near the ceiling scarce resources, once committed, may not find complements, however long we may wait. Excessive railway construction may not merely be a waste of present resources, it may also have the
effect of depriving the railways of future traffic by depriving railway customers of capital.
There can be little doubt, for instance, that the industrial development of Switzerland between 1850 and 1900 was gravely hampered by excessive railway building owing to the fact that the oligopolistic nature of the railway market made it necessary to stake claims by building lines long ahead of any possibility of their profitable use. The resulting capital shortage for a time provided a serious obstacle to the growth of Swiss industries.
In a strong boom investment plans are started for which adequate resources do not exist and which must therefore fail. What is worse, even the available resources are wasted by being given a form which for its effective use would depend upon the support of other resources which are not available. Not only are there too few resources, but the few are ‘scattered’ over too wide a field; they lack the support of other resources which would have rendered them more productive in their present uses than they are now.
We doubt whether the distinction between autonomous and induced investment can easily be made in practice; in many cases it would be most difficult to make it. We used this Hicksian pair of concepts merely to show that even if the distinction could be made conceptually as regards the form of investment, it cannot meaningfully be made as regards the resultant capital types. The resultant capital resources will have to be complementary, and if the two capital changes are inconsistent with each other there will be trouble, viz. interruption of the continuous investment process, even if the resource needs of the two types of investment did not clash at the ceiling. Neither autonomous nor induced investment is in fact independent of forces which, in their turn, depend on the degree of mutual consistency of the two types of investment.
Exception may be taken to the view here set forth on the ground that we have failed to distinguish between ‘economic growth’ and ‘cyclical fluctuations’, and confused phenomena
of the ‘long run’ with those of the ‘short run’.
*87 But in the light of what we have just said such an exception cannot be sustained.
The distinction between the long and the short run referred originally to the change in resources which occurs in the former, but not in the latter, where such change means purely quantitative change. The distinction between ‘given resources’ and ‘resources adapted to demand’ is unambiguous only where the adaptation means addition or subtraction. Where regrouping exists as an alternative mode of change the matter is no longer quite so simple. The whole notion is clearly linked to a purely quantitative conception of capital.
For Keynes of course the trade cycle means essentially fluctuations in the degree of utilization of existing resources. The short period is to him the period during which capital under construction emanates the multiplier effect, the long period that in which the new capital begins to produce output. In the Hicksian theory the juxtaposition of autonomous and induced investment amounts to an admission that if we have to explain why there is a ceiling and a bottom we cannot ignore ‘long run’ factors.
To us the whole division is artificial and unacceptable. Once we realize that industrial fluctuations are not merely a matter of utilizing existing resources in the short run, or increasing and possibly decreasing them in the long run, but also of regrouping them as well as increasing or decreasing them
in certain directions, the whole
fundamentum divisionis falls to the ground.
The possibility of multiple use of existing resources in successive periods blurs the simple line of distinction. This of course does not mean that time does not enter into the problems of capital. Time is germane to them, but not merely as the dimension in which the ‘quantity of Capital’ changes, but also as the dimension in which capital resources are turned from one mode of use to another. It was just for this purpose that in Chapter III we found it necessary to apply a form of period analysis to capital problems.
Moreover, economic progress in the long run also depends
on the productivity of the new capital which, in its turn, depends on the concrete form of the new resources. This form is determined by investment decisions made ‘in the short run’, and progress can be hampered by intersectional maladjustment.
In Chapter I (p. 10) we explained that investment opportunities really mean ‘gaps’ in the existing capital pattern. With respect to progress therefore we may say that the direction of progress depends on where these gaps are, thus using two concepts which would be meaningless in a world of homogeneous capital.
Once the homogeneity hypothesis has been abandoned the distinction between growth and fluctuations loses its meaning. This distinction finds a place in a theory which confines itself to asking whether and to what extent existing resources are being used, whether, and perhaps at what speed, such resources can be augmented, and what are the circumstances in which such augmentation is likely to take place. Once we have learnt to ask how, and in what order, existing resources are being used, and what are the implications of such multiple use, once we have begun to understand the importance of the concrete form of resources in limiting the scope of multiple use, we can easily dispense with the all too simple distinction between economic growth and cyclical fluctuations.
Thus far our approach to trade cycle problems has been mainly critical. In examining the Hicksian model we found certain weaknesses. A critical analysis of these provided us with an opportunity to set forth certain ideas which might help us to overcome them. We now have to turn to a more constructive task. The ideas mentioned have to be subjected to a test of coherence by welding them into a model, or at least as much of a model as is necessary to see whether and how they fit together.
The Hicksian theory is a theory of the strong boom. The only other model of the strong boom which has been worked out with any degree of precision is the body of ideas set out originally by Cassel and Spiethoff,
*88 and later developed by
*89 and Hayek,
*90 which has come to be known as the Austrian Theory of Industrial Fluctuations.
*91 Let us see whether with its help we can further develop and apply to the trade cycle the main ideas set forth in this book.
In expounding the Austrian theory there is no need to start
de novo. It is a body of ideas which has gradually evolved over the last fifty years. But it is necessary first to make some preliminary remarks about its nature, and thus to obviate certain misunderstandings which, experience teaches us, stand in the way of its correct understanding.
In the first place, we do not maintain that the Austrian theory could explain every and any industrial fluctuation that has ever occurred. Such a view of course would be incompatible with our plea for eclecticism. The Austrian theory is a theory of the strong boom, it deals with its causes and consequences. Undoubtedly, weak booms which ended when consumption failed to keep in step with production have occurred in history; America from 1929 to 1932 seems a prominent example. To account for them a different kind of model is required. All we contend here is that an underconsumption theory, which might account for the end of a weak boom, is not exactly a suitable instrument for analysing strong booms. And there is now good historical evidence to show that strong booms were a more or less regular feature of the expanding world economy in its ‘normal’ conditions from 1870 to 1914.
Secondly, the Austrian theory does not, as is often suggested, assumes ‘Full Employment’. It assumes that in general, at any moment, some factors are scarce, some abundant. It also assumes that, for certain reasons connected with the production and planned use of capital goods, some of these scarcities become more pronounced during the upswing. Those who criticize the theory on the ground mentioned merely display their inability to grasp the significance of a fundamental fact in the world in which we are living: the heterogeneity of all resources. Unemployment of some factors is not merely compatible with the Austrian theory; unemployment of those
factors whose complements cannot come forward in the conditions planned is an essential feature of it.
The Austrian theory does not rest upon a stationary model. Saving and investment play a prominent part in it, while of course in a stationary society there can be no such thing. It is also set against the background of economic expansion. Like Mr. Harrod and Professor Hicks it views industrial fluctuations essentially as deviations from the dynamic equilibrium path of progress. The condition of this dynamic equilibrium path is that planned savings equal planned investment. But the nature of the economic forces which in dynamic equilibrium, i.e. when all plans are consistent with each other, keep the economic system on an even keel by preventing planned investment from either exceeding or falling short of planned saving, and of those counter-forces which in certain circumstances prevent the first set of forces from operating, certainly requires further examination.
Finally, the Austrian theory not merely views industrial fluctuations as the result of maladjustment between planned savings and planned investment, but also as the result of structural maladjustment caused by the first type of maladjustment. In this way it is linked to a dynamic theory of capital of the kind outlined in Chapter V. In economic progress the degree of specialization of resources, in other words the scope there is for multiple use, is linked to the rate of accumulation of capital. Where the latter is misjudged, sooner or later the former will turn out to have been miscalculated.
We said that the conditions of stability in expansion, which keep planned investment within the bounds, and up to the extent, of planned savings, deserve further study. Mrs. Robinson has outlined a model in which these conditions hold and serve to keep the system on an even keel.
When an economy is expanding at the rate appropriate to the given conditions, all prices are equal to long-period average costs(including in cost, profit on capital at the given rate) and all capital equipment is working at the designed capacity. In each sector conditions of rising short-period supply price obtain, so that any increase in output relatively to capacity would be accompanied by a rise in price above long-period average cost.
The capitalists expect the rate of profit to continue in the future to rule at the present level.
Now, if we postulate that the capitalists’ expectations of future profits have great inertia and do not react to passing events, the system can be regarded as being in equilibrium from the short-period point of view. A chance increase in consumption would cause the output of consumption goods to rise above designed capacity, prices to rise above normal costs and so profits to rise above their long-run level. But since this state of affairs is not expected to last, investment is not stepped up, and no ‘acceleration’ occurs. Similarly, a chance increase in investment does not raise expected future receipts (in spite of a rise at the moment, due to the operation of the short-run multiplier). But the prices of capital goods have risen above the normal long-run level, the rate of profit to be expected on funds invested at these prices is less than the accustomed rate, and so, we may suppose, investment is checked. If investment chanced to fall, the price of capital goods would fall, the rate of profit to be expected on funds invested ‘at those prices would rise, and investment would pick up again. Thus, the postulate that expectations do not vary with current events may be considered to endow the system with short-period stability, and (combined with faith in future profitability of capital) to provide a presumption that the rate of investment tends to be maintained at a level which continuously corresponds to the gradually growing capacity of the investment-good industries.’
Mrs. Robinson says that she obtained this model by grafting Marshall’s analysis on to the Harrodian model. But the reader will not fail to notice how easily the Austrian conception of dynamic equilibrium fits into this mould. To be sure, Mrs. Robinson never mentions savings; to her, as to any Keynesian, savings are of course a mere residual magnitude. But it is obvious that a ‘chance increase in consumption’ means a fall in planned savings relatively to planned investment; otherwise why should prices rise? And a ‘chance increase in investment’ which leads to a rise in the prices of capital goods must mean a rise in planned investment relatively to planned saving. The essence of the matter is not merely that ‘expectations do not vary with current events’, but that prices moving away from their normal and expected level
provide entrepreneurs with a storm signal; they indicate
ex ante disequilibrium between savings and investment and tell investors when, and when not, to make new investment decisions. We may add that among these prices and costs rates of interest must play a significant part.
The economic system outlined in this model is kept on an even keel because every deviation from the equilibrium path will, when in an upward direction, release forces which via a rise in costs check investment, and when in a downward direction, stimulate it via a fall in costs. Thus, if the conditions of the model existed in reality, there would be no trade cycle. But at the same time the study of the model has proved worth our while; we now understand better what happens when its conditions are not fulfilled.
Where prices, wages, and rates of interest are inflexible, investment decisions to a large extent have to be made in the dark. Where interest rates are kept constant, no hint of
ex ante disequilibrium between savings and investment can transpire. Where raw material prices are ‘controlled’ and no rising wages give a hint of approaching labour shortage, we need not be surprised if we hit the ceiling with full force. The sensitive mechanism which emits the storm signals has been switched off; the deviation of actual prices from their normal and expected level can no longer serve as a measure of dis-equilibrium and a signpost for action. All this follows simply from what we said in Chapter IV about the role of flexible prices in the sensitive network of communications on which a market economy so largely depends. In such a situation entrepreneurs cannot correctly assess the relative availability and scarcity of the factors to be employed in their investment plans; the impenetrable smoke-screen of a deceptive ‘price stability’ shrouds them all equally. These plans of course cannot be carried out and disaster is the natural result. Dynamic equilibrium requires consistency of plans which, in its turn, depends on a flexible price system. Inconsistency of plans is an inevitable feature of a world in which prices (including forward prices and share prices) no longer tell an intelligible story.
We now have arrived at a decisive point of our argument. We have seen that in a strong boom entrepreneurs, deluded
by factor costs which are not equilibrium costs and therefore can say nothing about available supply, embark on investment projects the resources for which do not exist and cannot be created by a transfer of resources from consumption. Thus far we have not gone beyond what Cassel and Spiethoff said fifty years ago and most economists knew by, say, 1913. At best we have merely explicated the implications of their thesis. The specifically ‘Austrian’ element, the link with the theory of capital, has now to be brought into our picture. If all capital were homogeneous we should have little more to say about the trade cycle. As it is, the link with the main thesis of this book has now to be forged.
All entrepreneurial decisions are specifying decisions. Investment decisions determine not merely, as Keynes would have it, the ‘rate of investment’, but also determine the concrete character of each new capital good, whether building, plant, machine, etc. Each new capital good forms part of a whole and has to fit into a capital combination. We pointed out in Chapter III (pp. 49-50) that new investment depends on nothing so much as on the availability of cheap complementary resources of labour and capital. The complementarity of old and new capital goods shows itself within the firm and its production plan, but just as much in the structural overall complementarity of capital resources in the economy. Decisions on the construction of new capital thus involve all these complementarities. The entrepreneur in making his decision will be guided by his expectations about what complementary capital resources will be created during his investment period, and what other already existing resources will then be available in a complementary capacity, in just the same way as he will be guided by expectations about the future supply of labour. A railway is built in the expectation that the economic development of the area served by it will produce enough demand for its services. Engineering industries expand in the expectation that the industries who are their customers will expand at a certain rate. This fact is of some significance for the trade cycle in general and the strong boom in particular.
In Chapter V we learnt that economic progress involves the division of capital along with the division of labour. The more highly developed the economy, the more intricate the degree of
complementarity. Thus anything which gives a wrong picture of resources available for investment, and of the speed at which the economy as a whole can expand, will lead to wrong decisions about the degree of specialization of the new capital. That the economy ‘hits the ceiling’ may mean that a new railway line cannot be completed, or cannot be completed within the time planned, or at the cost planned. But it may also mean that even if it is completed as planned, it will lack complementary factors in the rest of the economy. Such lack of complementary factors may well express itself in a lack of demand for its services, for instance where these factors would occupy ‘the later stages of production’. To the untrained observer it is therefore often indistinguishable from ‘lack of effective demand’.
Rates of interest which are too low, i.e. fail to establish
ex ante equilibrium between savings and investment, are apt to convey such a misleading picture and thus to lead to wrong specifying decisions. In the Keynesian world in which idle resources of all kinds are abundant, and are to be had at constant cost in terms of ‘wage units’, all this does not matter. Here resources can be treated with impunity
as though they were homogeneous, simply because in any case there are always more than enough of them of whatever category. But in a world in which anything is scarce, and in which construction costs rise with investment demand in a strong boom, all this can hardly be ignored.
The essence of the matter is that investment decisions are not merely irreversible in time, so that excessive investment in period 1 as a rule cannot be offset by disinvestment in period 2, but that they are also irrevocable in kind. Even if, at a later point during the boom, interest rates start to rise, the message comes too late for those who have made their irrevocable decision before. If all capital were homogeneous there would be no sub-ceilings and the advance would not be halted until all resources had become equally scarce—Keynes’ point of full employment and full utilization. As it is, capital is heterogeneous, and the first sub-ceiling reached will necessitate not merely the revision of plans for the construction of new capital, but also the revision of plans for the use of existing capital.
We now turn to our last task in this chapter: to survey the problems which emerge on the morrow of the crisis, and to study ways and means by which they might be solved. The reader, we trust, will not expect to be told of an ‘adjustment mechanism’; in the realm of human action there is no such thing. A market economy, to be sure, has great resilience and can adapt itself to many needs, sudden as well as long foreseen. But this is not because of any automatic mechanism ‘built-in’ but because it serves in general to put the right man on the right spot. Successful adjustment to new conditions no less than whatever ‘stable progress’ there might exist, depend ultimately on the entrepreneurial qualities of mind and will which manifest themselves in response to challenge. Situations like those we shall study provide a crucial test of entrepreneurial ability.
We may remind the reader that we are primarily dealing with the situation which arises when a strong boom has collapsed as a result of the encounter with a ceiling, or at least a sufficiently tough congeries of sub-ceilings. This does not mean, however, that the situation which follows the end of a weak boom is none of our concern. It is of interest to us to the extent to which it gives rise to the need for capital regrouping; this need it shares with the aftermath of a strong boom. But this is not to say that capital regrouping by itself will suffice to overcome the situation following the end of a weak boom. As a rule it will not, and there may be much scope for the Keynesian nostrums. No doubt such an underconsumption crisis, due to a flagging ‘effective demand’, can be at least mitigated by increasing this demand, though it must remain a matter for decision in each case whether its investment or its consumption component needs to be strengthened. The need for capital regrouping arises from the peculiar characteristics of these circumstances.
Another point calls for notice: Any sudden and unexpected change in the ‘real situation’ will probably affect the demand for and the velocity of circulation of money. In the case under discussion widespread disappointment with the results of the past, the need for plan revision, the general loss of confidence,
will almost certainly have this effect. In a monetary system based on bank deposits, i.e. debts, in which the mere maintenance of any given quantity of money requires continuous agreement between creditors and debtors, the quantity of money can hardly remain unaffected. In particular, where the banks are involved in some of the investment schemes which have gone astray, the danger of
secondary deflation is always present. When that happens the ‘recession’ which succeeded the strong boom will turn into a ‘depression’, a cumulative process of income contraction, as has often happened in the past. Of course it need not happen. But to avert the danger must always be the primary aim of monetary policy in a recession. The reader is asked always to bear this in mind in the discussion which follows.
We now turn to the main issue. The situation which the economy faces on the morrow of the collapse of a strong boom clearly calls for capital regrouping on a large scale. On this score the reader who has followed the main line of thought of this book will have no doubt. The questions which arise are rather, what form the regrouping shall take and what are the circumstances favourable or detrimental to it. Plans have gone astray, hopes have been disappointed; capital combinations have to be dissolved and reshuffled. But what is the principle governing these changes?
Some planned combinations cannot come into operation because of lack of complementary factors; these factors have to be created now. But in general the rate of expansion of the whole economy will have to slow down. Yet the situation we confront is not, like that of a war or post-war economy, one of universal shortage. The scarcities are relative only. Something might be done by shifting resources to where they are most needed. The critical sectors are those sub-ceilings which lie in the path of expansion. Here more investment is required in order to ‘lift the ceiling’. To this end not merely must investment in other sectors be curtailed; additional factors able to help in lifting the ceiling must be recruited from wherever they happen to be, and this means as a rule that they must be withdrawn from those combinations of which they form part. Mobile resources from everywhere, even from the consumption goods industries, will have to be drawn to the critical sectors.
All this plays no part in the Hicksian theory. For Professor Hicks the slope of the ceiling in general, and the configuration of critical sub-ceilings in particular, are the long-run product of autonomous investment, something that lies entirely beyond the grasp of short-run action. It is true of course that this configuration of sub-ceilings is, at each moment, the cumulative result of forces that have operated in the past, but it does not follow that therefore nothing can be done about it in the short run. To us it is impossible, for the reasons mentioned, to separate the forces of progress from those of the cycle. To us the configuration of particular sub-ceilings is something that can be re-moulded by changing its composition, though it might be unwise to expect early results to transform the situation entirely.
In general the chief remedy for recession lies in increased investment in the critical sectors, even where such investment does not yield early results, and in the concomitant withdrawal of mobile resources from existing combinations. These mobile resources have to be detached from the specific and non-mobile resources with which so far they have co-operated, and this will lead to dissolution and reshuffling of existing combinations.
Such action creates a number of practical problems. The owners of the mobile resources of course act under the stimulus of the high earnings obtainable at their place of destination, but the owners of the specific resources to which until yesterday the mobile resources were complements, may have good reason to mourn their departure. Even where these mobile resources are not irreplaceable the earnings of specific factors are certain to be reduced.
The owners of a factory are unlikely to close it and let their plant lie idle merely because their liquid capital could earn a higher rate of interest elsewhere. But here the control structure is of some importance. In certain cases the creditors may compel their reluctant debtors to part with their mobile assets. In general we may say that where the division between the firm’s own capital and its debts corresponds most closely to that between its first-line, and its second-line and reserve assets, i.e. its fixed and mobile resources, the chances of successful withdrawal of the mobile resources are highest,
precisely because all the gain will accrue to the creditors and all the loss to the debtors. While, where the creditors own also part of the fixed first-line assets, they may be reluctant to incur the capital loss here involved. But in any case there will be enough resistance to all attempts to mobilize resources and disintegrate existing combinations to make the withdrawal of mobile factors a slow and precarious business.
This creates a problem for monetary policy. In all probability mobile resources cannot be withdrawn and capital combinations will not be reshuffled without pressure being brought to bear on owners and managers of specific resources. In some cases it may not be possible at all without actual bankruptcy. To this end a ‘severe’ credit policy is required. But a credit policy sufficiently severe to ‘crack open’ the tougher kind of unsuccessful capital combinations may discourage investment in the critical sectors of the economy.
Clearly this is a problem of policy which does not admit of a general answer. In such a situation there is much to be said for a ‘selective’ credit policy which need not be arbitrary if it merely reflects the degree of imperfection of the capital market which is the natural product of the past record of success and failure of individual firms.
The problem of surmounting intersectional maladjustment must not, however, be viewed exclusively as falling within the narrow context of the firm and its internal complementarities. We saw in Chapter IV that the same capital may give rise to service streams of very different kinds. An industrial economy has often a high degree of flexibility which may operate without much visible disintegration of combinations. Often the change is brought about by using the same plant to produce different products. Suppose our critical sub-ceiling is in mineral mining. It is surely unnecessary to deprive existing coal mines of their mining equipment. Our purpose of moving mobile resources to the critical area may be as well served by the heavy engineering industries switching their plants from pro-
ducing equipment for coal-using industries to producing mining (or ‘mineral-economizing’) machinery. In this way existing combinations may be moved bodily to ‘another stage of production’ without the painful need for disintegration. In such cases internal complementarity is preserved at the expense of structural complementarity. The complementarity relationship which hitherto linked the engineering industries to their customers in the consumer goods trades, the complementarity between successive stages of production, will now be broken; it is impossible to have change and to maintain all existing relationships of complementarity. Those changes which are necessary to rectify the inconsistent capital changes of the boom must not be expected to leave incomes and asset values intact.
It follows that any policy designed merely to restore the
status quo in terms of ‘macro-economic’ aggregate magnitudes, such as incomes and employment, is bound to fail. The state prior to the downturn was based on plans which have failed; hence a policy calculated to discourage entrepreneurs from revising their plans, but to make them ‘go ahead’ with the same capital combinations as before, cannot succeed. Even if business men listen to such counsel they would simply repeat their former experience. What is needed is a policy which promotes the necessary readjustments. No doubt, if the actual break-up of existing combinations with its consequences for control structure and portfolio structure can be avoided, readjustment will be much easier, but it is clearly impossible to maintain all those asset values which were based on inconsistent plans. What happens during a strong boom is that resources are being given a concrete form which, but for the misguided expectations of the boom, would not have been chosen. Somebody has to take the consequences.
We saw that even where the breaking-up of existing firms can be avoided, there will nevertheless have to be a break in structural complementarity. It follows that a policy endeavouring merely to ‘maintain effective demand’ by stimulating
consumption will simply defeat the very purpose of readjustment by making it profitable for those who should deflect the flow of their services elsewhere, to let them flow where they did before.
Thus far in this chapter we have been concerned with the strong boom. The strong boom is the result of plans involving inconsistent capital change, and this inconsistency is the result of the fact that where prices are inflexible they convey misleading information about available resources. As it is clearly impossible to have completely flexible prices in reality, or even an equal degree of flexibility throughout the economic system, investment decisions based on erroneous assumptions about the future availability of resources cannot easily be avoided. Even if all prices were completely flexible and sensitive to all present changes in demand and supply they would, in the absence of a fairly comprehensive system of forward markets which would make all relevant expectations consistent with each other, not necessarily reflect future scarcity of resources. Those who have to make investment plans might still be misled about the future availability of the resources they need if they relied exclusively on these present prices. The strong boom is thus an almost inevitable concomitant of an expanding industrial economy, and the system-wide regrouping of capital is its necessary consequence and corrective.
This does not mean that nothing can be done to mitigate it. There is one price in particular which, owing to its strategic importance, we should attempt to make as flexible as possible: the rate of interest, the general rate of exchange between present and future goods. When present goods are withdrawn from immediate use and turned into sources of future output, i.e. in the later stages of a strong boom, the rate of interest should rise. Any attempt to prevent it from rising means to spread misleading information about present and impending scarcities and future abundance. This is clearly seen on the Stock Exchange which discounts future yield streams on the basis of the present rate of interest. A sensitive and well-informed market witnessing the spectacle of a strong boom will of course in any case sooner or later have its misgivings about future yields and the cost of present projects. But we need not doubt that where this is not so, a rising rate of interest
would strongly reinforce the discounting factor and thus damp excessive optimisim.
Capital regrouping is thus the necessary corrective for the maladjustment engendered by a strong boom, but its scope is not confined to this kind of maladjustment. Where a weak boom has ‘petered out’ before ‘hitting the ceiling’ capital regrouping is just as necessary. That this is not at once obvious is due to the unfortunate habit of viewing all these problems in ‘macro-economic’ terms. If we come to believe that the only reason why expansion suffers a check is the combination of a weak accelerator with a feeble multiplier, we shall of course conclude that the only proper remedy is an increase in effective demand, and that it does not really matter how this is accomplished.
Even in a weak boom, however, some industries expand more than others, new products come into the market, consumers change their preferences, and in general, with or without full employment, some factors are scarcer than others. The new capital combinations will change the capital structure and the new products modify the market structure. Here again price inflexibility will for a time tend to hide the facts from the entrepreneurs, but the inconsistencies will show themselves in the end. This situation is best viewed in terms of Schumpeter’s model in which the ‘innovating’ new firms expand into ‘new economic space’ but also restrict the range of action of the older firms. As a result the latter have to adjust themselves to the new conditions, and this entails capital regrouping.
At the end of a weak boom the new capital resources begin to pour out output. Here there is no difficulty in obtaining and keeping together the complementary factor combinations since the ceiling has not been hit. But some firms may find it difficult to dispose of the new output. Prices will tend to fall, employment may decline and unsold stocks accumulate. Excess capacity (of the ‘real kind’!) may make its appearance. The notion that in such a case we could simply restore the
status quo by ‘maintaining incomes’ is, however, here just as futile as in the case of the recession following the strong boom. It is true of course that there is a danger that such an underconsumption crisis may degenerate into a cumulative
depression. If so, a budget deficit may help. But such a policy would have to be supplemented by strong pressure for the necessary capital regrouping to take place. Attempts, which are very likely in such situations, to ‘stabilize’ prices will, by reducing consumers’ real incomes, simply make adjustment more difficult. The best remedy for the excess capacity mentioned is to make it unprofitable for the owners of such resources to maintain them. Except in the case where there is excess capacity everywhere, the case in the contemplation of which the Keynesians specialize, existing capital combinations must be broken up and their fragments removed to wherever they are still useful. As in the case of the aftermath of the strong boom, a selective credit policy which reflects the degree of imperfection of the capital market, appears to be called for.
The significance of capital regrouping thus transcends the phenomena of the strong boom. But it also transcends the Austrian Theory of Industrial Fluctuations and its logical basis, the theory of capital we expounded in Chapter V. The Austrian theory, as most other models except Schumpeter’s ignores the effects of innovation and technical progress. It views economic progress primarily as taking place along the lines of ever greater division of labour and specialization of capital equipment, of ever higher degrees of complexity of factor combinations. But technical progress may cancel some of these effects by making some specialized skills and other specific characteristics redundant. ‘Technological Unemployment’ of highly skilled craftsmen is of course a well-known manifestation of this tendency. Technical progress, however, while making some capital equipment redundant, raises the demand for other types and their complements. All this simply means that in a world in which the forces of progress are manifold there are more, and not fewer, forces abroad which make the regrouping of capital an ineluctable task.
Technical progress means unexpected change. It is by no means the only form of unexpected change which entails modifications of the capital structure. In reality the capital structure is ever changing. Every day the network of plans is torn, every day it is mended anew. Plans have to be revised, new capital combinations are formed, and old combinations
disintegrate. Without the often painful pressure of the forces of change there would be no progress in the economy; without the steady action of the entrepreneurs in specifying the uses of capital and modifying such decisions, as the forces of change unfold, a civilized economy could not survive at all.
A Contribution to the Theory of the Trade Cycle, 1950, p. 108.
Ekonomisk Tidskrift, September 1950.
Quarterly Journal of Economics, May 1950.
Economic Journal, March 1954.
International Economic Papers, Vol. 3, pp. 75-171.
Economica, May 1940.
Economic Journal, March 1952, pp. 47-8 (by permission of the Royal Economic Society).
not a datum but a result of the market process.
Human Action, pp. 575-6.