§ 9. INTRODUCTORY.

In the preceding chapters an attempt has been made to acquaint the reader with the general phenomenon of crises and cycles, both from a morphological and from an historical approach. Our next and infinitely more difficult task must be to find out the cause or—what even prima facie seems more probable—the causes, of the said phenomenon. This is not only in order that we may satisfy our scientific curiosity, but also that we may discover, through an understanding of the causation, what policy should be followed in order to control and moderate the driving forces behind economic fluctuations and disturbances.

The history of doctrines which have attempted to clear up the mystery of recurrent disturbances of economic equilibrium is now more than one century old, Jean Baptiste Say and David Ricardo being perhaps the first to grasp the real problems involved. But though present-day economic science may be said to have made great advances towards a better and deeper understanding of the intricate questions to which the study of economic disequilibrium gives rise, and of the working of the mechanism of our economic system under the influence of disturbing factors, it would be unduly optimistic to assume that there has been found a solution of the problem of the causation of crises and cycles which leaves nothing to explore or which is satisfactory from all points of view. As in every other branch of economic science, a communis opinio doctorum on the origin of crises and cycles cannot be expected, and even on some of the most important points serious divergences of opinion still exist, though with a fair chance of being diminished as time goes on.

On the other hand, however, the theory of crises and cycles has that also in common with economic theory in general that the extent of actual agreement is much greater than is generally supposed by those who are not thoroughly acquainted with the subject, and who often betray their very ignorance by censuring economic science for hopeless divergences of opinion. It would be easy to show that the science of medicine gives the appearance of a still wider disagreement on very important questions of pathology and therapeutics, but every medical man will tell us that the bitter warfare in his science is being conducted on a rather broad groundwork of universally accepted principles. More often than not, it is not the principles in medicine as in economics that are at stake but the conclusions to be drawn from them, their arrangement and combination, the emphasis to be placed on the different factors, the interpretation of new experiences, and the expediency of practical measures—though, in fairness to the outside observer, it has to be conceded that the embittered spirit in scientific warfare must sometimes give the impression that there are unbridgeable abysses hopelessly separating the opponents. Finally, there is the human element which accounts for the fact that the real differences of opinion are often much exaggerated, common vanity and pugnacity playing a prominent part. It is all too common that an economist—and the writers outside of academic circles are sometimes even worse—wants to make himself immortal by presenting a brand-new theory of cycles and making room for it by abusing all the others and emphasizing the differences rather than the similarities. In direct contrast to these tendencies, it is the special aim of the present author to keep watch for every possible reconciliation and to endeavour to combine elements from different and even opposed schools of thought. This is not to be done in a spirit of a kind of scientific pacifist who shuts his eyes to secret armaments, nor is the author unaware of the danger of easy compromises. What he is striving after is a synthetic attitude rather than an eclectic one. This is, he believes, an urgent necessity in present circumstances.

It would at first seem to be an attractive idea to start the analysis by giving a synoptic account of the doctrinal history of the theory of crises and cycles. On second thoughts, however, this idea loses much of its attractiveness. In order to be of any value, such a synopsis must be more extensive than is compatible with the scope of this book.¹ But even if this difficulty were overcome there would still arise the more serious difficulty of how to arrange the different theories in accordance with their most significant characteristics, which is essential if the synopsis is not to be merely a pointless chronology. A systematic classification of the theories of crises and cycles is difficult to attain because almost every theory so far developed, save the most eccentric ones, contains some elements belonging to another type of explanation so that an attempt at systematic classification inevitably leads to numerous overlappings and cannot do justice to the theories in question. To quote only one instance, it is almost impossible to draw a clear dividing line between the under-consumption theories and the over-capitalization theories, there being a large margin of overlapping, and even the division between monetary and non-monetary theories, though indubitably a very important one, has many loopholes. This accounts also for the fact that none of the several attempts which have been made at systematic classification of the theories of crises and cycles seems really satisfactory.

This is true even of the most successful attempts at classification among which those made by Wesley Mitchell and Alvin H. Hansen are especially noteworthy. Mitchell² divided the different theories into three large groups: firstly, theories which trace crises and cycles to physical processes (cycles of crop yields, &c.), secondly, theories which relate crises and cycles to emotional processes (especially with mass alternations of optimism and pessimism), and, thirdly, theories which connect crises and cycles with institutional processes, i.e., with factors arising out of the structure and the institutions of our present economic system. The large third group is then subdivided into a great number of subordinate groups with different points of view. Hansen,³ while clearly demonstrating the difficulties and shortcomings of any classification, distinguishes (1) the Capitalistic Economy schools, which stress either the capitalistic process of distribution or the capitalistic process of production; (2) the Exchange Economy school, which considers crises and cycles to be a function of the competitive exchange economy and points particularly to recurrent errors of judgment (the psychological school in Mitchell’s classification), and (3) the Money Economy school which believes that the root of crises and cycles lies in the monetary and financial mechanism of our economic system, a school which is, broadly speaking, equivalent to what is more generally known as the monetary theory of crises and cycles.

In view of these difficulties, the present writer has preferred to forego a classified presentation of theories and theorists, and to adopt a different approach. The programme of this chapter is to examine all the elements and circumstances which offer any chance of explaining the origin of crises and cycles and to set aside for a final analysis those elements which, in the process of sifting, prove useful in this respect. In the last two paragraphs, special attention will be given to the problem of the origin and course of the present crisis. Following this programme, it is to be hoped that the analysis may seem less scholastic than is all too often the rule in these matters and may seem more realistic without losing its rigidly scientific character.

There remains the cumbersome but indispensable task of dealing with certain methodological questions which the problem involves. Among these the most important and also the most controversial question is to ascertain whether the method of deductive analysis or that of empirical description offers the better chances of bringing us nearer to a solution of the problem.

While the older literature on crises and cycles was principally based on deductive reasoning, though of necessity starting from the facts presented by the actual history of economic fluctuations, there has sprung up, in the last decades, a new school which claims, emphatically and with ill-concealed scorn for “theoretical” analysis, that the collection of statistical data, their intelligent arrangement, and their mathematical analysis constitute the only method appropriate to the problem. This method is sometimes spoken of as “quantitative analysis” (in contrast to the “qualitative” analysis of the deductive school), as “descriptive analysis” (Mitchell) or in the German literature, as “Konjunkturforschung” pure and simple (in contrast to the “Konjunkturtheorie” of the deductive school). Its general background is always the curious but ineradicable misconception of scientific methodology which was the characteristic of the late Historical school and which has been evolved further in its modern reincarnations, especially by the Institutionalist school in the United States.⁴ The recrudescence of the fundamental error inherent in this whole methodological trend is closely connected with the activities of the numerous research institutes which have been founded in the last decade in the United States and also in many other countries, with a view to obtaining a better diagnosis of economic conditions and their fluctuations and also, in the eyes of the more optimistic workers in this field, a prognosis of changes in economic conditions. These institutes—among which the Harvard Committee of Economic Research, the National Bureau of Economic Research in New York (with Wesley C. Mitchell as one of the directors), and the Institut für Konjunktur forschung in Berlin (with E. Wagemann as president) are the most noteworthy⁵—have been developed into large economic laboratories where, with a most refined apparatus of mathematical distillation, the raw material of statistical facts is worked up until we get the finished products in the form of all kinds of “cycles,” of “coefficients of correlation,” of “trends,” of “moving averages,” of logarithmic scales, of business “indices” and “barometers,” of “isolated time series” and what not. It is easy to understand that, in this atmosphere of institutionalized science with its air of exactness and progressive technique, not many escape the temptation to look down upon old-fashioned “theory,” with its small business unit of the private study, as something hopelessly behind the times and to claim for their descriptive and statistical method a wider field of application than is compatible with the logic of scientific methodology. Instead of confining themselves to the more modest but very meritorious task of collecting, arranging, and interpreting all the available facts and statistical data and of verifying the results of deductive reasoning, these men thought it possible to replace deductive reasoning entirely by their empirical and statistical method. Deductive theory, essentially centring around the static conditions of economic equilibrium, was declared to be totally unsuited to explain the disturbances of this equilibrium; the new “Quantitative Economics” was claimed on many sides, as the only suitable method for complying with the desideratum of Dynamic Economics. Other writers of this same school—for instance, Wagemann and to some extent also Mitchell—go so far as to deny that the search for causal relations in crises and cycles is a legitimate object of economic science.

Now, there cannot be the slightest doubt that the development of statistical research work, with its refined methods of observation and interpretation of the facts of cycles and crises (“Konjunkturforschung”), has been, in its proper field, a great advance, and there is nobody who does not appreciate very highly the work done by the several research institutes. The picture of the phenomenon of crises and cycles given in the second and third chapters of this book has been largely based on this work. The empirical basis on which the theoretical analysis of crises and cycles must necessarily be constructed has been broadened by it in a manner which leaves less room for mere guesswork and vague hypotheses, and also there is no disputing the educative influence emanating from the work of the empiricists on the mental sobriety of the theorists. On the other hand, however, it is equally indubitable that the scope of the work that can be usefully done on these lines is rather more limited than was believed in the early enthusiasm of the ’twenties. The present crisis, which has reduced so many extravagant expectations to more reasonable proportions, has also exploded the more far-reaching hopes of the cycle statisticians. The attempts at prognostication especially have lamentably failed as was most conspicuously demonstrated by the breakdown of the “Harvard Barometer.” Since this is a striking example of the inherent defects of this kind of work, it may be useful to explain that the diagnosis and prognosis given by the Harvard Committee was based on the combination of three time-series the temporal order of which was supposed to follow a regular course so that, at a given moment, the point in the business cycle corresponding to the constellation of the curves at that moment could be determined.

It was indeed an ingenious idea to apply the principle of nautical astronomy to economic forecasting, but there was one fatal flaw. For as long as we have not made a thorough investigation into the causal relationships between the time-series, the mere temporal sequence does not tell us any more than that something has happened in the past which might not happen in the future if some variables in the causal mechanism should change. But in investigating the causal relationships we are thrown back from statistical empiricism to “theory” in the deductive and analytical sense. Hence this whole procedure of the empiricists falls little short of the reasoning of the legendary army doctor who, being informed that a foot-soldier sick with typhoid fever had quickly recovered after having eaten by mistake a liberal portion of sauerkraut, decided that sauerkraut is a cure for typhoid fever. And many hasty attempts at introducing some sort of theory, after the flaw in the empirical procedure has been detected, come quite near to the reasoning of the same army doctor who, after having applied his recipe on an unfortunate lancer with fatal effect, corrected his first deduction by saying that everything depends on the kind of arms. A great deal more could be said on this subject to show that the whole idea of forecasting, whatever its methods, is based on a number of misconceptions which go far to rule it out as a useful undertaking.⁶ Even the accomplishment of the more modest task of economic diagnosis encounters great difficulties. The errors in diagnosis and prognosis made by the Cycle Institutes in the last five years have, indeed, been so many that it is really to be wondered whether the world would not have been better off without these more ambitious activities, considering the harm done in economic policy by misleading diagnosis and forecasting. These experiences are closely connected with methodological misconceptions about the merits of statistical empiricism and the demerits of analytical theory. It can be safely assumed that, to-day, there is perhaps a better understanding everywhere of the fact that no intricate statistical research work, no calculations of “coefficients of correlation,” “trends,” and “moving averages” can bring us a single inch nearer to the solution of the problem of the causation and of the mechanism of crises and cycles. By the statistical method, we ascertain facts, but we cannot explain them, i.e., bring them into logical order so that we “understand” them. Only analytical theory can do that, and if there has been, in recent years, any furthering of our insight into the mechanism of crises and cycles, this has been the work of the theorists and not of the empiricists.

While it can be safely assumed to-day that the legitimacy of the method of analytical theory is hardly contested any longer (with the possible exception of very hard-boiled Institutionalists in the United States), there still continues a lively discussion about the real nature of the problem involved in the phenomenon of crises and cycles and about the type of theory appropriate to it.⁷ For all our aversion from every kind of “prolegomenism” and from philosophizing endlessly about tools without really getting to work, it is well to be clear on two points : first, that the phenomenon of general economic disequilibrium as revealed by crises and cycles constitutes a problem unexplained by general economic theory which is essentially a theory of economic equilibrium and of the forces working interdependently for its continuous restoration, and, second, that, in spite of this, there cannot be any doubt that the problem constituted by the recurrent disruption of the economic equilibrium must be tackled by the same type of analysis as that applied to the problems of economic equilibrium. To decry this type of economic analysis as “Static Theory,” and to clamour for a “Dynamic Theory” instead, is to confuse a question of method with a question of subject and to misconceive the real nature of analytical economic theory which must of necessity be the same whether applied to questions of Statics or to those of Dynamics. The real task and purpose of the theory of cycles and crises, therefore, is to modify and enlarge the scope of general economic theory so as to include the explanation of the recurrent disruptions of economic equilibrium. Since practical economic life has been up till now continuously caught in the ups and downs of disrupted equilibrium, the theory of crises and cycles is of special importance in making the whole body of economic theory more realistic. Such a theory, it must be repeated, is not a new and independent kind of theory with methods of its own, but a part of the general body of economic theory, based, as a kind of superstructure, on the theory of economic equilibrium and modifying or enlarging its general propositions. It must be, in the celebrated words of Böhm-Bawerk, “the last, or the last but one, chapter of a written or unwritten system of Social Economics and the ripe fruit of the knowledge of all economic processes and their functional connections.” It must be noted, however, that in order to harmonize with the general body of economic theory, the theory of crises and cycles must be able to explain in a really convincing manner why the very strong forces always working for equilibrium are recurrently counteracted so that the whole system of balancing factors breaks up. Not all theories so far advanced have proved able to do this, for they point to disturbing factors which, in the light of general economic theory should be easily digestible by the economic system. It is well to remember this point in the subsequent sections.

Something more must be said in order to describe more exactly the real nature of the theory of crises and cycles. Mention has already been made in § 4 of the fact that the scientific analysis of economic fluctuations began with the phenomenon of crises and only at a comparatively late date was it realized that the theory of crises must be incorporated into the more general theory of cycles, Juglar being the outstanding pioneer in spreading this new conception. It must not be thought, of course, that the earlier writers up to the last quarter of the nineteenth century were ignorant of the phenomenon of cyclical fluctuations; it was only that their scientific curiosity was predominantly directed towards the more dramatic phases of the crisis and the depression. The obvious reason for this narrowness of viewpoint lies in the fact that the boom phase of the cycle, characterized as it is by full employment of the factors of production and by an altogether healthy aspect of economic life, gives the outward appearance of perfect economic equilibrium—except for the last stage of hectic speculation which aroused enough suspicion even among the earlier writers. Hence it was generally thought that this phase was just that phase which corresponds to Equilibrium Economics, without there being anything peculiar about it demanding a special explanation, and again from a practical point of view there seemed nothing pathological about the boom phase to give rise to concern and uneasiness. The only trouble was that it did not last; it did not only just come to an end but was usually followed by the worst kind of economic disequilibrium. This had to be explained, but it was only by a slow process that the origins of the crisis and of the depression were traced to the preceding boom and thus the real pathological character of the latter discovered. In all this, the attitude of the earlier writers corresponds closely to the attitude of the general public in these matters : the boom is generally taken as something normal and enjoyable with nothing the matter with it, and when the crisis breaks out the causes are sought in all directions save for the one direction in which they should be sought, i.e., in the unhealthy and abnormal conditions created by the boom. The regular temporal sequence of boom and crisis (or depression) experienced during one century of capitalistic development has, of course, done much to shake this attitude and to suggest that the temporal sequence implies a causal one.

There remains one final question. Is it really true that the phenomenon of crises and cycles is a general one like the phenomenon of price or interest, or is every crisis and every cycle just an “historical” event more or less unique in its kind, with a kaleidoscopic change of constellations? If the latter should be the case, then a special theory of crises and cycles has evidently no real raison d’être since nothing useful could be said on crises and cycles in general; every crisis and every other phase of the cycle would create a new concrete situation which would have to be analysed with the tools of general economic theory. This is, in fact, a view held by several writers.⁸ The present writer, however, is not prepared to accept it. Though he would concede that many cycle theories have the defect of making too sweeping generalizations, it seems that the radical view referred to overlooks the fact that all crises and cycles of our economic system can, if their ever-changing accessory traits are eliminated, be reduced to a uniform pattern of typical constellations of data. The common denominator of all crises and cycles is to be found in the typical reaction of the mechanism of our economic system to certain disturbing factors by which the economic equilibrium is totally upset. This typical reaction is such a regular occurrence that it calls for a special explanation, which it is the task of the theory of crises and cycles to give. We may conclude, then, that this last and perhaps most serious attack on the legitimacy and necessity of the theory of crises and cycles can also not be said to have been really successful.