Concerning Economic Prediction

      AN INTRODUCTION

      It is the business of science to predict. An exact science like astronomy can usually make very accurate predictions indeed. A chemist makes a precise prediction every time he writes a formula. The nuclear physicist advertised to the world, in the atomic bomb, how man can deal with entities so small that they are completely beyond the realm of sense perception, yet make predictions astonishing in their accuracy and significance. Economics is now reaching a point where it can hope also to make rather accurate predictions, within limits which this study will explain.

      In these pages we shall be primarily concerned with a new approach to economics and the problem of economic forecast, with the near-term future of the United States particularly in mind. This approach moves partly through some avenues that in the past have been the province of other sciences as various as biology, psychology, and mathematics.

      The study here falls into two parts. First, it shows that rhythm and periodicity exist in the natural world, and that our economic world, analyzed with similar statistical tools, also displays curvilinear forms and distinct rhythms. Second, it deals with some of the ideas which underlie these facts, suggests implications which seem safely implicit in them, and indicates some meanings which such facts hold for all of us.

      The debt of the authors to those whose names, equations and graphs line the pages of this book — and to many others unnamed — is without end. Theirs is the pioneering that is moving economics out of the blind alley where it stood for many years, so that it can take its rank as a true science.

      There are those who, admitting that economics has not been an exact science, also insist that it cannot be, in the sense of predicting outcomes in human affairs. There are even some who consider prediction regarding human life as a kind of impiety — or fakery, at best.

      That prediction regarding human affairs so often stands in ill repute with sober men (regardless of whether it “comes true” or not) often stems from techniques used in formulating it. It is less the forecast than the questionable methodology that often lacks scientific credibility.

      The reader will be introduced to a method of thinking about the future which — new though it may be to him — seems definitely to have proved of value. It is this method which is of fundamental importance — an importance greater than any specific conclusions to which it may lead. For on its validity depends the whole value of the conclusions.

      We shall find, as we go forward, that in this approach to economic phenomena we are abandoning the classical approach based on endless argument over cause and effect. It is hoped that the reader’s reward will be the discovery that in economics, as in other sciences, we are apparently dealing with laws regarding rhythmic human response to certain stimuli that give a remarkable working tool to any man who is responsibly concerned with future outcomes — whether he be businessman, community leader, or statesman.

      Law in nature, of course, is not the kind of law that is handed down by an authority. It is merely a summary of our observations concerning what has consistently happened, and what we may therefore expect to continue in a consistent universe. Such law permits us, in effect, to assay certain probabilities. The ability to calculate probabilities is a vital part of all our modern scientific progress. As Eddington has pointed out, in speaking of his own field of physics, “The quantum physicist does not fill the atom with gadgets for directing its future behavior, as the classical physicist would have done; he fills it with gadgets determining the odds on its future behavior. He studies the art of the bookmaker, not of the trainer.”*

      [* A. S. Eddington, The Nature of the Physical World (p. 301), The Macmillan Company.]

      Probabilities have to do with averages. When the physicist predicts that a given group of atoms will act in a certain way, he is relying — as physicists now believe — on a knowledge only of what the average atom will do. Quite similarly, when an insurance company more or less accurately predicts the number of deaths for next year, it is relying on statistical averages of a like kind.

      The discovery that the law of averages applies to humanity — that certain activities of people, viewed en masse, fall into definite patterns, some of which repeat themselves with periodic rhythm — promises to be of great aid in making economics function as a true science.

      These patterns will not tell us what any given individual will do — any more than laws in physics will tell what a particular atom will do, or life insurance statistics will reveal whether a given individual will die. But the patterns do reveal how masses are likely to act at given times. And to that extent they are a formulation of law.

      This discovery has had to wait upon the development of the science of statistics, the invention of index numbers, and the compilation of statistical series over long periods. Today the work is only beginning.

      Adequate statistics have not been kept for more than a few generations. Thus we are like doctors following Pasteur, who had incontrovertible evidence that germs exist, but in only a few instances had isolated the germs involved in particular human states.

      To assume that from our present limited store of knowledge we can henceforth predict the course of life for humanity would be childish. But we already have discoveries of importance. As the community becomes better informed of their value it will doubtless help us to extend them. Most important, to whatever extent the limited knowledge we do have can be put to practical use, it would seem that now is the time of times to bring it forth.

      To be completely exposed to surprise