Pitfalls of Forecasting

November 22, 1948

From time to time readers of this column suggest that I should forecast the future of business—tell whether we are headed for a boom or a bust, and exactly when it will come, how high or deep it will go, and how long it will last. I have refrained from such crystal gazing. The fate of the political experts who predicted the certain election of Dewey will help to explain why.

The economic future, like the political future, will be determined by future human behavior and decisions. That is why it is uncertain. And in spite of the enormous and constantly growing literature on business cycles, business forecasting will never, any more than opinion polls, become an exact science.

The reasons for this ought to be clear. The future of business never depends on any single factor. It depends on a combination of countless factors. None of us is near enough to omniscience to keep track of them all and to give each its exact weight. The typical professional business forecaster tends to assume that he can make a chart of past price movements, past volume of sales, past wage payments, or whatnot, and somehow pull out of it automatically the answer to the future. We ought to have learned by this time that he can’t. All statistics, even the most recent, necessarily refer to past situations. It is true that the past and the present, when carefully studied, can throw great light on future probabilities. That is how we must study them. But they can never tell us future certainties.

The factors from the past or present that we select to study are arbitrarily selected. To know even completely the situation about inventories, say, a few weeks ago, was not to know that Mr. Truman, and not Dewey, was going to be our next President. And to know completely the state of inventories today is not to know what Mr. Truman is now going to do, what Phil Murray is now going to advise or demand, what Stalin’s next move will be, or what will happen in Britain, France, or China. Yet such countless developments all go to determine the future state of business.

In brief, a full knowledge even of all business statistics put together cannot give the full answer to the business future. For the future of business is today affected even more by what happens “outside” of business than by what happens “inside.”

There is one more reason why business forecasting is so hazardous. When astronomers calculate the orbit and appearance of a comet, their predictions do not influence the comet. In human affairs, on the other hand, predictions about the future influence that future and change that future. It may have been precisely because Dewey’s political victory was so generally taken for granted that he was defeated. For it led him to make a disastrously wrong type of campaign and it led his supporters to relax their efforts.

In business, a general belief that there will be a rise in prices tends to bring about that rise. If, on the other hand, the business community were to become convinced that prices would be stable until next February and then start to break, it would start to act on that assumption immediately. Therefore the expected break would come not in February but right away. For what each of us is seeking as an individual consumer, speculator, or producer is not primarily to be right about the future but to be right sooner than anybody else. That is how exceptional gains are made. But that is something that can only be achieved exceptionally and individually, and not collectively.

I do not mean to disparage all efforts at business forecasting. All of us, no matter what our occupations, are forced constantly to make guesses about the future, and to act on them. We must try to make them as well informed as possible. But we will save ourselves from some of the worst jolts if we frankly recognize that our guesses are only guesses, and that we cannot reduce the human future to a scientific certainty.