CROSS-CUT ANALYSIS: THE CAUSES OF
It would be very desirable at this point, before summarizing our conclusions as to the best method of forecasting, to present a statement of the causes of business cycles which would serve as a basis for rational forecasting by the method of cross-cut analysis. As we have indicated, we do not believe that forecasting can ever become an exact science until we get beyond empirical deductions from the sequence of past phenomena, and ground our analysis in a study of the effects which may be expected to flow from causes which we observe at work. An adequate cycle theory should be of great value for this purpose. The differences between cycles are great enough to make it impossible to fit them all into any simple frame-work of theory. But the resemblances are great enough so that we are unwilling to resign ourselves to a search for the causes of each individual fluctuation without a unifying theory.
Unfortunately, we cannot offer such a theory. During the past 15 years a host of books have been written concerning the cycle, and a score of theories of its causes have been advanced, but none seems at the same time thoroughly sound in logic, con-