By the mid-1920s, the deductive method had long since become the accepted mode of inquiry for discovering laws relating to the behavior of market phenomena. There was little concern about reinforcing deductive analysis with empiricism beyond the casual sort that used actual (or conjectural) data for purposes of example and illustration. Wesley Mitchell’s work at Columbia University and at the National Bureau of Economic Research in business cycles, along with Irving Fisher’s development at Yale of index numbers to measure price-level changes, were exceptions to the preference of most economists in the United States toward deductive economics. Marshall’s Principles and his strong reservations about the application of mathematical methods to economics influenced most economists to rely on deductive analysis for their own work, and teach it to their students. Thus, mathematics and statistics existed as disciplines that remained quite separate from economics

The creative spasm of theoretical innovation known as “high theory,” which dates from the mid-1920s, also encouraged new methods for studying the behavior of the economy, though these were not primarily mathematical or statistical. Concern about cyclical phenomena and the usefulness of the ex ante-ex post

construct of the Stockholm School are among the intellectual breakthroughs of the period. Unlike the neoclassical concept of equilibrium, which focused on the requisites for an economy’s return to stability, the ex ante-ex post construct offered a way of conceiving of an economy in the process of changing from one phase of the business cycle to another.1 Once suggested, this idea implied the need to invent a method to evaluate the relative merits of one plausible cycle theory as opposed to another equally plausible theory. The challenge was taken up by the League of Nations, which commissioned Jan Tinbergen, a Dutch scholar, to evaluate their relative merits empirically. Jointly, with the Norwegian Ragnar Frisch, he became the 1969 recipient of the Nobel prize in economics. Tinbergen’s 1939 statistical verification of alternative business cycle theories, which pioneered the method of least squares and regression analysis, marks the beginning of econometrics as the sister discipline of economics. It also marks the beginning of the third and present stage of numeracy, in which economics has emerged as a predictive rather than as a moral science.