The Methodology of Economists

Thirty-five years ago, E. C. Harwood, AIER's founder, wrote the following about the relation of economic conclusions to the methods of inquiry used by many economists: "That economists frequently do not agree has become so commonplace that some economists no longer seem to be troubled by the suggestion that such a state of affairs is scandalous. That many economists do agree on certain analyses and conclusions is equally scandalous from the viewpoint of modern science, however, because that agreement rests on methods of inquiry that have been found unreliable and have been discarded by capable scientists, the fact that a few conclusions on which some economists agree do have adequate scientific bases emphasizes by contrast the more fundamental disagreement among economists generally regarding the methods of inquiry that can be expected to yield useful results." During subsequent years, we have developed and refined our views about procedures of inquiry. Among their key aspects, as presently evolved, are the following: (1) An emphasis on the importance of "firm naming" in economic inquiry. Such naming, although firm in the sense that much terminology in developed fields is firm, is not final As inquiry progresses, improvement in naming is to be expected, just as occurred with the name "atom" in physics. Much of the conventional work in economics is plunged almost immediately into a semantic swamp by the use of naming that is inconsistent, incoherent, and frequently based on ancient epistemological ideas that long have been outmoded. (2) An emphasis on the entire pertinent transactional field, rather than on presumed separate and interacting "reals" making up that field. The "reality" of the various aspects and phases of the transaction generally depends on the field itself. For example, borrowing cannot exist without lending, and vice versa. Concentration on too limited an area of transaction is associated with much of the inquiry conducted by conventional economists. For example, Keynesian economists have tended to emphasize the volume of current consumption to the exclusion or minimization of its effects on future production and consumption. (3) An emphasis on the crucial importance of continuous testing of the conjectures (often called hypotheses) that are developed about what happens under specified circumstances by careful observation of available data and events. Conventional economic inquiry often proceeds by the development of elaborate conjectures far in advance of any significant testing of them or of their underlying "assumptions" against observed facts. Conventional economic inquiry relies on logical consistency, initial plausibility, assumed truisms, etc. and has resulted in elaborate "theories" or "models" in which the developers have great confidence. Unfortunately, the predictions based on such models often have been notably inaccurate, as subsequent events demonstrated. The facts and notions with which we begin an inquiry also may turn out to be inaccurate, which requires that assertions following therefrom must be carefully qualified. Inquiry, then, neither begins with certainty nor attempts to achieve certainty; rather, the objective is the development of warranted (but not final or certain) assertions adequate to solve the problem at hand. We also eschew the general procedure of inquiry, which has achieved dominance among academic economists, in which great emphasis is placed on the deductive and mathematical elaboration of a few "axioms." After the model is developed, supposedly it is to be tested against the facts, but in practice the major emphasis is placed on the "internal" deductive and mathematical elaboration. To summarize, we long have maintained that the "quest for certainty," whether through deduction, revelation, intuition, or any other means, has not been a reliable source of useful solutions to human problems. That policy recommendations based on the results of such procedures often have proved disastrous supports this view.