The Empirical Status of Business-Cycle Theory

Abstract

M UCH criticism has been directed against business-cycle theorists for their failure to test hypotheses with available empirical data. In part, tests are not made because they always tend to be inadequate; but often even refutable hypotheses are not tested. As Friedman expresses it, Because we cannot adequately test the validity of many hypotheses, we have fallen into the habit of not trying to test the validity of hypotheses even when we can do so. We examine evidence, reach a conclusion, set it forth, and rest content, neither asking ourselves what evidence might contradict our hypothesis nor seeking to find out whether it does.' basic significance of such tests to the further development of business-cycle theory has also been emphasized by Tinbergen: The business cycle is largely a measurable phenomenon, showing itself in the movement of many economic variables. It is a coherent movement but not rigorously cyclical. A large number of theories aim at explaining its nature. Most of those current in economic textbooks and literature run in qualitative terms. need for testing these theories is now generally felt, because of their multiplicity, their mutual inconsistency, and their incompleteness or indeterminacy from the quantitative point of view. Testing will hence be of great significance, both for our factual knowledge of economic fluctuations and for the further development of the theories concerned.' Before additional tests of businesscycle theories are made, however, it may be profitable to assess the current status of empirical research in business fluctuations. But it must be evident from the scope and breadth of this task that a thorough evaluation is not possible in a brief survey. This paper brings together various approaches which have been made to the testing of hypotheses, along with results and critical comment on both results and methodology. paper is developed in three sections. Section I summarizes certain past results of the testing of business cycles, along with some critical comment. Section II describes various projects currently under way and provides a brief evaluation of each. Finally, Section III suggests measures designed to increase the usefulness and significance of quantitative-historical research in business fluctuations.3