Tinbergen's Cycle: An Arithmetic Error? - TINBERGEN'S CYCLE: AN ARITHMETIC ERROR?

Abstract

Because of one algebraic mistake in the works of Tinbergen, Neuberg comes to the conclusion that the pre-Haavelmo research ultimately failed to understand changes in business variables as cyclic harmonic motion and Morgan does not convey the failure. What went wrong in the pre-Haavelmo research, or in Tinbergen's work, or in Morgan's reconstruction of it? As a historian of science, let me unscramble some confusion. The point is that the mistake did not play any significant role in Tinbergen's argument and therefore did not supply, as Neuberg thinks, any ultimate failure in the research program. Tinbergen's program for business cycle research of the 1930s can be summarized as follows: Tinbergen was in search of an endogenous theory of the business cycles. The core of the business cycle theory was the mechanism. By a mechanism, he meant both a closed system of equations (of which at least one must be dynamic) and the eventual economic meaning of each of those equations. Both aspects were relevant for the explanation of the business cycle-only together did they form a genuine business cycle theory. A mechanism served as an explanation for a business cycle when it generated a cycle with the same period as the business cycle itself. Although Tinbergen used the insights gained from this research in the development of his model of The Netherlands (1936) and of the United States (1939), they were not meant to be business cycle theories (or schemes) but, literally, models of the Dutch economy and the economy of the United States. The Dutch model was built in response to a request from the Dutch Economic Association to present a paper on the question,