Die verschiedenen Konzepte des Akzelerators

Abstract

In theoretical analysis the accelerator concept is attractive because it can explain changes in investment which do not run parallel to demand. Empirical tests of the accelerator usually show unsatisfactory results. This is partly due to the specification used for the estimation: the relation between capital stock and production (the first level of estimation) is not popular because of the time trend in the variables; instead frequently a regression of investment expenditure on the change in production (second level of estimation) is applied. But because of the cyclical variations in the capital-output ratio this specification is not a good substitute for the first one. This paper shows that with the help of the relation between capital stock and output (“naive” accelerator) the highly different phases of investment activity can be explained, but that the individual annual observations follow another set of rules. Within the “flexible” accelerator the attemps that use only flow magnitudes do not represent a further development, but rather a counter-hypothesis of the original accelerator concept: flow magnitudes can explain medium-term variations only to a very small extent. This is contrary to the accelerator concept which explains considerable investment variations with the help of relatively small output variations. Attempts which explain investment variations through a connection between output and capital stock, expand the accelerator theory in so far as they take account of disequilibrium situations. The econometrically estimated equations give an excellent picture of short-term and medium-term investment activity. The economic interpretation of this equation shows, however, that in addition to the explicitly used variables (capital stock and output) other magnitudes contribute implicity to these excellent results. The merit of the accelerator theory then is not to be seen in a complete “explanation” of investment behavior, but rather in the fact that it shows highly variable investment demand for capacity purposes which again is dampened by other factors in such a way as to exhibit medium-term variations in capacity utilization.