Research Discussion Paper – RDP 12 Testing Two Theories of Business Fixed Investment


The principal aim of this paper is to test two alternative theories of gross business fixed investment: a neo-classical theory; and an accelerator hypothesis. Equations based on these theories are developed for two categories of gross private fixed capital expenditure in Australia: “other new building and construction”, which it is convenient to refer to as construction investment; and “all other”, which is referred to as equipment investment.

Lags between investment expenditure and changes in its determinants are given particular attention. Two alternative methods are used to estimate these lags: the Almon variable technique; and the rational lag distribution method.

The plan of the paper is as follows. The two theories of net investment are examined in section 1. A theory of replacement investment is presented in section 2. The two theories of net investment and the replacement theory are drawn together in section 3 to form the two theories of gross investment. Some comments are made in section 4 on the two methods used to estimate the lag distributions. The notation and sources of the data are given in section 5. The regression estimates for the equations for equipment investment and construction investment are reported in sections 6 and 7 respectively. The results of the study are summarised in section 8.