Reverse Capital Deepening

It has long been taken for granted that there is an inverse monotonic relationship between the rate of interest (or the rate of profit) and the quantity of capital per man. This belief was founded on the principle of substitution, whereby ‘cheaper’ is substituted for ‘more expensive’ as the relative price of two inputs is changed. In the field of capital theory, the principle of substitution persuaded many economists, such as E. von Böhm-Bawerk (1889), J.B. Clark (1899) and F.A. von Hayek (1941), that a lower rate of interest (which is equal to the rate of profit in equilibrium) is associated with the use of more ‘capital intensive’ techniques, and thus with the substitution of capital for other productive factors, such as labour or land. This process is called capital deepening. Recent discussions have shown that this is not necessarily true, since a lower rate of interest might be associated with lower, rather than higher, capital per man. This phenomenon is called reverse capital deepening. The paper discusses the relevance of reverse capital deepening and its implications for the traditional view that technical choice is a monotonic function of the rate of interest . It is maintained that there is as yet no full agreement as to the main consequences of the discovery. For example, Christopher Bliss has noted that reverse capital deepening makes it impossible to see the accumulation of capital as a process associated with ‘a continuous increase in consumption per capita,…a continuous decline in the rate of interest and …a continuous increase in the real wage rate’ (Bliss, 1975, p. 279). He also called attention to the fact that the ‘extended accumulation history’ of an economic system moving through real time is normally different from the hypothetical history we can tell by ‘travelling’ across steady states (Bliss, 1975, pp. 194 and 280-1). In particular, he emphasized that the statement that the rate of interest may be expected to fall as capital deepening takes place ‘cannot be interpreted’ in the case of extended accumulation history, as we would have, in that case, ‘a whole structure of interest rates…not a single rate of interest’ (Bliss, 1975, p. 294). A different point of view has been expressed by Pierangelo Garegnani, who has maintained that capital paradoxes in general, and reverse capital deepening in particular, by making traditional beliefs untenable, suggest a ‘correction’ to traditional theory, which would make it reasonable to expect ‘instabilities or tendencies to zero of wages, or of net returns on capital’ (Garegnani, 1990, p. 76). This author’s view is that, rather than introducing such a correction, one should drop any idea of a causal connection from marginal products to the distribution of the social product, and further develop the conjecture that distribution is brought about by ‘more complex economic and social forces like those envisaged by the old classical economists’ (Garegnani, 1990. p. 76-77). As it is common in theoretical sciences (see Kuhn, 1970, 2000), the discovery of an apparent anomaly has induced economists to look for a more general theory, or to switch to an altogether different framework.