From the 1900s to the 1940s: disagreements and uncertainties

The period from 1900 to the 1940s witnesses the rise to practically complete dominance of the marginalist approach; but it also witnesses the paradox of economists who agree on interest as the reward, determined by a supply-and-demand equilibrium, of the factor “capital”, but are unable to agree on how to make the idea rigorous; each different view is sharply criticized by others.

There is almost general agreement that the quantity of capital must be measured as a quantity of value; but the Clarkian conception of value capital as perfectly analogous to labour or land is criticized not only by Bohm-Bawerk (1907), who argues that it is indispensable to refer to the delay between inputs and output and to time preference to understand why advanced labour has a marginal product that pays not only its wages but also interest; it is also and more radically criticized by Veblen (1908: 160-67) who questions its value measurement noting that prices are not given. The given endowment of value capital is admitted to be problematical by Wicksell too, as we saw; a radical rejection comes from Irving Fisher who argues that the value of capital derives entirely from capitalization of its prospective returns, and that the rate of interest is determined, not by equality between supply and demand for capital, but by equality between supply and demand for savings, in turn determined by “impatience” and by the existence of “investment opportunities”, which he treats as determinable independently of the rate of interest, as if there were just one aggregate commodity “income” that can be consumed or “transferred” across periods by being employed in investments that produce “income” in subsequent periods. In some passages Fisher admits the influence of the rate of interest on relative prices but (without proof) dismisses this effect as not affecting his conclu­sions.

Disagreements are strong also on the period of production, which is generally rejected (even, as we saw, by Wicksell) because of needing simple interest; but the idea that decreases of the rate of interest will induce the adoption of in some sense “longer” or “more roundabout” production processes is widely accepted: it is found, for example, in Fetter, Marshall, Hicks, and Kaldor. However, Knight criticizes this idea in a series of articles in the 1930s. He is answered by Hayek (1936) with an attack on all conceptions of capital as a single factor, a “fund”, the Clarkian conception as well as the Bohm- Bawerkian one, all accused of needing the measurement in terms of value and therefore of suffering from logical circularity; capital, Hayek argues, can be described only through a complete enumeration of all physical capital goods. Knight (1936) replies that one does not need the “fund” capital to determine interest, which can be determined at the margin of new investment; but in so doing he follows Fisher in implicitly taking prices as given in determining the return to marginal investment; thus he too falls under the objection advanced, for example, by Lutz (1956 [1967]) that given prices imply a certain rate of interest, and therefore a determination of the rate of interest as the rate of return on marginal investment taking both present and future prices as given entails circularity. Actually Knight still believes in capital as the value factor, as shown, for example, by his 1946 entry “Capital and interest” in Encyclopaedia Britannica.

The uncertainties are confirmed by a tendency to limit the presentation of formalized general equilibrium theory to acapitalistic production. This is what Pareto does in the Manuel (1909) where he drops Walras’s equations of “capitalization” he had initially accepted, without replacing them with anything else. Marshall too presents only (parts of) the acapitalistic general equilibrium model in the Mathematical Appendix of the Principles; Stackelberg, Zeuthen, and Wald, who in the 1930s try to give a more rigor­ous proof of existence of solutions to the general equilibrium equations, again limit themselves to the acapitalistic model. Apparently there is faith that the approach can be extended to include capital and interest, and therefore a limitation of the analysis to the acapitalistic model is not perceived as a decisive deficiency; but no one seems to know how rigorously to perform the extension.