In early neoclassical theories, the distributive variables were understood as the prices firms have to pay for the employment of the factors of production. Income distribution was thus conceived as a market phenomenon: as the relative prices of commodities are determined by the equilibrium between their supply and demand, so income distribution comes out of the equilibrium between supply of and demand for factors of production.

The rate of interest was accordingly intended as a variable reacting to discrepancies between supply of and demand for capital, in more or less the same way as the price of a commodity reacts to the difference between its supply and demand. Therefore, the equilibrium level of the rate of interest was thought to involve the equality between the quantity of capital demanded by firms and the stock of capital supplied by households. Moreover, since the quantities of available labour and land were considered exogenous magnitudes within the theory, similarly, the existing stock of capital was understood as a given amount. More recently, the given stock of capital that characterized the initial versions of the neoclassical theory has been interpreted as due to a ‘missing equation’ in those equilibrium systems. In particular, re-reading those early attempts from a neo-Walrasian standpoint, various scholars identified the missing equation with a condition of zero net savings which is required by the stationarity of the system and which relates intertemporal households’ decisions about current and future consumption with firms’ choices of the optimal production plans.

In the paper, we set out to address the issue of the neoclassical conception of the market for capital, with its given existing stock, in reverse order. We shall start from the neo-Walrasian stationary equilibrium model in order to use it as a benchmark in the analysis of the early equilibrium models. In particular, we shall consider both a Wicksellian model, in which the existing stock of capital is taken in value terms, and Walras’s theory without and with capital formation, in which, instead, the endowments of capital goods are included among the data on the same footing as the endowments of non-producible inputs. The comparison with the neo-Walrasian stationary model will allow us to analyse the logical working of Walras’s and Wicksell’s theories, which therefore are not merely considered from point of view of the history of economic thought, but also from a methodological perspective.