Alternative routes of adjustment of saving to investment in the long period

Introduction The recent period has witnessed a surge in the analysis of economic growth and development on neoclassical lines. New Growth Theory (Romer, 1987, 1990; Lucas, 1988) has proved influential both in theory and in practice (see e.g. Bardhan, 1995). One of its distinct differences from Old Growth Theory (Solow, 1956) is that in New Growth Theory different saving ratios affect the rate of output growth while in Old Growth Theory they do not influence anything other than the level of the accumulated capital stock relative to labour. By this is strengthened a typically neoclassical conclusion that the path of growth of an economy is causally affected by the saving behaviour of the constituent members of that economy; investment does not have a causal impact on economic growth and development, for investment is to adjust to saving for macroeconomic equilibrium. Economic growth is basically supply-led.² In stark contrast, the Keynesian conception reverses the causality of economic growth and development: in the long period as well as in the short period, it is saving that adjusts to investment, not the other way around: investment is determined autonomously, that is, independently of the saving corresponding to the capacity output; then, macroeconomic equilibrium requires saving to adjust to investment. Economic growth is demand-led (Setterfield, 2002). The Keynesian conception, however, has identified several alternative routes through which saving is generated to match investment in the long period. Two such routes are widely known, in reference to which different Keynesian positions are distinguished. The first position is usually called ‘Neo-Keynesian’, with Kaldor, Robinson and Pasinetti as its early representatives. This position holds that, in the long period, saving is brought into line with investment through variations in the state of income distribution. A higher level of investment causes a rise in the general price level so that wage earners, who are paid at a fixed money wage rate and have a lower propensity to save than profit earners, find themselves with a lower real wage rate while profit earners are consequently remunerated at a higher rate of profits; the resulting level of saving is higher corresponding to the higher level of investment. The second position is commonly called ‘Kaleckian’ and its early inspirations include Kalecki, Steindl, and Baran and Sweezy.³ The source of saving required to match varying levels of investment is varying levels of output ensuing from variations in the equilibrium degree of capacity utilization; in the long period, the degree of capacity utilization tends to settle at a certain level which is usually different from the given ‘planned’ or ʼnormal’ level.⁴ In this adjustment, the real wage need not, if it may, change while the rate of profits changes in accordance with the degree of utilization. This route of adjustment closely resembles the short period adjustment of saving to investment. There is, however, athird ‘Keynesian route’ of adjustment, which was identified some time ago but has not attracted as much attention as it deserves. This route is variations in the size of productive capacity and, consequently, in the volume of output (and, in principle, without changes in the relation between the real wage rate and the rate of profits). This route was brought to attention first by Garegnani (1962, 1979, 1982a, 1982b, 1992) and later elaborated by Ciccone (1984, 1986), Vranello (1985), Kurz (1990, 1994), Park (1997a), Palumbo and Trezzini (2003), and others. The present chapter aims to make an explanatory and constructive contribution along these lines. The next two sections provide brief criticisms of the two well-known Keynesian routes of adjustment of saving to investment. The following sections expound the third alternative route and provide a simple model which accords with that route. The final section concludes with a short comment. The Neo-Keynesian position The Neo-Keynesian mechanism of adjustment of saving to investment - the adjustment through changes in the distribution of income - is based on three characteristic assumptions. One is the differentiated propensities to save out of profits and out of wages.⁵ It is due to this differentiation in the propensities to save (such that the propensity to save out of profits is higher than that out of wages) that changes in the distribution of income manifest themselves in the form of the corresponding changes in the level of aggregate saving. Another assumption is that, in the long-period equilibrium the available labour force is fully employed and/or productive equipment is utilized at the normal level.⁶ It is implied, then, that the relationship among output, labour and the material means of production in the long-period equilibrium will be as specified by the dominant technique in use (see, e.g., Robinson, 1962, p. 11). Thus, except in the case of normal utilization of productive capacity alone (Joan Robinson’s ‘Limping Golden Age’ growth), the theory does not intend to explain the existence of the long-period, structural unemployment, i.e. unemployment due to the shortage of productive equipment to which labour is applied.