Throughout the 1950s and 1960s, the predominant Keynesian view was that of the Neoclassical-Keynesian Synthesis of enriched Hicks-Modigliani IS-LM models. The Pigou Effect led to a great debate in monetary theory, but as the debates had led, effectively, to the conclusion that the Pigou Effect had to work on a very narrow band of assets, it was presumed that, even if it existed, the power of the Pigou Effect could be empirically ignored. The conventional wisdom was that the Neoclassical model was true in the long-run, the Keynesian model true in the short run when there are sticky wages, liquidity traps and interest-insensitive investment - all of which, it was argued, were reasonable assumptions about the real world, thus "let us get on with policy". Yet, because of this, the theoretical validity of the Keynesian system was disabled as it seemed that unemployment could not maintain itself in a "perfectly working system" as Keynes (1936) had envisaged and instead relied on some imperfection or other. In order to restore its theoretical validity, more faithful Keynesians would have to at least remove the "sticky prices" qualification. Important in this regard, as insisted upon famously by Abba Lerner (1952), was to prove that high price flexibility, far from being stabilizing and returning the economy to full employment, was in fact, "destabilizing" and maintained or worsened unemployment. The challenge was thus laid down for the Keynesian faithful: if that could not be proved, then Keynes's system was truly of no theoretical importance. The point of departure was Keynes himself. In Chapter 19 of the General Theory, Keynes confronted the "Keynes effect" which, recall, argued that declining money wages and price levels would reduce money demand and thus interest rates thereby taking the economy to full employment. However, Keynes did not ignore the inherent heterogeneity of agents in the economy. As he notes falling money wages and price levels will lead to redistributions of income - firstly, from wage-earners to non-wage earners and, secondly, from debtors to creditors, the net effect of which would be the reduction in the economy-wide marginal propensity to consume (Keynes, 1936: p.262). The income redistribution effect was taken up by Michal Kalecki (1939: Ch. 3; 1942) in an imperfect competition model . Specifically, if money wages decline, then the mark-up between prices and wages would increase (i.e. increasing "degree of monopoly" in Kalecki's language). This would result in a redistribution of income from wage-earner to profit-earners. If profit-earners have a lower propensity to consume than wage-earners (a reasonable assumption given social reality), then the average marginal propensity to consume in the economy declines and thus aggregate demand declines. Thus, far from being stabilizing, the reduction in money wages in a situation of unemployment can lead to reductions in aggregate demand and thus more unemployment. This type of relationship between aggregate demand and income distribution was pursued by Cambridge Keynesians such as Nicholas Kaldor (1956) and Joan Robinson (1962) and, along somewhat different lines, Sidney Weintraub (1958, 1965), Kenneth Boulding (1950) and Frank Hahn (1950, 1951). The second effect was already expressed in Keynes (1931) and in Irving Fisher (1933) and is known as the "Debt-Deflation Effect". Effectively, if prices decline, then the real value of private wealth increases, which implies that the liabilities of debtors and assets of creditors increase in real terms. Again, complying to social reality, debtors have a higher marginal propensity to consume relative to creditors (which might help explain how they became debtors in the first place!), consequently, there is a reallocation of real wealth from debtors to creditors and thus the aggregate marginal propensity to consume declines. The result of price flexibility in situations of unemployment, then, is once again a decline in consumption demand and thus a further reduction in aggregate demand and employment. This Debt-Deflation Effect was given a central role by James Tobin (1980), J. Caskey and Steve Fazzari (1987) and Thomas Palley (1996). A third effect was also proposed by Keynes: namely that "if the reduction [in money wages and prices] leads to the expectation, or even to the serious possibility, of a further wage-reduction in prospect.. [then] it will diminish the marginal efficiency of capital and will lead to the postponement both of investment and of consumption" (Keynes, 1936: p.263). This includes what has since become known as the "Tobin-Mundell effect", in reference to the analysis of James Tobin (1965) and Robert Mundell (1963) . Effectively, if prices decline, then by a form of adaptive expectations, there will be expected future deflation. Consequently, when making a portfolio allocation between money and capital, expected deflation will lead to a rise in money demand. This will shift the LM curve to the left - and thus reduce output and employment. The destabilizing influence of the Tobin-Mundell effect is laid out in James Tobin (1975, 1992), J. Bradford de Long and Larry Summers (1986) A fourth effect proposed by Keynes is that deflation and the consequent rise in the debt burden will both depress the "animal spirits" of indebted businessmen if not wholesale bankruptcy outright (Keynes, 1936: p.264). This possibility, which is also related to the debt-deflation argument, has been most fully explored in the Hyman Minsky (1975, 1982, 1986). Minsky proposes that firms issue debt to finance production in the expectation of future profits. During booms, the debt burdens of firms increase and therefore firms become highly leveraged with relatively lower and lower quality debt. The system becomes, in Minsky's language, "financially fragile" in the sense that small shocks that could otherwise be absorbed (e.g. small deflations) can have mass repercussions, depending on the amount of debt outstanding. The amount of bankruptcies this causes will immediately reduce aggregate demand and thus a whole new bout of deflation and bankruptcies. This "debt-bankruptcy" spiral, originally suggested in Keynes (1931), is but one of the various Minskian variations on a quintessentially Keynesian theme. A fifth effect, more implicit than explicit in this portion of the General Theory, is the existence of "endogenous money". This old concept was revived in the work on financial intermediation by John G. Gurley and Edward S. Shaw (1960) and James Tobin (1963), and was a central feature of the Monetarist controversy (esp. Nicholas Kaldor, 1970, 1982). It has since become a standard feature of Post Keynesian economics, such as in Hyman Minsky (1982), Basil Moore (1988), Randall Wray (1990), Marc Lavoie (1992), Thomas Palley (1996) and others. If money supply is endogenous, then the money supply function is "horizontal" (or nearly so) and not vertical. Consequently, reductions in the money wage can very well lead to reductions in money demand as hypothesized by the Neo-Keynesians but, with money supply endogenous, there will be a corresponding reduction in money supply so that, in effect, there will be no resulting fall in interest rates. As a consequence, the "Keynes Effect" is eliminated completely and the economy will stay at unemployment equilibrium. Finally, a sixth effect was proposed by Abba Lerner (1936, 1939, 1944, 1952) which was quite ingenious. Namely, if all factors have highly flexible prices, it becomes effectively impossible for money wage reductions to increase employment. The basic notion is a kind of macroeconomic "non-substitution" idea: suppose that there is a decline in wages, then firms will try to get rid of machines and substitute towards labor. But as the prices of machines (and all other factors) are highly flexible or at least more flexible than money wages (a credible assumption), then these prices will collapse immediately - to the point where it is no longer desirable to make the substitution. Thus, Lerner concludes, there will be no substitution between labor and other factors if there is high price flexibility - consequently, lower money wages create no incentive for firms to employ more labor. Paradoxically, Lerner suggests, high (but not perfect) price flexibility is consistent with - nay, necessary for - Keynes's theory of unemployment equilibrium to hold and it is only when prices are rigid or sticky or sluggish, that Keynes's theory begins to unravel. Lerner's argument relies in good part on an effective demand constraint in a general equilibrium model and thus can be regarded as an early expression of what later became known as the "disequilibrium" approach to Keynesian theory most closely associated with the research programme of Robert Clower (1965), Axel Leijonhufvud (1967, 1968), Robert J. Barro and Herschel Grossman (1971, 1976) that had a brief but brilliant life in the late 1960s and early 1970s..
|
All rights reserved, Gonçalo L. Fonseca