Recent Developments in the Mainstream
Following the contributions by Hicks and Modigliani, the IS-LM model became the shared theoretical framework for all mainstream macroeconomic debates. Although quite heated and sometimes involving strong divergences concerning the appropriate short-run macroeconomic policy, these debates never addressed the theoretical foundations of the model and the associated tendency towards full employment (or towards the equilibrium unemployment rate determined by frictions and imperfections) in the long run.
Debates initially centred on the elasticities of the IS and LM curves and hence the respective impact of monetary and fiscal policies as countercyclical policy tools. With the rise to dominance of monetarism the focus of debates has become the degree of flexibility of money wages and prices and their speed of adjustment (hence, the speed of adjustment towards potential output), which in turn depends on how expectations are formed. The possibility of “Keynesian” outcomes in the short run (that is, a role for aggregate demand in the determination of output and employment) has therefore come to depend on the assumptions about expectations. Fluctuations in output and employment, however, were still seen to depend on aggregate demand shocks, and hence the models required real wages to move counter-cyclically to remain consistent with decreasing marginal product for labour and profit-maximizing firms. Yet the labour market side of these analyses has encountered problems on empirical grounds. A number of empirical investigations have confirmed, with greater subtlety and sophistication of analysis, the results originally obtained by Dunlop and Tarshis, that is, that real wages are either moderately procyclical or exhibit no consistent pattern over the cycle (see Brandolini 1995, for a survey). These results may be regarded as one of the factors that favoured the emergence and acceptance of real business cycle and new-Keynesian macroeconomic models.The peculiarity of real business cycle models is that the labour market is continuously in equilibrium, so that the very concept of unemployment disappears from the analysis: cyclical fluctuations in the employment level reflect short-run voluntary changes in labour supply. These are the response of inter-temporally maximizing workers to shortrun changes in real wages determined by (temporary) supply shocks, the nature of which is not usually discussed in any depth, but are generally associated with technical changes. While long-run labour supply is generally admitted to be inelastic with respect to permanent changes in wages, short-run changes in labour supply in these models are described as highly responsive to wage changes perceived as temporary, owing to the fact that agents maximize utility over the life-cycle and hence substitute labour and leisure inter- temporally. This could in principle render real business cycle models consistent with the observation that employment fluctuations are usually large, while real wages tend to be moderately pro-cyclical. Empirical analyses however, perhaps not surprisingly, have not confirmed the above short-run behaviour of labour supply (Altonji 1982, among others; see also Romer 2012: 228, 254 5).
On the other hand, the so-called new-Keynesian strand in mainstream macroeconomics purports to provide sound micro-foundations for real and nominal price rigidities which can explain, given the accepted theoretical framework, the existence of persistent involuntary unemployment and the short-run role of changes in aggregate demand in determining output and employment fluctuations. In their labour market analyses in particular, new-Keynesians have provided a large (perhaps too large) number of alternative or complementary models which could explain rigidity in real wages, thus explaining the possibility of some persistent, structural unemployment, the size of which ultimately depends on the institutional arrangements concerning unemployment benefits and employment protection.
At the same time they have tried to provide foundations for price and/or money wage rigidities that could account for the temporary effects of aggregate demand on employment levels. With respect to consistency with observed facts, these models have the advantage that, since some structural unemployment exists, the short-run increase in labour supply does not require a movement along a standard labour supply curve (and hence strongly pro-cyclical real wages). Conversely, the assumption that firms do not necessarily change the price level over the cycle (even though marginal costs are changing) means that (in the short run) they are not moving along their labour demand curves (which would require countercyclical movements of real wages). The new-Keynesian approach therefore seems to have the advantage of greater flexibility in the analysis of the short-run and in rendering the models consistent with empirical observations of labour market and macroeconomic behaviour, though at the cost of losing clear theoretical foundations, particularly with regard to the theory of income distribution. In these models the macroeconomic equilibrium real wage turns out to be determined, not only by the marginal labour product, but also by the markup of the representative firm determined by the elasticity of its product demand curve, thus overlooking the input-output interdependence between firms and hence between prices and costs in the system as a whole. Labour market institutions in turn determine the equilibrium unemployment rate, that is, the unemployment required to keep the real wage at a level consistent with marginal labour product and the given mark-up.During the late 1970s and the 1980s another approach to employment theory was influential, particularly in Europe, based on the introduction of wage and price rigidities in general equilibrium models. Within this approach, Malinvaud’s “equilibria with rationing” proposed a taxonomy according to which unemployment could be either “Keynesian” or “Classical” (Malinvaud 1977).
In the first case, price rigidity determines “low” real wages and households’ consumption rationing which in turn causes underutilization of existing capacity and unemployment. Classical unemployment, on the other hand, features full utilization of capacity, with the latter unable to employ the available labour force. This situation is generally associated with exceedingly high wages, which discourage investment. The wage policies appropriate to deal with these two types of unemployment are therefore completely opposite, while in any given economy both may coexist in different industries or regions. Malinvaud’s approach was criticized by some (mostly mainstream economists) on the grounds of the arbitrariness of the assumption of fixed wage and prices, while others (mostly non mainstream economists) expressed dissatisfaction both with the interpretation of Keynes and with the notion of classical unemployment and its roots in neoclassical assumptions (Kahn 1977; Schefold 1983).