Inflation and output: a review of the wage-price mechanism
BIS Economic Papers No 24
Since the late 1970s a medium-term strategy has formed the basis for macroeconomic policies in most industrialised countries. This strategy has four components:
- a reduction in fiscal deficits and in the government share of output
- a shift in the distribution of factor income in favour of profits
- removal of rigidities in labour and product markets
- relatively tight monetary policies to reduce and then stabilise inflation at a low rate.
A key assumption behind the medium-term strategy is that there is no long-run trade-off between inflation and the level of output. This assumption is also known as "the long-run neutrality of money" or the "natural rate hypothesis" (NRH). Thus expansionary monetary policies may temporarily increase output and lower unemployment, but the economy will eventually return to the natural rate of unemployment at a higher rate of inflation. Conversely, deflationary policies will over a certain period reduce output as well as the rate of inflation, but in the end only the reduction in inflation will remain as the economy again returns to its natural rate.
So far the medium-term strategy has been a success in many respects. Government deficits have been reduced and profit shares are in many countries back to pre-1973 levels. Moreover, the restrictive monetary policies adopted almost universally have contributed to the largest deceleration of inflation in the post-war period and have also helped to maintain inflation at a low and stable rate for the last two to three years. Yet in two respects the strategy has been disappointing. The fall in inflation was accompanied by very large output losses and the response of inflation to the policies adopted was subject to long lags. Secondly, in Europe and, to a lesser extent, Japan there have so far been no signs of a return to a natural rate of unemployment. As can be seen from Graph 1, the NRH seems to hold for the United States, as the policies adopted in 1979-80 initially reduced inflation as well as employment, but by 1987 the rate of unemployment had returned to its initial level and during 1988 it fell below what many previously considered to be the natural rate. However, in Europe and Japan it is difficult to find evidence of the NRH. Obviously, in both cases the rate of inflation has been substantially reduced, but the rate of unemployment has stabilised at a level which is considerably higher than in 1979-80.
When attempting to explain these disappointing aspects, it is natural to look to the wage-price mechanism in industrialised countries and this is the main topic of this paper. Following a brief review of short and long-run trends in output, inflation and money supply growth in Section I, Section II surveys major models of the wage-price mechanism. In particular, it discusses those theories which can explain why a change in monetary policy is not immediately reflected in a parallel change in the rate of inflation but is accompanied by real output effects, which may last for four to five years. Section III turns to the empirical evidence with respect to the parameters of the wage and price equations which are responsible for the persistent output effects of monetary policy changes, while Section IV surveys various hypotheses concerning the particular shape of the European Phillips curve. Finally, Section V deals with some unresolved issues and attempts to draw conclusions and policy implications.
In focusing on the wage-price mechanism, the paper leaves out a number of subjects which are also closely related to the problem of inflation. It does not discuss the longer-term aspects of an ongoing and steady rate of inflation, such as whether it is entirely neutral and whether the optimum rate is positive, zero or negative. Moreover, the paper has little to say on the causes and effects of inflation, and the formation of inflationary expectations is treated only superficially. The paper also neglects several international aspects of inflation, especially the role of exchange rate movements. Export and import prices are mentioned as components of aggregate price equations and as possible sources of supply shocks, but the transmission of inflation between countries is mentioned only briefly and hypotheses regarding the "world rate of inflation", which were very popular some years ago, are not discussed. Finally, many technical aspects of the wage and price adjustment process have been left out in order not to overburden the text with equations and mathematical proofs. Readers interested in a more rigorous presentation and discussion of the various issues are referred to two recent surveys (Blanchard (1987b) and McCallum (1987a)l from which many aspects and ideas in the following text have been taken.
Because of these limitations, this paper is by no means a survey of inflation. It is rather an attempt to review certain aspects of the inflation process which are thought to be of interest to policy-makers concerned about the real and nominal effects of their actions and about the process by which policy changes are transmitted to the "real" economy.