Inflation, relative prices and nominal rigidities - BIS Papers No 19, part 3, October 2003

This paper examines the distribution of Belgian consumer prices and its interaction with aggregate inflation over the period June 1976-September 2000. Given the fat-tailed nature of this distribution, both classical and robust measures of location, scale and skewness are presented. We found a positive short-run impact of the skewness of relative prices on aggregate inflation, irrespective of the average inflation rate. The dispersion of relative prices also has a positive impact on aggregate inflation in the short run and this impact is significantly lower in the subsample starting in 1988 than in the pre-1988 subsample, suggesting that the prevailing monetary policy regime has a substantial effect on this coefficient. The chronic right skewness of the distribution, revealed by the robust measures, is positively cointegrated with aggregate inflation, suggesting that it is largely dependent on the inflationary process itself and would disappear at zero inflation. These results have three important implications for monetary policy. First, as regards the transmission of monetary policy, our results are in line with the predictions of menu cost models and therefore suggest that this type of friction can be an important factor behind the short-run non-neutrality of monetary policy. Second, as regards the design of robust estimators of core inflation, economic arguments based on menu cost models tend to highlight the importance of the absence of bias. We have proposed an unbiased estimator by taking the time-varying degree of chronic right skewness explicitly into account. Third, as regards the optimal rate of inflation, the chronic right skewness found in the data provides no argument against price stability, as it appears as an endogenous response of optimising price setters and would disappear when targeting a zero inflation rate. This conclusion contrasts sharply with the implications of the exogenously assumed downward rigidity of Tobin (1972), which would justify targeting a sufficiently positive inflation rate in order to facilitate the adjustment of relative prices. Our empirical findings contradict the latter type of downward rigidity, which implies a negative correlation between skewness and inflation. Therefore, the cross-sectional properties of Belgian inflation data do not provide strong arguments against a price stability-oriented monetary policy such as the one pursued by the Eurosystem.