Uncertainty of inflation and inflation rate: Does credibility of inflation policy matter?

Abstract

We subject country level data to causality tests within a panel cointegration framework, to examine the relationship between inflation rates and the uncertainty of inflation for a panel of twenty two emerging economies, over the time period 1968 to 2010. Using the GARCH framework, a time series of conditional variances of inflation has been generated to serve as proxies for uncertainty of inflation. This series has been utilised to determine whether and how the relative degrees of central bank independence (CBI) influence the relationship between inflation rates and the uncertainty of inflation across the selected countries. Results of cointegration tests in the total sample confirm a robust long run equilibrium relationship between inflation and the uncertainty of inflation. For causality tests, although the long run outcomes are similar, short run results show some differences along the lines of the degree of CBI. For high CBI countries, the causal link in the short run is unidirectional and supports the Friedman-Ball hypothesis. By contrast, the low CBI countries display bidirectional causal links in the short run and provide support for both the Friedman- Ball and Cukierman-Meltzer hypothesis. In the long run, the entire panel and the two subgroups of countries all provide significant evidence affirming the Friedman-Ball hypothesis. The unambiguous conclusion of the empirical investigation suggests that the Friedman-Ball hypothesis is a long run phenomenon, regardless of differences in degrees of central bank independence.