Inflation convergence in the EMS : some additional evidence : a comment

As in Caporale and Pittis, this paper finds significant evidence supporting the hypothesis of long-run equilibrium relationships between inflation rates in countries which participate in the ERM. However, the results differ in several important respects. First, the evidence rejects a dynamic specification in terms of inflation differentials against Germany and in at least one important case, Table 5, it is clear that imposing this restriction may lead to invalid inferences on the role of the ERM as a mechanism to achieve inflation convergence. Second, on the issue of German leadership the results given in Tables 4 and 5 suggest that the German inflation rate cannot be considered weakly exogenous. Rather it shares a long-run relationship with inflation in both ERM and non-ERM economies and responds to deviations from these equilibria. Finally, as these results also hold for a sample period twice the length of that used by Caporale and Pittis they cast considerable doubt on their assertion that cointegrating relationships are unlikely to be detected when “the dynamic process of convergence is still going on”.