The Nature of Persistence in Euro Area Inflation: A Reconsideration

Recent empirical studies find little evidence of a change in euro area inflation persistence over the post-1970 period. Their methodology is primarily based on standard unit root and structural break tests on the persistence parameter in an autoregressive specification for the inflation process. These procedures are, however, not designed to detect a change in persistence when a sub-sample of the data has a unit root, i.e., when the process shifts from stationarity to non-stationarity or vice-versa. In this paper, we use four classes of tests for a change in persistence that allow for such shifts to argue that euro area inflation shifted from a unit root process to a stationary one at some point in the sample. Statistical methods to select the break date identify the change in the second quarter of 1993, around the time of the Maastricht Treaty which established the groundwork for the European Monetary Union, with an explicit mandate for price stability as the primary objective of monetary policy. Bootstrap estimates of the persistence parameter, half-life estimates and confidence intervals for the largest autoregressive root all suggest a marked decline in persistence after the break. We also illustrate that the hypothesis of stationarity with a mean shift but a stable persistence parameter is not compatible with the data. The evidence presented is therefore consistent with the view that the degree of inflation persistence varies with the transparency and credibility of the monetary regime.