The Fisher effect in the presence of time-varying coefficients

Abstract

A resolution of the Fisher effect puzzle in terms of statistical inference is attempted. Motivation stems from empirical evidence of time-varying coefficients in the data generating process of both the interest rates and inflation rates for 19 OECD countries. These time-varying dynamics crucially affect the behaviour of all the cointegration estimators considered, especially in small samples. When employing simulated critical values instead of asymptotic ones, the results provide ample evidence supporting the existence of a long-run Fisher effect in which interest rates move one-to-one with inflation rates in all countries under scrutiny except for Ireland and Switzerland.