Half-Life Estimation under the Taylor Rule∗

This paper addresses two perennial problems in the existing Purchasing Power Parity (PPP) literature, namely, unreasonably long half-life estimates of PPP deviations (Rogoff, 1996) and extremely wide confidence intervals for half-life point estimates (Murray and Papell, 2002). One answer to these problems can be found in the work of Kim, Ogaki, and Yang (2003), who used a system method and estimated much shorter half-lives than Rogoff’s “Remarkable Consensus ” of 3 to 5 year half life, though with limited efficiency improvements. Rather than following these authors in using a money demand function, we incorporated a forward-looking version of the Taylor Rule into a system of exchange rates and inflation. In so doing, we obtained a substantial improvement in efficiency, as well as reasonably short half-lives for PPP deviations. Our model also indicates that real exchange rate dynamics may differ greatly, depending on the pattern of systematic central bank responses to the inflation rate