Abstract
This article constructs and estimates a measure called perceived inflation persistence that can be used to determine if professional forecasters’ inflation forecasts indicate there has been a change in inflation persistence. This measure is built via the implied autocorrelation function that follows from the estimates obtained using a forecaster-specific state-space model. Findings indicate that U.S. perceived inflation persistence has changed since the mid-1990s with more consensus among forecasters at lower levels of persistence. When compared to the autocorrelation function for actual inflation, forecasters typically react less to shocks to inflation than the actual inflation data would suggest.
ACKNOWLEDGMENTS
The author is very thankful to Gregor Smith for his guidance, patience, and supervision throughout this article. The author is grateful to two anonymous referees and an associate editor for very constructive comments. The author also thanks Michael Boutros for skilled research assistance and Allan Gregory, Allen Head, Jonathan Wright, Nicolas Vincent, Rodrigo Sekkel, Norm Swanson, Brigitte Desroches, Antonio Diez de los Rios, Kim Huynh and participants at the CEA Conference (2010), the Midwest Macroeconomics Meetings (2011), the Bank of Canada Seminar Series (2011), the Atlanta Fed Seminar Series (2013), the St. Louis Fed Applied Time Series Econometrics Workshop (2013), and the European Economics Association Meetings (2014) for helpful comments. The views expressed in this article are those of the author and do not necessarily represent those of the Bank of Canada.