Essays on the dynamics of inflation expectations

Abstract

This thesis investigates the dynamics of inflation expectations with a particular focus on survey data. It aims to further the understanding of what drives inflation expectations and what are the implications of changes in inflation expectations for economic choices. The first chapter examines to what extent monetary policy moves household inflation expectations. More specifically, I study the effect of different types of monetary policy announcements on household inflation expectations based on micro data from a survey of German households. As unique feature, interviews of the survey were conducted both shortly before and after monetary policy events. This timing provides a natural experiment to identify the immediate effects of policy announcements on household inflation expectations. In contrast to most existing studies, the availability of the survey over a period of 15 years also allows me to exploit the time-series dimension to estimate how policy announcements affect household inflation expectations over the medium-term. I find that policy rate announcements lead to quick and significant adjustments in household inflation expectations with the effect peaking after half a year. Announcements about forward guidance and quantitative easing, on the other hand, have only small and delayed effects. My results suggest that monetary policy announcements can influence household expectations but further improvements in communication seem to be necessary to reach the general public more effectively. In particular, in an environment where policy rates are constrained by the effective lower bound, it may be very hard for central banks to influence household expectations. In the second chapter, joint with Evi Pappa and Alejandro Vicondoa, we focus on expectations about inflation in the medium to long run and study the implications of changes in these expectations for households’ economic choices. We identify in a SVAR shocks that best explain future movements in different measures of underlying inflation at a five-year horizon and label them as news augmented shocks to underlying inflation. Independently of the measure used, such shocks raise the nominal rate and inflation persistently, while they induce mild and short-lived increases in economic activity. The extracted inflation shocks have differential distributional effects. They increase significantly and persistently the consumption of mortgagors and homeowners. Differently from the traditional monetary policy disturbances, news augmented shocks to underlying inflation induce a positive wealth effect for mortgagors and homeowners, driven by a reduction in the real mortgage payments and a persistent increase in real house prices that they induce. The third chapter, joint with Jonas Fisher and Leonardo Melosi, is also about long-run inflation expectations but in this case the focus is on professional forecasters. We use panel data from the U.S. Survey of Professional Forecasters to estimate a model of individual forecaster behavior in an environment where inflation follows a trend-cycle time series process. Our model allows us to estimate the sensitivity of forecasters’ long-run expectations to incoming inflation and news about future inflation, and measure the coordination of beliefs about future inflation. We use our model of individual forecasters to study average long-run inflation expectations. Short term changes in inflation have small effects on average expectations; the sensitivity to news is over twice as large, but is still relatively small. These findings provide a partial explanation for why the anchoring and subsequent de-anchoring of average inflation expectations over 1991 to 2020 were such long-lasting episodes. Our model suggests coordination of beliefs also played a role, slowing down but not preventing the pull on average expectations from inflation running persistently below target. We apply our model to the case of a U.S. central banker setting policy in September 2021. Our results suggest the high inflation readings of mid-2021 would have to be followed by overshooting of the Fed’s target generally at the high end of the Fed’s Summary of Economic Projections to re-anchor long term expectations at their pre-Great Recession level.