UC Berkeley

This dissertation consists of three chapters with a common theme of expectations and beliefs in the macroeconomy. The first chapter introduces a micro-level measure of consumer inflation uncertainty. Literature on cognition and communication documents that people use round numbers as a communicative tool to convey uncertainty. I construct an uncertainty measure that exploits consumers' tendency to round their inflation forecasts to multiples of five on the Michigan Survey of Consumers. I document cross-sectional and time series properties of the measure and provide support for its validity. Mean inflation uncertainty is countercyclical and positively correlated with inflation disagreement, inflation volatility, and the Economic Policy Uncertainty Index. Inflation uncertainty varies more in the cross section than over time, so a major benefit of this new measure is its cross-sectional dimension which enables micro-level analysis of the relationship between uncertainty and consumption. More uncertain consumers are more reluctant to spend on durables, cars, and homes, and their spending attitudes are less sensitive to interest rates. The measure also has applications to inflation dynamics and monetary policy. For example, the expectations of more-certain consumers can be used to improve Phillips curve estimation.

The second chapter focuses on central bank communication with households. Transparent communication with the general public is a stated goal of the Federal Reserve. While most research has focused on central bank communication with financial markets, this paper evaluates the effectiveness of Federal Reserve communication with the public at large. While professional forecasters are attentive to Federal Reserve communications regarding the price stability objective, including the announcement of a 2% inflation target, many households are not. Consumers' inflation expectations are weakly-anchored, especially among less-educated, low-income, and female consumers. Anchoring has not improved notably since the late 1990s. News and media data reveal that Federal Reserve communications are not widely propagated through traditional or new media channels to the public and that that consumers do not proactively seek information on monetary policy. Evidence collected from dozens of surveys from the 1950s to 2014 exposes a lack of public awareness of the Federal Reserve and its objectives and a decline in public opinion of central bankers.

The third chapter studies expectations in an important episode of economic history. Competing interpretations of the Great Depression depend on the behavior of inflation expectations in the onset and recovery. A number of papers have examined whether the deflation of 1930-32 was anticipated and when positive inflationary expectations reappeared. I review and compare the various statistical, narrative, and market-based approaches that have been used to estimate inflation expectations in the Great Depression era and supplement these approaches with additional methods and narrative evidence. I introduce a new approach using Phillips curve estimation. Reconciling the disparate findings of the previous literature, I conclude that the deflation was mostly unanticipated until mid-1930 and that a regime change occurred at the start of Roosevelt's presidency, prior to monetary expansion.