Date

2019

Department or Program

Economics

Primary Wellesley Thesis Advisor

Daniel Sichel

Additional Advisor(s)

Eric Hilt

Abstract

This thesis tests empirically whether the political preferences of Federal Open Market Committee (FOMC) members, indicated by party affiliation, the partisan direction of donations to political campaigns, and the party of the President affect their voting behavior when setting monetary policy. I use two main empirical strategies in this project. The first is a linear probability model that examines the correlation between a range of background characteristics of FOMC members--including political affiliation, educational attainment, and work background--on the probability of casting a dissent vote against the majority decision of the FOMC at a particular meeting.

The second approach controls for the state of the economy and focuses on whether an FOMC member’s vote on interest rates at a particular meeting was for an increase, a decrease, or no change. To control for the state of the economy and its effect on FOMC interest rate decisions, I use predictions from Taylor-like rules that translate measures of economic activity and inflation into prescriptions for interest rates. I then use a multinominal logit specification to assess how partisan affiliation (and several other factors) affect voting choices after controlling for the Taylor-Rule prescriptions. To implement both empirical strategies for this analysis, I constructed a unique data set that ranges from 1970 to 2018, where each observation is a person-meeting.

My somewhat surprising results indicate that partisanship emerges based on the party of the sitting President rather than through the party affiliation of FOMC members. In particular, during Republican Administrations, FOMC members downweight the signal from economic conditions when considering decreases in interest rates and also are considerably more likely to vote for rate decreases than is the case during Democratic Administrations. Additionally, I find that my bank president variable is no longer significant, which is surprising because the prior literature finds that bank presidents are hawkish.

Share

COinS