Polarized Expectations, Polarized Consumption

This paper argues that political polarization plays a key role in shaping the economic expectations and consumption behavior of households. Using a combination of survey and consumption data of U.S. households, we document five facts. First, household beliefs are well-described by a single factor, which behaves like sentiment. Second, at any given time there is wide dispersion in household sentiment, largely driven by political affiliation. Third, household sentiment is highly persistent, with one exception: following elections when the White House switches parties, optimistic households become pessimistic and vice versa. Fourth, the magnitude of this switching behavior has increased over time. Fifth, consumption responds differentially along party lines following changes in the White House. We show that standard theories of expectation formation struggle to simultaneously rationalize these facts