When Guidance Changes: Government Inconsistency and Public Beliefs

Abstract

Governments often provide inconsistent guidance to the public. Does inconsistency affect how much people believe the latest recommendations? Using an incentivized online experiment with 1,900 US respondents in early April 2020, we present all participants with the latest official projection about coronavirus death counts. We randomize exposure to information that highlights the government’s inconsistency about the coronavirus threat. When inconsistency is salient, participants have reduced propensity to revise prior beliefs about death counts and lower self-reported trust in the government. These results align with a simple model of signal extraction from government communication, and have implications for the design of changing guidelines in other settings.

Keywords: signaling, Bayesian updating, coronavirus

JEL Classification: D78, D83, D91, H12, I18