Endogenous Benchmarking and Government Accountability: Experimental Evidence from the COVID-19 Pandemic

Abstract

Can citizens hold governments accountable for their management of the COVID-19 pandemic? While benchmarking theories argue that cross-national comparison in the media can provide the necessary information, we relax the assumption that people are exogenously exposed to the same benchmarks. Integrating politically selective news exposure into the theory captures a key countervailing mechanism. Using the pandemic as an important test case, we embedded survey experiments capturing self-selection and exogenous exposure to factual information in representative surveys in France, Germany, and the UK. We find that exogenous benchmarks have consistent effects on individuals' evaluations. But when given the opportunity to select the benchmark, there is significant self-selection based on respondents' pre-treatment political views. Importantly, self-selection is not complete and in a second experiment we show that self-selected types do not differentially respond to information. Altogether, our results suggest that self-selection limits, but does not break, benchmarking as an instrument of accountability.

Keywords: democracy, accountability, cross-national benchmarking, COVID-19

JEL Classification: D72, D91, H12