Trust and accountability in times of pandemics

Trust and accountability in times of pandemics

Series: Working Papers. 2306.

Author: Monica Martinez-Bravo and Carlos Sanz.

Topics: Uncertainty | Health, education and welfare | Population and ageing | Productive sectors | Governance.

Full document

PDF
Trust and accountability in times of pandemics (1 MB)

Abstract

The COVID-19 pandemic took place against the backdrop of growing political polarization and distrust in political institutions in many countries. Did deficiencies in government performance further erode trust in public institutions? Did citizens’ ideology interfere with the way they processed information on government performance? To investigate these two questions, we conducted a pre-registered online experiment in Spain in November 2020. Respondents in the treatment group were provided information on the number of contact tracers in their region, a key policy variable under the control of regional governments. We find that individuals greatly over-estimate the number of contact tracers in their region. When we provide the actual number of contact tracers, we find a decline in trust in governments, a reduction in willingness to fund public institutions and a decrease in COVID-19 vaccine acceptance. We also find that individuals endogenously change their attribution of responsibilities when receiving the treatment. In regions where the regional and central governments are controlled by different parties, sympathizers of the regional incumbent react to the negative news on performance by attributing greater responsibility for it to the central government. We call this the blame shifting effect. In those regions, the negative information does not translate into lower voting intentions for the regional incumbent government. These results suggest that the exercise of political accountability may be particularly difficult in settings with high political polarization and areas of responsibility that are not clearly delineated.

Previous Do Renewables Create Local... Next A production network model...