Investigating the 'Bolsonaro effect' on the spread of the Covid-19 pandemic: An empirical analysis of observational data in Brazil

Abstract

Brazil counts among the countries the hardest hit by the Covid-19 pandemic. A great deal has been said about the negative role played by President Bolsonaro's denialism, but relatively few studies have attempted to measure precisely what impact it actually had on the pandemic. Our paper conducts econometric estimates based on observational data at municipal level to quantitatively assess the 'Bolsonaro effect' over time from March 2020 to December 2022. To our knowledge, this paper presents the most comprehensive investigation of Bolsonaro's influence in the spread of the pandemic from two angles: considering Covid-19 mortality and two key transmission mitigation channels (social distancing and vaccination); and exploring the full pandemic cycle (2020-2022) and its dynamics over time. Controlling for a rich set of relevant variables, our results find a strong and persistent 'Bolsonaro effect' on the death rate: municipalities that were more pro-Bolsonaro recorded significantly more fatalities. Furthermore, evidence suggests that the president's attitude and decisions negatively influenced the population's behaviour. Firstly, pro-Bolsonaro municipalities presented a lower level of compliance with social distancing measures. Secondly, vaccination was relatively less widespread in places more in favour of the former president. Finally, our analysis points to longer-lasting and damaging repercussions. Regression results are consistent with the hypothesis that the 'Bolsonaro effect' impacted not only on Covid-19 vaccination, but has affected vaccination campaigns in general thereby jeopardizing the historical success of the National Immunization Program in Brazil.

Fig 1. Growth in mortality, mobility and vaccination rates in Brazil.

Sources: Ministry of Health, Secretarias Estaduais de Saúde. Brasil. For the mobility data: Google mobility report (change relative to January 2020), for Bolsonaro’s statements: https://www.aosfatos.org/, Authors’ own elaboration. Note: Data considered are the seven-day moving average for each variable.

Fig 2. Covid-19 pandemic timeline in Brazil.

Source: Ministry of Health, Secretarias Estaduais de Saúde. Brasil, Authors’ own elaboration.

Fig 3. The ‘Bolsonaro effect’ on the mortality rate over time (2020–2022).

Sources: Ministry of Health, IBGE, TSE; authors’ calculations. Note: Model (1) (see Table 2, col.3). Confidence interval at p < 0.05.

Fig 4. The ‘Bolsonaro effect’ on the change in mobility over time (2020–2021).

Sources: Ministry of ealth, IBGE, TSE, Facebook Movement Range; authors’ calculations. Note: Model (2) (see Table 3, col.3). Confidence interval at p < 0.05.

Fig 5. The ‘Bolsonaro effect’ on the vaccination rate (2^nd dose) over time (2021–2022).

Sources: Ministry of Health, IBGE, TSE; authors’ calculations. Note: Model (3) (see Table 4, cols. 6a). Confidence interval at p < 0.05.

Fig 6. The ‘Bolsonaro effect’ on the vaccination rate (2^nd dose) by population categories.

Sources: Ministry of Health, IBGE, TSE; authors’ calculations. Note: Model (3) (see Table 4, cols.6a). Confidence interval at p < 0.05.