No other event has affected the world economy during this century like the COVID-19 pandemic. Not even the financial crisis of 2008. And we still do not know the effects it will have in the medium and long term. In this context of negative economic indicators and extensive measures restricting mobility (which have generated relevant collective levels of anxiety, stress and even fear among the population and investors in general), one question stands out: will it affect the decision to vote for part of citizenship? The right to vote in a democracy is an individual act, with benefits and costs. If the former outnumber the latter, the person will see this act as a worthwhile course of action and will vote accordingly. But the pandemic may be increasing costs, as a result of the increase in the chances of contagion, so the result of this calculation may vary.
To answer the previous question, we propose an empirical exercise based on the regional elections of July 12, 2020 in the Basque Country. The coronavirus generated in this Autonomous Community more than 14,000 affected and 1,500 deaths until that moment, but the incidence was not the same in all municipalities. Specifically, of the 251 municipalities in the region, 220 (87%) have been infected, causing 115 of these (42% of the total) also deaths from coronavirus. Fortunately, in 31 (13%) municipalities there were no infections or deaths from coronavirus. This supposes a (quasi) natural experiment, which makes it possible to compare and identify how participation in the elections of these first two types of municipalities has varied with respect to the last ones, those in which there were no infections (which will mean the control group versus the than to compare the effects).
Maps: Municipalities with infected and deceased by COVID-19 in the Basque Country. July 2020
As we have anticipated, our objective is to evaluate how electoral participation varied, with respect to previous years and to municipalities not directly affected, as a result of the fear that citizens have of being infected by the disease. And that fear is more palpable the closer the voter is to the sources of infection.
To quantify this, we have created a database containing information on regional elections from 2005 to 2020 in the Basque Country. We propose a model that tries to estimate what factors explain that in a given municipality there is a level of participation based on sociodemographic variables, characteristics of the municipality and, of course, those related to the expansion of COVID-19 in the municipality. Applying a difference estimator (difference-in-difference), controlling previous trends and applying placebos to confirm robustness in the estimates, the results are as follows:
In the first place, in the municipalities where there have been cases of COVID-19, voter turnout fell by an average of 3.2 percentage points (always with respect to those municipalities where there were neither deaths nor infected, which are the control group) , which represents an average drop of 4.7% compared to the 2016 elections. And this reduction is greater the higher the rate of infected in the municipality.
Second, in those municipalities where there have been deaths from the pandemic, the effect on abstention is greater. Specifically, voter turnout fell by an average of 4.5 percentage points, which represents a decrease of 6.7% compared to 2016. As before, this effect is greater the higher the death rate in the municipality .
Ultimately, fear of voting caused a reduction in citizen participation. And this effect was much greater when voters took a closer look at the negative effects of the pandemic, especially due to the death of neighbors.
What can we conclude?
These results show the effect that the fear of being infected has had on participation in elections. Our results are consistent with a similar analysis of Fernández-Navia, Polo-Muro and Tercero-Lucas, who estimate a drop of 2.5-3.8 percentage points when measuring the effect of those infected on participation. However, we found a greater effect when also measuring the effect of the deceased on participation.
In any case, these increases in abstention have important implications for modern democracies, and may translate into losses of political inclusion, representation and legitimacy of democratic systems as a result of fear of contagion. Therefore, we could make three public policy recommendations: 1) increase security against contagion in voting centers; 2) use non-contact forms of voting (aware of the risks that these can present); and 3) reinforce measures to overcome the pandemic as soon as possible and prevent new outbreaks. The benefits of these measures transcend their economic value, since they also protect the health of democratic systems.
There are several electoral processes that are coming up soon (plebiscite in Chile, presidential in the United States, etc.) and that will be carried out at different stages of the pandemic in their respective countries. If the results of this work are maintained, abstention will be higher than in previous elections, especially in those cities where the coronavirus has had a greater incidence. Who will win for it? Democracy, surely not.
Note: The content of this article is part of the Final Master’s Thesis prepared by Miguel Vázquez-Carrero, defended in September 2020 at the Carlos III University of Madrid. A version of it can be found in the working papers of the journal Covid Economics (vol. 50), of the Center of Economic Policy Research.