Compulsory Voting Diminishes the Relationship Between Winning and Satisfaction With Democracy

Compulsory voting is used in over two dozen countries and regions. Chile recently reinstated mandatory voting after just over a decade without it. In Belgium, Flanders will soon begin conducting municipal and provincial elections on a voluntary basis after generations of compulsory voting. In the United States, compulsory voting bills have been introduced California, Connecticut, Massachusetts, New York, and Washington state—all in the past four years.

It is clear from decades of research that compulsory voting increases turnout, and a burgeoning literature has begun shedding light on the downstream effects of a requirement to vote. In my recent contribution to the JOP, I add to this literature by offering a theory of compulsory voting’s moderating impact on the well-known link between winning elections and satisfaction with democracy (SWD).

The relationship between electoral victory and SWD is thought to be driven by a policy mechanism and an emotional mechanism. Per the policy mechanism, people become satisfied because their party will enter government and form laws. According to the emotional mechanism, people become satisfied because winning affects feelings and emotions that shape democratic attitudes.

I argue that compulsory voting weakens both mechanisms. Starting with policy, those who are externally pressured to the polls are relatively unlikely to base their votes on political ideology or parties’ issue positions, and anticipated legislation thus matters relatively little. Consider an analogy from gambling. If an enthusiastic bettor has a large amount of money riding on a particular result, their level of satisfaction with the game will depend heavily on the outcome. Alternatively, if a casual gambler has only a trivial sum at stake, their degree of satisfaction will be relatively unaffected.

As for the emotional mechanism, individuals who act due to intrinsic motivation are more likely to be emotionally invested in tasks and their outcomes, and individuals who act due to external coercion tend to be less intrinsically motivated. Thus, voters in compulsory systems are less likely to experience the psycho-emotional benefits and discomforts of victory and loss. Here there’s a sports analogy. For a passionate football fan, the outcome of a match can dramatically impact their emotional state and thus their evaluation of the match or even football as a sport. Now consider someone with no interest in football who is coerced by friends to watch a game. For this person, the outcome of the contest will have little or no effect on their emotional state or evaluation of the match or sport.

I test my theoretical predictions in two separate studies. In Study 1, I leverage a natural experiment whereby individuals are quasi-randomly assigned to compulsory voting based on their birthdate. In Argentina, turnout is mandatory by law, but only for those aged 18-69; for individuals aged 16-17 and 70 and up, participation is elective. Leveraging the cutoff ages, I show that the association between voting for the winner in the 2019 presidential election and SWD is weaker for those who were just old or young enough to be required to vote (though the effect is not statistically significant in the analysis of those around the age 70 cutoff).

By design, Study 1 focuses on people in particular age bands in one country. It is thus important to show that the effects generalize. To do so, in Study 2 I supplement the causal evidence from Argentina with a cross-national analysis of post-election survey data from the Comparative Study of Electoral Systems. Results show that the findings of Study 1 extend broadly. The association between voting for winners and SWD is weaker in countries where nonvoters are subject to penalties. Further, this is most evident among individuals who are less likely to have turned out volitionally.

My findings update the understanding of the consequences of compulsory voting beyond turnout, while also calling into question the universality of the strong association between voting for winners and satisfaction that has been detected by dozens of studies. The weakness of the “winner’s boost” under mandatory voting has an optimistic interpretation: losers are less dissatisfied relative to winners. As Anderson et al. note, accepting loss is part of the democratic bargain, and losers’ consent is probably more likely when there is a smaller gulf in SWD between the victorious and defeated.