About half or more of the eligible voters typically participate in national elections in established democracies. Millions of citizens vote, even though the odds their ballot will decide the outcome is vanishingly small. What sustains voter participation is, upon reflection, not obvious.
During a career of pioneering work, André Blais has studied the decision to vote from a variety of angles. Among his many contributions, Blais has shown in previous work that variation in beliefs about the odds of casting a decisive ballot—the main focus of the classical version of the rational theory of voting—has only a modest influence on participation (Blais, Reference Blais2000). In his illuminating and well-crafted new book The Motivation to Vote, Blais and co-author Jean-François Daoust demonstrate the importance of four individual-level factors on electoral participation: a voter's level of political interest, civic duty, the perceived ease of voting and how much the voter cares about a particular election. The first two factors are enduring individual predispositions that are shaped early in life, while the latter two are election-specific considerations. The book draws on the highly ambitious Making Electoral Democracy Work (MEDW) survey project and reports the results from surveys conducted in five countries over 24 elections between 2011 and 2015. The main analysis is based on the 26,000 interviews conducted prior to these elections.
Although all four of the factors studied by Blais and Daoust predict the decision to vote, Blais and Daoust report that the individual's level of political interest and civic duty are much more important than election-specific factors. As they write, “Our claim is that these two basic predispositions, one's level of interest in politics and one's feeling that voting is or is not a moral obligation, are the two most powerful individual-level determinants of the decision to vote or abstain. Because these predispositions are stable, there is strong stability in the propensity to vote” (103). This important finding suggests that scholars should devote greater attention to investigating how the enduring psychological commitment to voting develops and what the root causes are of political interest and a sense of civic duty.
Even the best datasets have limitations, and the excellent MEDW surveys are no exception. One concern is how the key variables are operationalized and measured. The survey item used to measure political interest, which is intended to capture an enduring individual predisposition to like politics, is the response to a single question about interest in “the current federal election.” The variable measuring how much the individual cares about the current election, which is used to distinguish the predisposition of political interest from how much the individual cares about the current election, asks how much the individual cares about “which party will form the government after the election.” The relatively large effect of “political interest” on turnout reported in the analysis is the turnout boost measured in models without the “care” variable, while the impact of “care” is measured as the incremental effect once the variable “interest” is included in the model. This approach to distinguishing the effect of long-standing “interest” from that of caring about the particular election is well justified under a theoretical model in which the individual's political interest is fixed prior to the current election and if the relevant constructs are well measured.
More generally, the observed stability of an aggregate outcome variable like turnout or vote share may be due to either the insensitivity of individual behaviour to the current political context or due to the relative stability of contextual influences acting on the individual. The latter would suggest that factors proximate to the election can generate a substantial change in outcomes if the context shifts in significant ways. The recent US political experience, in which the 2018 midterm election turnout climbed to 50 per cent (up from 40 per cent in recent prior midterms), while the 2020 general election turnout increased to 67 per cent, a level not seen in over a century, provides suggestive evidence that contextual factors can be quite important. Although contextual effects appear relatively modest in the sample of elections studied by Blais and Daoust, such effects may be large and politically important at other times.
Blais and Daoust's work suggests the importance of gaining a better understanding of how the incentives of political elites and the structure of political competition shape the context in which citizens develop and maintain an enduring individual-level attachment to voting. Many once-popular activities that seemed rock solid, including church attendance and watching the World Series, have shown serious participation declines in recent years, but voting remains strong. This could be the good fortune of democracy. Alternatively, the fact that voting has not gone out of fashion may reflect the strategic interests of politicians, parties and groups in increasing voter engagement through emotionally engaging issues or enhanced mobilization. If interest in voting fell, then elites would find that they would need to mobilize fewer voters to gain power. This would provide a strong incentive to search for ways to motivate supporters to vote, and we might see individual-level attitudes about voting change in response to these mobilization efforts.
Overall, Blais and Daoust have produced a major contribution to our understanding of voter turnout. Their book suggests many avenues for future research and should be read by all scholars of voting and political behaviour.
