07.11.2019 | Susana Aguilar and Andrés Santana
Most models of voting incorporate the costs of voting, which may include both the informational costs necessary to decide whom to vote and the actual costs of going to the voting booth on election day, as one of the key factors to explain why people cast a ballot or not. The truth is that we do not know yet why Downs was wrong and why most people bother to vote. Why are some people especially inclined to feel that voting is a duty? Why are some citizens particularly prone to overestimate their influence as voters? Little is known as well about why some people perceive higher or lower costs of voting than others. To cut a long story short, voting is a rather puzzling activity and there is a gap in the literature as to what are the determinants of the costs of voting at the individual level.
We have addressed this gap with the help of the “Making Electoral Democracy Work” database, which contains survey information on national elections in France, Germany, Spain, Switzerland, and Canada. By connecting the rational choice model of voting with the sociological and psycho-sociological models, we have been able to test whether C is affected by socio-demographic and/or attitudinal factors.
Figure 1 shows the results. It has not come as a surprise to find that C is lower for people who identify with a party, have a higher education, are unionised, have lived longer in the region where the election takes place, and are middle-aged, since all these factors cut down the informational costs of voting. The positive sign of the age squared coefficient (which is statistically significant, although in the graph it is hard to notice) is probably due to the increase in the physical effort that is required to go the voting booths beyond a certain age threshold, and means that the costs of voting display an overall U-shaped relationship with age. C is also lower for urban dwellers (possibly, because the transportation costs of going to the polls are smaller in cities than in rural areas) and men but, unexpectedly, it has not been found to be affected by the presence of kids at home. Finally, it falls with political interest and the importance attached to elections.
We have also detected that all the factors that increase C decrease the propensity to go to the polls. There is however one interesting exception: women have higher C than men and yet vote as much as them. Alongside sex, those in certain age groups (the youngest and the oldest), together with the less educated, the rural dwellers, and the newcomers in the region have higher C and therefore suffer from a disadvantageous position when it comes to exercising the main right in democracy: the right to vote. Take the case of age and education: whereas the probability of experiencing high costs is only about 0.14 for higher educated 58-year-olds, it is almost 0.35 for lower educated first-time voters (see Figure 2).
This situation might constitute a serious breach of one of the main tenets of our political regimes, that is political equality. It also places five consolidated democratic countries on a somehow similar footing to the United States, a country in which abstention has been long ascribed to ethnic minorities and worse-off groups. Our results bear therefore important political and normative implications: if we want to increase the electoral turnout and reduce the comparatively high costs of voting that the women and the youth (and the other mentioned groups) perceive, policy measures aimed at easing the electoral administrative procedures as well as “luring” these groups into going to the polls, should be promptly adopted. World-wide different campaigns to augment the electoral turnout have been implemented in countries as diverse as New Zealand, Peru, Sweden and Ukraine. Only a few have however addressed the youth (Canada and Brazil) and none, to our knowledge, has been tailored to women.