We build a model of voluntary and costly expressive voting, where the relative weight of ideology and valence
issues over voting costs determines how people vote and if they actually turn out to vote. In line with the conventional
rational calculus approach, the model predicts that the cost of voting depresses voter turnout. Against the conventional
wisdom, though, high voting cost/low turnout elections tend to have a larger share of voters for whom the common value
signal on candidates’ valence matches their private value views, thus raising the chances that high valence candidates are elected.