Consistent evidence of variation in the participation gap between education or income groups in developing and developed democracies has begun to accrue. This points to varying disparities in participation between the haves and have nots that occasionally reach alarming levels, potentially triggering breakdowns in political representation. A few cross-sectional analyses identify institutional factors, such as voting complexity or state capacity, or economic ones, like income inequality, as driving the difference. Few explanations currently try to address why this participation gap varies over time. This is the question I take up here—I examine the extent to which the turnout gap has changed over time, and what the most robust explanations are for this temporal trend out of a set of factors grouped into 3 “families”. These sets of explanations refer to mechanisms that operate through voters’ (1) resources, (2) motivation to participate, or (3) likelihood of being mobilized at election time. Using an original pooled data set, with individual-level turnout data from 170–180 elections in 21 OECD countries, and party placement data from the MARPOR project, I show that the magnitude of the turnout gap between lower-SES and higher-SES citizens has increased over time, and that this trend has mainly been driven by the demobilization of lower-SES citizens. A two-stage analysis reveals that union density along with legislative fractionalization are the most consistent correlates of the participation gap between SES groups. The results are obtained from the most comprehensive evaluation to date of the way in which the SES-based turnout gap in advanced democracies has evolved over time. The findings contribute to our understanding of the long-term consequences for individual political behavior of institutional transformations in advanced democracies, and the democratic implications of these changes.