Disentangling the Turnout Effect: The Case of Taiwan's 2012 Concurrent Elections

Abstract

It is generally agreed that when a relatively low-profile election is held simultaneously with a high-profile election, voter turnout rate of the former tends to be higher than it would be if it were held separately. The central concern of this study is twofold. The first one is to estimate the size of this effect, and the second one is to explore whether partisan effects of voter turnout exist. The case of Taiwan’s 2012 concurrent elections is employed in our study. For the first question, we estimate that the overall effect of concurrent presidential election on legislative election turnout rate is roughly 16.7%. As to the second research question, we find that partisan effects differ depends on voter types. For those who usually turnout to vote in high-profile elections, the pull effect of concurrent election is nearly universal regardless of demographical characteristics and partisanship. For those who rarely turnout to vote in stand-alone elections but are mobilized in a highly competitive concurrent election, the pull effect does show partisan difference. Surprisingly, such effect is somewhat beneficial to the pan-Green camp in Taiwan’s 2012 legislative election. Despite the exploratory nature of this study, our findings have important theoretical implications. Future studies of partisan-turnout effects should specify the mechanism behind higher (or lower) turnout before inferring its effects. Furthermore, there may be different “layers” of voters drawn to the voting booths in a high turnout election and their voter compositions, including partisanship, may be also quite different.