An enduring puzzle in comparative politics is why voters in some democracies continuously support dominant parties in elections, and whether their support is based on policy congruence or non-policy factors like valence. Smith and his team consider the preeminent case of a dominant party—Japan’s Liberal Democratic Party (LDP)—and investigate whether voters’ support for its policies can explain its recent landslide election victories. They first introduce a new measurement strategy to infer individuals’ utility for parties’ policy platforms from conjoint experiments. Unlike most other uses of conjoint designs, their approach quantifies individual preferences for entire platforms rather than the average effect of any one component. Using this measure, they then show that many voters supported the LDP in the 2017 and 2021 elections despite preferring the opposition’s policies. To understand what accounts for this disconnect, Smith and his team experimentally manipulate party label and decompose its effect, revealing that trust is an important non-policy factor motivating LDP voters. Together, their findings support the argument that the LDP’s recent dominance can be attributed to its valence advantage over the opposition rather than voters’ support for its policies.

This event is part of APARC's Contemporary Asia Seminar Series.

Daniel M. Smith is Associate Professor of Political Science at the University of Pennsylvania. His research interests cover a range of topics in comparative politics and Japanese politics, with a core focus on elections and democratic representation. He is the author of Dynasties and Democracy: The Inherited Incumbency Advantage in Japan (Stanford University Press, 2018), and articles appearing in the American Political Science Review, American Journal of Political Science, The Journal of Politics, and Comparative Political Studies, among other journals and edited volumes. He also co-organizes the Japanese Politics Online Seminar Series (JPOSS), and co-edits the Japan Decides election series.