Working Papers

Theories of democratic representation assume that politicians’ constituency-building behavior is shaped by electoral incentives, but empirical evidence on this relationship is limited. We use detailed data from Japan covering five decades of itemized campaign expenditures, municipality-level vote shares, and digitized records of parliamentary meetings, to measure politicians’ constituency-building behaviors. We estimate how changes in vote-seeking incentives affect campaign spending and legislative behavior with a difference-in-differences design exploiting within-individual variation stemming from court-induced seat reapportionments (which alter vote thresholds, and hence the size of the constituency needed, to win a seat). The results show that politicians respond to changes in incentives through the activities they directly control: when a district loses a seat, candidates spend more in campaigns and broaden their geographical mobilization efforts. In contrast, we find no evidence that legislators respond through participation or policy specialization in parliamentary activities, which are more constrained by parties and legislative institutions.

Abstract: Low voter information can disadvantage female candidates, especially in commonly used preferential voting systems where intraparty choice reduces the heuristic value of party labels. We argue that such disadvantages can be overcome by (1) providing voters with candidates’ biographical information, and (2) making preference voting compulsory. The former can disabuse voters of (negative) gender stereotypes, while the latter can counteract inequalities in the use of preference voting by men and women, and encourage voters to consider candidate information. We test our expectations with two survey experiments conducted during recent open-list proportional representation elections in Japan, where women’s representation has notoriously lagged. The results confirm that each experimental treatment (information and compulsory preference voting) increases voters’ support for women, both among the actual candidates running and in a conjoint experiment featuring hypothetical candidates. These findings underscore how simple rule changes might reduce barriers to women’s electoral success in preferential voting systems.

Abstract: The share of women in politics is higher, on average, under closed-list proportional representation (PR) electoral systems compared to majoritarian systems. Yet, even in PR systems, progress toward gender parity has been slow and uneven. We consider the role of a common institutional feature of party organization—seniority-based promotion—and argue that gender gaps in career progression can emerge either due to direct bias in the seniority system, or because majoritarian offices (such as local mayor and list leader) serve as important steppingstones that create bottlenecks in women’s career paths. Using more than a century of detailed candidate-level data from Norway, we find that advancement is generally gender-neutral across stages of a typical political career, but that gender gaps emerge at majoritarian bottlenecks. We also document how parties can employ workarounds to mitigate the adverse effects of these bottlenecks on women’s progression into higher offices.

Abstract: 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 preferences or non-policy factors like valence. We consider the preeminent case of Japan’s Liberal Democratic Party (LDP) and investigate whether voters’ support for its policies can explain its recent landslide electoral victories. We first introduce a new measurement strategy to infer individuals’ utility for parties’ policy platforms from conjoint experiments. Estimating a model of vote choice with this utility reveals that many voters support the LDP despite preferring the opposition’s policies. To understand what accounts for this disconnect, we then experimentally manipulate party label and decompose its effect, finding that trust is an important non-policy factor motivating LDP voters. Together, our findings support the argument that the LDP’s continued dominance is due to its valence advantage over the opposition rather than voters’ support for its policies.

Abstract: Studies of voting behavior in some settings may be hampered by poor data availability or unsuitably large units of aggregation for reported turnout. We propose and demon- strate a practical big-data solution to these kinds of challenges, using fine-grained cell-phone mobility data on millions of GPS locations for more than 300,000 eligible voters in Tokyo. Our approach uses the geolocations of polling stations, combined with GPS data points recorded on election day and a reference day, to measure patterns in individual-level (but anonymized) voting behavior. We first test the validity of the measure by comparing it to official aggregated data on turnout, and then illustrate its substantive utility with an application exploring the well-known relationship between turnout decisions and the cost of voting, proxied by the distance between a voter’s residence and the polling station. Finally, we discuss the potential limitations of the approach and provide step-by-step instructions for other researchers.