Objectives. In this article we explore how the geographic location of a proposed public good on the ballot in a local referendum influences voting turnout. We argue that voters who live farther away from the good, and are thus likely to bear the cost of the good but have no access to it, would be more motivated to turn out in the election. Drawing on the cost-orientation hypothesis, or negativity effect, ''that people are more strongly motivated to avoid losses than to approach gains,'' we expect these voters to derive higher expressive benefits from the act of voting relative to those of voters located closer to the good. Methods. We examine voting turnout in the 2002 referendum in the City of Seattle on the proposed construction of a monorail. We conduct our study at the precinct level using spatial tools of analysis. We evaluate the effect of accessibility on turnout by means of a curvilinear model that incorporates demographic and political variables. Results. We find that voting turnout is determined partly by accessibility. Turnout is higher in precincts located farther away from the monorail line than in precincts located relatively closer to the line. Partisanship conditions this effect. Conclusions. This study provides tentative support for linking voter turnout to the negativity effect via expressive benefits. Voters' location in relation to a public good may affect directly their political behavior by means of their perceived net gains or losses from the good. . The corresponding author will share all data and coding information with those wishing to replicate the study. The second and third authors are listed alphabetically. The authors are grateful to Cristina Bradatan, Scott McClurg, and the SSQ editor and anonymous reviewers for helpful comments and constructive feedback.

To explore the mechanism underlying the formation and persistence of cleavage structures the author applies a contextual approach in the case of Norway where regional variation in political cleavages persists over time. This study focuses on the time period between 1890 and 1930, the formative years of the Norwegian party system. The premise of the argument is that a factor contributing to persisting patterns of electoral mobilization is the location of new voters in patterns of social interaction deriving from the class composition in different regions. Initial electoral mobilizations coupled with enduring social structures can be carried on for a long time. These social conditions may pose high costs of mobilization to new parties in a political system. The article shows the persistence of a distinct class composition across regions in Norway and demonstrates the contextual effect by examining the behaviour of new voters in the 1900 election. It evaluates the contextual argument against other party incentives for mobilization by means of a statistical model that incorporates contextual incentives in addition to incentives deriving from electoral rules and political competition.

Contests over public goods remain at the forefront of urban political battles in nearly every major city in the United States. The spatial location of the good can play an instrumental role in understanding the contours and outcomes of such conflicts. The authors explore a particular case—voting for a growth-related development project, the monorail, by referendum in the city of Seattle—and examine how a grassroots campaign successfully mobilized voters by targeting both their particularistic and collective interests. The authors conduct analysis at the precinct level and use spatial tools of analysis and ecological inference, finding that voter support for the monorail stemmed from the location of the proposed route and the campaign's progressive appeals to environmental, social justice, and high tech concerns. Although cost overruns ultimately derailed construction of the monorail in 2005, when passed in 2002, the monorail was the most expensive infrastructure project in Seattle's history.

No abstract

In this article we inquire into the strategic intent behind the design of election laws. Presuming that institutional designers are strategic and rational, we identify the extent of information incompleteness as determining their objectives for rule choice. For different levels of information incompleteness, we assess empirically the validity of explaining actual choices with designers' electoral goals. Using parameter-specific predictions for players' institutional preferences (obtained from a game-theoretic model which assumes electoral success as the design goal) we find that design does not always meet the criterion of elections-aimed rationality (electoral rationalizability). We identify a structural difference between levels of electoral rationalizability in two sets of European electoral reforms, those that accompanied the late 19th-early 20th-century franchise expansion, and those during the post-communist franchise creation in East-Central Europe. For the later sub-sample we rely on a qualitative assessment of electoral rationalizability since extremely high levels of information incompleteness in these cases do not allow us to apply the model reliably. We provide evidence linking the structural difference in electoral rationalizability between the two sub-samples and the sub-sample variation in electoral rationalizability in the late 19th-early 20th century to the quality of information available to the designers.KEY WORDS . endogenous institutions . election laws . Europe . franchise expansion . incomplete information . institutional design