Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: The thesis positing a connection between party competition and electoral turnout has reached the status of conventional wisdom in American politics. While we do not argue with the theory itself, we suggest here that its empirical validation is based on a poor connection between measurement and theory. Specifically, previous studies examine the competition-turnout linkage by using state legislative competition measures to explain turnout in presidential and other national elections. We find a theory of state-level party competition driving turnout in national elections to be implausible, and see little reason to believe that competition in low visibility and low salience races would be the mechanism motivating citizens to vote in high salience and high visibility presidential elections. What is lacking for an appropriate examination of this relationship is a suitable measure of national party competition in the states. In previous work (Brown and Bruce, forthcoming) we developed such a measure, and use it here to examine variations in state turnout rates in presidential, senatorial, and U.S. House elections. Our analyses support the conventional wisdom, finding a generally robust relationship between national electoral party competition and national turnout rates across the states. Yet even in supporting the conventional wisdom, our analyses are instructive, as they represent a significant improvement in the link between measurement and theory in the competition-turnout literature.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper tests David Laitin's competitive assimilation hypothesis with the aid of Latvian school attendance records. Aggregate data appear to confirm a moderate rise in the number of Russian-speakers sending their children to Latvian schools. Disaggregate data, however, paint a more complicated picture, one which questions a main assumption of the Laitin model. Minority parents do not send their children to Latvian schools in Latvia's economically vibrant cities and continue to opt for education in Russian, despite the apparent economic rewards for knowing Latvian in these cities. I explain this paradoxical outcome by operationalizing and measuring the three variables that Laitin cites as affecting rates of assimilation: economic rewards, in-group status, and out-group acceptance. I then propose an alternative game theoretic framework for analyzing school choice.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper represents an initial effort to model the evolution of party systems, to account for stability and change in party systems actors and policy space. It employs both a simple simulation and a game theoretic heuristic to generate a series of hypotheses about party systems stability. The paper focuses on the distribution of voter preferences, the level of strategic voting, and the degree of party competition as important determinants of party systems change in established parliamentary systems, while controlling for variation in electoral institutions. Evidence from a longitudinal case study on Canada and a time series, cross-national (TSCS) data set support some of the hypotheses derived from the game theoretic heuristic. Rules for a more complicated simulation are presented, which, when programmed, will help evaluate the validity of the hypotheses in more complex party systems.

Publication Type: Proceeding
Abstract: For vote-maximizing candidates in two-party contests, basic Downsian theory argues for candidate convergence away from their own party’s supporters and toward the views of the median voter in the district, and it also leads us to expect these centripetal pressures to be strongest when elections are expected to be close. Yet, the available evidence from the U.S. Congress – both the work we report here using DW-NOMINATE scores over the 1952-2000 period and that of many other scholars using closely related methodologies – challenges these predictions. Drawing on and extending recent work of Adams, Merrill, and Grofman (2005), we propose a neo-Downsian model to explain non-convergence re-sults. Our model allows us to understand the complex interactions of political competi-tion, partisan loyalties, and incentives for voter turnout that can lead vote-maximizing candidates, even candidates in close elections, to seek to mobilize their voter base in terms of turnout by adopting positions attractive to the party faithful, rather than courting the median voter by moving towards the center. Furthermore, in our model, the winners from the party whose supporters are more ideologically concentrated (in the U.S., the Re-publicans) can be expected to display even less relationship between their ideological lo-cations and election competitiveness in their districts than is true for the representatives from the party whose voters are more ideologically dispersed.

Publication Type: Proceeding
Abstract: This paper argues that recent representation research demonstrating that legislators appeal to subconstituencies rather than the district as a whole offers some leverage in helping to reconcile these counter-intuitive and apparently conflicting findings of past research on competition and representation. To date, all studies of the effect of competition on responsiveness rest on the assumption that politicians appeal to the average voter in their districts. However, scholars seldom find that this occurs.