Party Registration and the Geography of Party Polarization

Abstract

Are voters in the United States sorting into geographic “enclaves” of determined commitment to one party or the other, or do voting and public opinion patterns hide a fundamentally ambivalent electorate? The most common geographic measure of partisanship—the presidential vote—cannot identify strength of commitment to the parties because voters have no viable alternative to the major-party candidates. We use party registration statistics instead, since voters can safely register with a third party or as an independent without giving up their right to choose a major-party candidate in the fall election. These numbers suggest that the strongest trend by far is not growing polarization but the large and pervasive increase in registered independents. Although the 2008 election was extraordinary in many ways, its impact on these basic trends is ambiguous. We offer some preliminary analysis that suggests this trend is ideologically meaningful and bears important implications for polarization, and then make a case for further research on the independent phenomenon.