Party registration choices as a function of the geographic distribution ofpartisanship: a model of 'hidden partisanship' and an illustrative test

The Public Choice literature has identified conditions in which voters in m ulti-candidate contests would have an incentive to vote strategically rathe r than vote for the most preferred candidate or candidates. In the US, wher e party registration and party primaries play a critical role in the electo ral process-especially in states with closed primaries-the existence of mul tiple layers of elections across constituencies can induce strategic falsif ication of party registration that is tied to the geographic distribution o f electoral strength. Following V. O. Key, we should expect that a long his tory of one party dominance in local elections should encourage voters to r egister in the party whose elections are most determinative of electoral ch oices, even if that is not the party with which they most identify. However , in many states, while politics may be dominated by one pasty locally, the re may be real two-party competition for at least some offices at the state level and for the presidency. We use a 'natural experiment' to view the link between party registration a nd voting for president and obscure judicial offices in order to test the h ypothesis that, for whichever party is the minority party in the local unit , party registration will understate the voting support in presidential or other statewide elections, where that party's candidates have a realistic c hance to win. In the modern South this hypothesis can be shown to imply tha t the relationship between Republican party registration and vote shares fo r Republican candidates for president or statewide office ought to be curvi linear. To test this and other related hypotheses, we examine data on polit ical units (e.g. counties) with considerable variation in party registratio n and concomitant variation in the extent of one-party dominance of local p olitics by looking at county level data from North Carolina for the preside ntial elections and obscure judicial elections in 1984 and 1996. As hypothe sized, for the North Carolina data the relationship between party registrat ion and voting can best be fit by a quadratic function, but the strength of the quadratic term is much less for the 1996 data, reflecting the increase in Republican registration and the success of local GOP candidates in the 1990s. (C) 1999 Elsevier Science Ltd. All rights reserved.