The past two election cycles have not been kind to the Republican Party. Since 2006, the GOP has lost 14 seats in the U.S. Senate and 54 seats in the House of Representatives. So naturally Republicans have been looking forward to the 2010 midterm elections. Not since 1994 have they gone into an election cycle with such high expectations. With Democrats defending a large number of seats in traditionally Republican territory, unemployment hovering just below 10 percent, and a Democratic president with mediocre approval ratings, Republicans have high hopes of picking up a large number of House and Senate seats and possibly taking control of one or both chambers.
It is almost certain that Republicans will make substantial gains in the midterm election. But there is growing concern among some GOP leaders and strategists that the party may fall short of its goals because of a growing internal schism between the party establishment and conservative grassroots activists. In a number of races, insurgent candidates running under the anti-government Tea Party banner have challenged candidates supported by national and state Republican organizations. Several of these upstart candidates, like Rand Paul in Kentucky and Sharron Angle in Nevada, have already secured nominations by defeating candidates backed by the GOP establishment. Even where Tea Party candidates have lost, they have sometimes forced mainstream Republicans like Mark Kirk in Illinois and Carly Fiorina in California to veer to the right to shore up their support among conservative GOP primary voters.
According to conventional political-science wisdom, candidates who stray too far from the center of the ideological spectrum run the risk of hurting their chances in the general election by alienating moderate swing voters. But lately that assumption has been challenged by conservative strategists who argue that the best way for the GOP to regain voters' trust is to go back to the party's conservative roots by standing up squarely for smaller government, less regulation of business, and traditional "family values." What Ed Kilgore of The Democratic Strategist recently referred to as the "move right and win" strategy appears to be very popular among conservative Republican activists. But evidence suggests that a party doesn't always improve its electoral prospects by moving away from the center.
One way of addressing this question is to look at the relationship between the ideologies of congressional incumbents and their electoral performance. The advantage of focusing on incumbents is that their voting records can be used to gauge their overall liberalism or conservatism. For example, in the current Senate, based on a widely used scale called DW-NOMINATE, Russ Feingold, a Democrat from Wisconsin, and Bernie Sanders, an independent from Vermont, have the most liberal voting records, while Republicans Jim DeMint of South Carolina and Tom Coburn of Oklahoma have the most conservative.
In order to evaluate the impact of ideology on electoral performance, I conducted a statistical analysis of all contested Senate races involving incumbents between 1992 and 2008. I controlled for other factors that influence election results, such as the strength of the parties in the incumbent's state, the strength of the challenger, and the political climate at the time of the election.
The results showed that for Republican incumbents, conservatism had a significant negative influence on electoral support. A 10 percent increase in conservatism was associated with a decline of about 1 percentage point in the incumbent's vote. This may not sound like much, but a 10 percent decrease in conservatism might have saved seven of the 21 GOP incumbents who were defeated in these elections, including George Allen Jr. and Conrad Burns in 2006 and Norm Coleman and Ted Stevens in 2008.
What about the other party? Interestingly, for Democratic incumbents, liberalism did not have a significant impact on electoral performance. Only nine Democratic senators lost their seats between 1992 and 2008. Having a strongly liberal voting record neither helped nor hurt Democratic incumbents, which may reflect the fact that liberal Democrats generally don't emphasize ideological themes to the extent that conservative Republicans do.
It's not clear whether the ideologies of challengers and open-seat candidates have the same effects on electoral performance as the ideologies of incumbents. On that question, as we say in the social sciences, further research is needed. However, the evidence from two decades worth of Senate races involving Republican incumbents does raise serious doubts about the "move right and win" theory. This evidence suggests that while ideological moderates may have a hard time winning GOP primaries these days, they make stronger general-election candidates than hard-line conservatives.