« The Mother of Presidents? | Main | The Iraqi Election: Democratic Voting or Fraud? »

18 October 2005

The 2006 midterms

Every day seems to bring a new lowest low to President Bush’s approval ratings, and at the same time large majorities of the public think that the country is heading in the wrong direction and won’t get any better in the near future. Common sense would seem to suggest that the Democrats might be looking forward to some success in the 2006 midterms.

Political scientists might also have cause to agree. There’s an extensive literature on partisan “surge and decline� in Congressional elections. When a President is on the ballot, he generates a "coattails effect" that helps members of his party win House seats (the surge) - but this effect vanishes at the next midterm when he’s not on the ballot (the decline). But the 2002 midterms didn't exactly fit this pattern - Bush seemed to engineer a new coattails effect by campaigning heavily for several Republican candidates who went on to win close contests.

National disapproval of Bush and the GOP agenda might not translate into higher approval for Democrats anyway. While the public generally dislikes Congress as an institution, they tend to approve of their individual Representative at higher rates: voters claim to choose “the candidate, not the party�. So as political scientists who claim to have assembled an array of generalizations about political behavior, do we have any way of gauging what’s likely to happen next fall?

Gary Jacobson’s 1989 APSR article argues that “strategic politicians� selectively enter races when national conditions favor their party over the governing party. These quality challengers (measured by whether they have held elective public office before) have more of the political skills necessary to win elections - so even when voters cast their ballot on the basis of local candidates, aggregate election results reflect national conditions.

So how are the Democrats doing in terms of their candidate recruitment for 2006? Larry Sabato’s page of the thirty House races that are likely to be close in 2006 reckons that nine of those seats have Democratic incumbents, fifteen have Republican incumbents and eight are open seats. Looking at races with incumbents, four of the Democratic incumbents are already facing opposition from a Republican candidate who has held elective office before, while nine of the Republican incumbents face similar opponents (several races have no declared candidates, and who the eventual nominee is obviously depends on next year’s primary races).

This would suggest that we’re not seeing a surge in the number of quality Democratic challengers that national conditions might lead us to expect. are not necessarily recruiting enough quality candidates to take back the House. But two questions remain. First, Jacobson’s original article noted that

…the crucial period is not when individuals decide to run for Congress but when they can no longer gracefully change their minds and decide not to run for Congress. Initial decisions to test the political waters may indeed be made right after the last election – if not earlier; final decisions to run or withdraw can be taken right up to the primary.

The numbers above represent declared candidacies, not potential interest in a race. Perhaps high-quality Democratic potential challengers are merely waiting to see whether the GOP’s string of bad news continues before jumping into campaign season?

Second, are we measuring the quality of candidates correctly? The justification for using previous elective office is that these candidates likely have the political skills and ability to campaign effectively, as well as a base and connections that allow them to raise money and start the election with some name recognition. But suppose that the “national issues� affecting the midterms are not economic (as Jacobson assumes they will be) or competency issues like the federal response to Hurricane Katrina. What if the national conditions instead become political reform? Should we then continue to measure quality in terms of previously holding elective office? Or do we want to start understanding quality in the traits of those who come from outside the realm of “politics as usual�?

Posted by Phil Jones at October 18, 2005 10:54 AM