Bush Dogs innoculation votes fail

There are 39 Bush Dogs --- Democrats in Congress who, on big, partisan issues, vote to trust George W. Bush and his judgment against the rest of the Democratic caucus and 80%+ of the Democratic electorate. The Bush Dogs make up 16.8% of the Democratic House caucus. On national security issues, this group allows for conservative control of the process as Speaker Pelosi is operating under a majority of the whole instead of a majority of the majority rule set.

Most Bush Dogs are either Blue Dogs or New Dems, and they have formal internal party caucuses. The first purpose of a caucus is to have members help each other out. This often means electoral assistance with the goal of winning re-election. I have no problem with this, as the CBC, the Hispanic Caucus and the Progressive Caucus attempt to do the same thing with varying levels of effectiveness. People join caucuses on the basis of common identity and goals. So these two groups that vote against Dems believe that this is useful behavior to their own goals.

Why do they vote this way? It can be either that they personally believe in George W. Bush's judgment and vision for a cowering America, which is a harsh indictment of their own judgment. Or they believe that they are in districts which demand these reactionary votes, either from the point of view of voters, or more likely from their probable donor base. In this projected political calculation, a Bush Dog calculates that voting with the rest of the Democratic Party is very dangerous to their future political careers, so voting for Bush is a political innoculation to prevent a strong reactionary Republican challenge.

Yet this strategy is not working. One would expect that if Bush Dogs are voting their districts in voting for George W. Bush's policies and contortions of the Constitution, they would be on average, no more vulnerable to a challenge than most non-packed and stacked district Democrats. The National Republican Congressional Committee has released its initial target list of twenty four Democratically held seats it wants to field first string challengers against.

If the innoculation strategy works, one would expect roughly four Bush Dogs to be challenged, as that is the proportion of the Democratic Caucus that they compose. Instead a third of all of these 'first tier' challengers will be against Bush Dogs, for a relative risk of 2.0 for a Bush Dog compared to a generic House Democrat. That is a signifcant increase in risk if we assume that first tier challengers have a substantially higher probability of knocking off an incumbent than non-first tier challengers.

For the innoculation strategy to still be a viable strategy, one must assume that there are more Bush Dogs who the NRCC looked at and said 'Hmm, he is an entrenched incumbent with great constituent support, amazing fundraising and if only he had voted against telecom immunity we could beat him this year instead of the past eight attempts in a very favorable district....'

I don't think there are that many marginal Bush Dogs where one or two votes matter. So either the political calculation for innoculation is wrong, or their judgement is wrong as exhibited by supporting George W. Bush on a couple of crucial partisan matters.