How will the 2010 midterm election affect the ideology of the House Democrats? Without speculating or considering how it might have changed the ideologies of individual returning members, it probably is safe to say it will have a moderately large aggregate effect simply via replacement. As many commentators have pointed out  (for example here and here), the Democratic losses in the election were skewed toward the more conservative wing of the party’s House membership. After the jump, I use some rough-cut empirical data to shed further light on this point.

Below are D-Nominate scores for each Democratic member of the 111th Congress, ranked from most liberal to most conservative. In this ranking, the median Democrats in the 111th Congress were Rep. Jim Langavin  and Rep. Mike Doyle. Members colored in green and red represent seats won by Republicans for the 112th Congress; green indicates a seat that was held in the 111th by a Democrat with a D-Nominate score more liberal than the median Democrat; red indicates a seat held by a Democrat with a D-Nominate score more conservative than the median Democrat.

As can be seen, 59 of the lost (or currently losing) Democratic seats were from the conservative half of the caucus. Only 5 were from the liberal half of the caucus. Two obvious implications of this are: (1) the caucus as a whole is likely to be more liberal in the 112th Congress; and (2) the median member of the caucus in the 112th is likely to be more liberal than the median member of the caucus in the 111th.

Four caveats accompany this analysis: Seats being vacated by Democrats that were won for the 112th by Democrats are assumed to have unchanged ideologies. Second, the three Republican seats won by Democrats for the 112th are not included. Third, some of the elections from 2010 have not been decided. In these cases, I went with the current vote leader. Finally, I did not clean the data for deaths or party changes or resignations that occurred during the 111th. Consequently, as noted above, this is a rough cut at the data and the conclusions are generalizations rather than specific numbers.

One quick application to this entails leadership selection. Since the Democrats will be losing one leadership slot (Speaker), some member of the current leadership will need to step out of the leadership. If it is the case that nothing can be arranged informally, the party elections are likely to reflect, in part, the new, more liberal ideology of the caucus.