March 8, 2006
Will Republican Retirements Swing the House?

By Jay Cost

With the announcement of Bill Thomas’s retirement, many have begun to take closer notice of the number of open seats in 2006 – and how they seem to favor the Democrats. Most political observers have recognized that the number of open seats is a critical factor in the partisan composition of the House, and that this year the Republicans must defend more than the Democrats. However, few have correctly surmised that the Democrats enjoy little-to-no real advantage because of open seats.

All political scientists agree that open seats are a key method of party changes in the House of Representatives. Only a half dozen (or fewer) incumbents tend to lose every year, so an open seat is the only way for a challenger to get around what is known as the “incumbency advantage”. This is the massive advantage that incumbent politicians enjoy in terms of money, approval rating and name recognition; because of this advantage, contests where the incumbent is running tend to be referenda on the incumbent. It is in open seat races where national trends can influence the election and where partisanship becomes more important. Thus, it is in open seat races – and open seat races alone – that the Republicans could seriously suffer from today’s anti-Republican mood. A handful of Republican incumbents will bite the bullet because of today’s political climate, but if the Democrats hope to retake the House, they will need to net a dozen or so open seats.

Open seats are also an indication of how the parties think they are doing. The fact that one party has more open seats than the other is often a sign that politicians sense a negative political climate. Politicians are strategic, and will flee when a loss seems imminent. So, open seat announcements are noteworthy; they give us a sense of how each party surmises its position with the electorate.

As it stands, the Republicans have 17 open seats to defend. The Democrats have 9. Historically speaking, with an economy as strong as today’s, and the party of the President having to defend 8 more open seats than the opposition, we can expect the Republicans to lose about 9 net seats. My sense is that the final figure will be slightly lower than that. The reason for this is that this estimate takes into account the quantity, but not the quality, of open seats in play. In 2006, the quality of Republican open seats is very poor from the Democrats’ perspective. Consider the following:

· Of the 17 open Republican seat districts, Bush won 15 in 2000 and 2004.

· Bush’s median percentage of the vote in 2000 in all 17 was 55%. In 2004 it was 57%

· Of the 17 open Republican seats, Bush increased his percentage of the total vote by an average of 3% between 2000 and 2004.

· The median Cook Partisan Voting Index of these 17 districts is Republican +5. In other words, the median district of these 17 tends to vote Republican 5% more than the nation.

Simply stated, Democrats are not salivating at this set of 17. These are not the sorts of districts that have shifted against Bush and the GOP enough to give the Democrats a real shot. Sure, there are a few among these – CO 07, IA 01 – that will probably switch to the Democrats because the seats are open. And IL 06 might be a nail-biter for the GOP. Nevertheless, these types of open seats are very dissimilar to the type of open seats the Democrats had to defend in 1994. The Democrats had many more open seats that year. And, furthermore, the quality of those open seats was much more amenable to GOP gains. In 1994, Republicans won Democratic open seats in conservative districts. It is unlikely that the Democrats will win Republican open seats in conservative districts.

Further, as mentioned above, open seats are frequently a sign of how politicians view their chances. Thus, we can get a sense of how the parties will do in November based upon seat vacancies today. If, for instance, lots of Republicans are retiring, we could infer that the Republicans sense an anti-Republican climate and have decided to avoid losing. Open seats in 1994 were a sign of the anti-incumbent climate of that year, for instance. What can we infer about the GOP based upon these open seats? Consider the following:

· Of the 17 open Republican seats, 9 of them are open because the incumbents are pursing higher electoral office.

· The median share of the vote that incumbents in these seats took in 2004 was 60%.

· Of the 7 Republican seats that are open because members have decided to retire, 2 of those members faced no opposition in 2004. The remaining won their 2004 election by a median value of 60%.

· Of the same 7 seats, the median length of time each current member has served is 24 years. All of them have served through anti-Republican, anti-incumbent elections.

· 7.3% of the Republican delegation will be open in November. 3.9% of the Democratic delegation will be open.

Clearly, these vacancies do not exist because these Republicans fear losing in November. All of these incumbents are quite skilled at retaining their seats, many of them are actually seeking higher offices, and as a whole the Republicans are not defending that many more open seats than the Democrats.

It is likely that we will see a few more Republican vacancies. But we should also expect a few extra Democratic vacancies as well. Overall, there will not be many more – and we can expect that the Republicans will have to defend about 8 to 10 net open seats. A large number of additional Republican retirees seems extremely unlikely. Retirees would need to announce early enough to give their replacements time to file their candidacies. Filing deadlines in many states have now passed; they are quickly approaching everywhere else.

All in all, the following seems clear to me. While the larger number of Republican open seats seems, at first glance, to favor the Democrats, a closer look at them indicates it does not. The Democrats should expect to net about 2 seats thanks to Republican vacancies. Thus, the Democrats, to take the House, will have to do something that is now almost impossible: defeat a net 13 Republican incumbents.

Jay Cost, creator of the Horse Race Blog, is a doctoral candidate of political science at the University of Chicago. He can be reached at

 © 2000-2006   All Rights Reserved

Jay Cost

Author Archive
Email Author
Print This Article
Send Article To a Friend

More Commentary
Media Silent on Radical Islamic Events - T. Blankley
Build a Fence -- And Amnesty - Robert Samuelson

The Inescapable Facts on Education - John Stossel

More From Jay Cost
In Defense of Pork
No More '08 Polls!
A Prediction for the 2006 House Contest - Part I | II