The party in the White House usually struggles in midterms, but the Republicans could lose their majority without a huge swing
I don’t believe the Democrats will win back the House of Representatives in 2014. President Obama’s low approval rating, combined with the usual midterm loss and normal movement away (pdf) from the White House party on the national House ballot, should keep Republicans in control. Yet, there’s a difference between thinking whether the Democrats “will” win back the House or whether they “can” win it back.
Right now, the Democrats hold a lead of about 4-5pt per the HuffPollster and Real Clear Politics average. Many have concluded that this lead would not be enough to take back the House, if the election were held today. However, I believe that it quite likely would be enough.
How so? Let’s address a bunch of reasons people expect that a 4-5pt Democratic lead on the national House ballot would result in Republicans still holding the House – and then show why I think those could be wrong.
1. A uniform shift of 4-5pt on all House seats would still leave Republicans winning a majority of seats
North Carolina Republican Congressman Robert Pittenger was the “median” representative in 2012. Half the races were decided by more than his 6.1pt margin and half were decided by less. Given that Democrats won the national House vote by 1.4pt, a uniform swing across all districts would imply that Democrats would need to win the national House vote by 7.5pt to take back the House.
Count me as one of the people who does not believe in uniform swings. It’s not that the uniform swing is uninformative; it’s that it is very inexact. There are many factors that go into House races, including challenger quality, money spent, and whether or not the incumbent is running for re-election. Most of those are unknown at this point for key races.
You only need to look at the 2006 election to get an idea. Back in 2004, Republicans won the national House vote by 2.6pt. They won the median district by a little over 10pt. In other words, there was that same 7.5pt pro-Republican bias between the national House vote and the median district in 2004 as there was last year.
When we examine 2006, we see the bias simply didn’t hold. Democrats only won the national House vote by 8pt, which should have given them the thinnest of majorities per a uniform swing. Instead, they took 233, or 13 more, seats than a uniform swing implied.
The 218th seat won by the Democrats belonged to Leonard Boswell, who had actually taken the seat easily in 2004. He had health problems, which led to a closer than expected re-election campaign. Boswell, with a winning margin of 5.4pt, might have survived even if the national Democratic margin had been closer to 3pt.
My own math, taking into account redistricting in 2011, says a 3pt Democratic win in the national vote and a takeover of the House would not be nearly as likely as in 2006; but a 4 or 5pt victory would probably do the trick.
2. The experts say there are very few seats up for grabs
The indispensable Cook Political Report has only has 13 Democratic-held seats listed in the relatively competitive tossup or “lean” category. Of course, Democrats need to take 17 seats to win the House. The ratings reflect, among other things, a lack of strong challengers for the Democrats and lack of retirements by Republicans.
The thing is that expert ratings (like most polling) are not all that predictive a year out from an election. At this point in the 2006 cycle, there were 17 Republican seats in the lean or tossup categories (pdf). That’s well short of the 30 seats that Democrats would ultimately take from Republicans. At this point in the 2010 cycle, there were 28 Democratic seats in the lean or tossup category. Republicans, of course, went onto gain 63 seats in 2010.
It’s not until later in the cycle when individual seat rankings become quite useful. That’s when potential challengers and incumbents read the national environment and decide to run or not. Chances are that if the 4-5pt Democratic lead holds, the individual seat rankings will reflect that edge. For now, individual seat ratings probably aren’t all that helpful to understanding which way and how hard the wind is blowing.
3. The Abramowitz model says Democrats need something like a 13pt margin on the national House ballot
Alan Abramowitz’s national House ballot to seats model seems to have unusual sway among some. The model is elegant in the sense that it does a good job of trying to map the midterm penalty and how much exposure the majority party has, in a minimalistic fashion. The problem is that some don’t seem to quite understand how the model should work.
It’s not a straight national vote-to-seat equation. It’s built for early September of a midterm year. Abramowitz isn’t saying that a 13pt Democratic margin in the national House ballot on election day is what Democrats need to take over the House. What he is saying is that a 13pt lead in September is likely to shrink because of the natural movement away from the White House party on the national House ballot during the course of the election year.
Moreover, the model is inexact. It would be within the margin of error of the model for Democrats to take back the House with a 2pt September lead on the national House ballot. In 2010, the model forecasted a Republican gain of 45 seats per my calculation. That was 18 seats off the final Republican gain of 63 seats.
Abramowitz’s forecast is a good starting-point for understanding how uphill is the Democrats’ task in taking back the House, but it is far from perfect.
4. The final national House ballot surveys are biased against Republicans
Charlie Cook has a rule that you subtract 2pt from the Democratic margin on the final national House ballot to know how the national House vote is actually going to pan out. That may have worked over five years ago, though it doesn’t seem to work anymore. In 2008, 2010, and 2012, the Real Clear Politics average of the national House ballot underestimated the actual Democratic standing in the national House vote. So, there’s no reason to think the final national House ballot will overstate the Democrats’ standing in 2014.
There are plenty of reasons the Democrats won’t win back the House. But it’s not impossible that they will. If the same national environment that is producing a 4-5pt on the national House ballot still exists in a year’s time, Democrats may very well win back the House.