Elections with results as dramatic as this year’s are sometimes referred to as “realigning elections.” The term — although somewhat ambiguous and overused — usually refers to an election in which one party not only gains a significant amount of power, but coalitions also shift, the signature of which is usually that the rising party performs particularly well in certain geographic regions or among certain demographic groups.
The 1980 election, for instance, was arguably the beginning of a long-term shift toward Republicans in America’s suburbs, with Jimmy Carter’s share of the suburban vote dropping from 53 percent in 1976 to 37 percent in 1980: the 16-point swing against Mr. Carter was about twice the one he suffered in cities or rural areas. Likewise, in 1994, the shift against Democrats was particularly sharp in the South: 19 of the 52 representatives they lost came from that part of the country.
The 2010 elections, by contrast, were remarkable for their orderliness — and they tended to reinforce, to an almost uncanny degree, existing political coalitions.
Below is a chart that arranges America’s 435 Congressional districts from those (on the left) that gave the highest percentage of their vote to Barack Obama in 2008 to those (on the right) that gave the highest share to John McCain; the chart then compares which party each district had elected to the House before and after Tuesday night.
In 2008 — even though President Obama was on the ballot — his vote share was only a moderately good predictor of the results of the election to the House. Democrats won 16 seats in which Mr. Obama carried less than 40 percent of the vote, and 50 in which he had less than half of it.
After Tuesday, however — assuming that current results from the several races that remain uncalled by The Associated Press hold — Democrats will hold just 12 seats in which Mr. Obama failed to receive a majority of the vote (down from 50!) and just three seats in which he had 40 percent or less (down from 16).
Republicans, meanwhile, will now hold 55 seats in which Mr. Obama carried a majority in 2008 — up from 28. But mostly, they’re in places where Mr. Obama had only a slim majority. They’ll hold only 14 seats in which Mr. Obama had at least 55 percent of the vote (up from eight before). And they’ll hold just one seat, Illinois’s 10th Congressional District, in which Mr. Obama got at least 60 percent of the vote (he received 61 percent there instead); this figure is actually down from four seats before.
The next chart presents this data in a somewhat different way: by means of a logistic regression analysis that predicts the probability of Democrats controlling the seat based on Mr. Obama’s vote share. You can see that the regression curve is now much steeper. Whereas, prior to the election, every seat where Mr. Obama had won somewhat from 35 to 62 percent was in at least some reasonable degree of doubt — each party had at least a 10 percent chance of controlling such a seat — that range has now been narrowed to seats in which Mr. Obama took from 47 to 62 percent of the vote. (Put another way, Democrats lost almost every incumbent they had in districts where Mr. Obama had less than 47 percent of the vote.)
We can also go beyond the winners and losers to look at the share of the vote that each candidate received. In 2008 — in districts that were contested by both Democrats and Republicans — Mr. Obama’s vote share explained about 70 percent of the variance in the share of the vote for the Democratic candidate for the House.
In this year’s midterms, however, the two variables were brought into somewhat sharper alignment: the vote for Mr. Obama now explains 83 percent of the variance in the Democratic vote for Congress.
Rather than a realigning election, then, 2010 served as more of an aligning election: Congressional districts behaved less independently from one another, and incumbency status mattered less. Instead, they hewed tightly to national trends and the overall partisanship of each district. Most of the House incumbents whose districts had been outliers before (mainly Democrats like Representative Gene Taylor of Mississippi, whose district gave just 31 percent of its vote to Mr. Obama, but also a couple of Republicans like Representative Joseph Cao of Louisiana) were forced into early retirement.
What does this mean for 2012? Democrats — if they are expecting to do better than they did this time around — might actually be pleased that elections have become so strongly aligned to partisan orientation. They now have just 12 seats in which Mr. Obama won a minority of the vote to defend — whereas Republicans have 55 where he took the the majority. So if there is even a fairly modest shift back to Democrats in 2012, and the shift is again fairly uniform, they could be in a position to achieve quite a few gains.
Or, if the economy improves and — having facilitated a more even balance of power in Washington — the electorate becomes somewhat less angsty, the incumbent advantage could become stronger again, and the gains that Republicans made could prove to be relatively “sticky” — as they were, for instance, after 1994. Plus, Republican inroads in governorships and statehouses on Tuesday night should give them more leverage over redistricting, so they’ll be able to protect a few of their incumbents who otherwise might lose.
But generally it seems like we have entered a period in which races for Congress have become highly nationalized, and in which few potentially competitive races are conceded by either party and few incumbents are given a free pass. That could mean we’ll continue to see some wild swings over the next several election cycles.