No doubt we’ll be talking about the 2010 election for a long time, and dueling explanations for the Democrats’ defeat will abound. Although I plan to make my own contribution to this explanatory surfeit, my topic right now is more modest—to trace the contours of what actually happened on November 2.
Let’s begin with the basics. In the midterm election of 2006, Democrats received 52.0 percent of the popular vote cast for House candidates, while Republicans received 45.6 percent. This year, projections indicate that the Republicans will end up with 51.8 percent, versus 45.1 percent for the Democrats—in short, a Republican gain of 6.2 percent and a Democratic loss of 6.9 percent since 2006.
One might hypothesize that these results reflect a selective partisan mobilization: Enthusiastic Republicans showed up to vote while depressed Democrats stayed home and pulled the covers over their heads. Not so. According to the 2006 exit poll, those who voted were 38 percent Democratic, 36 percent Republican, and 28 percent Independent. This year the split was very similar—36/36/28—which accounts for only a small portion of the popular vote shift.
Or maybe some Democrats were so disgruntled that they broke ranks and supported Republican candidates. No again: 93 percent voted for Democratic candidates in 2006; 92 percent in 2010. And by the way, 91 percent of Republicans for voted candidates of their own party in 2006, and 95 percent in 2010. Partisan polarization is alive and well.
What about age? The conventional wisdom before November 2 was that seniors enraged or terrified by changes in Medicare would turn out in droves to punish those who voted for health reform while young people disillusioned by Obama’s failure to create the New Jerusalem would abstain. That did happen, but only to a modest degree. Voters of ages 18-29 constituted 12 percent of the electorate in 2006; 11 percent in 2010. Voters over 65 were 19 percent of the total in 2006; 23 percent in 2010—noticeable but hardly decisive. If 65 and overs had constituted the same share of the electorate in 2010 as in 2006, the Republicans’ share would have declined by only .7 percent—about one-tenth of their actual gains.
We get more significant results when we examine the choices Independents made. Although their share of the electorate was virtually unchanged from 2006, their behavior was very different. In 2006, Democrats received 57 percent of the Independent vote, versus only 39 percent for Republicans. In 2010 this margin was reversed: 55 percent Republican, 39 percent Democratic. If Independents had split their vote between the parties this year the way they did in 2006, the Republicans share would have been 4.7 percent lower—a huge difference.
But why did they change? Here we reach the nub of the matter: The ideological composition of the electorate shifted dramatically. In 2006, those who voted were 32 percent conservative, 47 percent moderate, and 20 percent liberal. In 2010, by contrast, conservatives had risen to 41 percent of the total and moderates declined to 39 percent, while liberals remained constant at 20 percent. And because, in today’s polarized politics, liberals vote almost exclusively for Democrats and conservatives for Republicans, the ideological shift matters a lot.
To complete the argument, there’s one more step: Did independents shift toward Republicans because they had become significantly more conservative between 2006 and 2010? Fortunately we don’t have to speculate about this. According to the Pew Research Center, conservatives as a share of total Independents rose from 29 percent in 2006 to 36 percent in 2010. Gallup finds exactly the same thing: The conservative share rose from 28 percent to 36 percent while moderates declined from 46 percent to 41 percent.
This shift is part of a broader trend: Over the past two decades, moderates have trended down as share of the total electorate while conservatives have gone up. In 1992, moderates were 43 percent of the total; in 2006, 38 percent; today, only 35 percent. For conservatives, the comparable numbers are 36 percent, 37 percent, and 42 percent, respectively. So the 2010 electorate does not represent a disproportional mobilization of conservatives: If the 2010 electorate had perfectly reflected the voting-age population, it would actually have been a bit more conservative and less moderate than was the population that showed up at the polls. Unless the long-term decline of moderates and rise of conservatives is reversed during the next two years, the ideological balance of the electorate in 2012 could look a lot like it did this year.