The 2010 midterm election saw a historically large percentage of voters claim to be conservative – 42 percent, compared to 32 percent in 2006 and 37 percent in 1994. Unsurprisingly, this has not escaped the notice of liberal analysts who promulgate the “Emerging Democratic Majority” thesis, which proposes that over time the electorate will naturally favor the Democrats. How do these analysts respond to this problematic (for them) trend?
Via a short essay for the Democratic Strategist, entitled “Is the Electorate Moving to the Right? Ruy Teixeira Says No,” we have their straightforward answer. The electorate has not moved in any significant fashion, and what we saw this November is nothing for liberals to worry about.
However, their reasoning on this line of inquiry is highly problematic. For starters, there is a subtle but significant change of subject from the title to the guts of the piece. Ed Kilgore (who introduces Teixeira’s argument) writes at the beginning:
It’s becoming more and more obvious that the big dispute at the heart of most arguments about the larger meaning of the 2010 midterms elections is whether the U.S. electorate is moving ideologically to the Right (sic) in a way that gives Republicans a natural majority in the future. And the very core of that dispute involves the behavior of self-identified independents, who obviously shifted towards the GOP between 2006-08 and 2010, and who seem to be exhibiting more conservative attitudes generally. [Emphasis Mine]
Ok, fair enough. My answer to that question is a pretty simple: The electorate has indeed moved to the right. The numbers are pretty stark.
Three points are evident from this chart. First of all, note the change in 1994, which remains an important election for understanding party alignment. Prior to 1994, the GOP would usually win the conservative vote with less than 50 percent of the points (somewhere between 65-35 and 75-25). Since then, the GOP has consistently won it by getting more than 50 percentage points, even in bad years for the party. That’s in part a product of the conservative South swinging from the Democrats to Republicans; 1994 was the first year since Reconstruction that the GOP won a majority of House seats in the South.
Second, more self-identified conservatives showed up to vote in 2010 than at any point since 1980.
Third, the Republican margin of victory in 2010 was greater with this group than.
So, open-and-shut case, right? The electorate is moving to the right. How to get around this? Here’s Teixeira’s answer (emphasis mine):
Has the public shifted sharply to the right ideologically? Conservatives say the 2010 election proves this, But (sic) careful analysis of available data shows there is far less to this argument than meets the eye.
Notice the shift in terminology – from “the electorate” to “the public.” And even then, the answer is still in the affirmative:
Conservatives turned out heavily for the 2010 elections but, among registered voters as a whole, the percentage of conservatives only increased by 3% between 2006 and 2010.
So, really the title of the piece should be: “Is the Public Moving to the Right? Ruy Teixeira Says Yes.” Their actual point of good news (for liberals) is that the public at large has moved to the right less than the electorate we saw in 2010. Their source for this claim is a Pew poll of registered voters taken in September. But what does Gallup say?
If we follow Gallup, we’d draw two conclusions: (a) the increase in self-identified conservatives in the broader public has increased more than Teixeira and Kilgore acknowledge; (b) conservatives were not over-represented as a share of the electorate in 2010, but rather were under-represented in previous cycles.
Now, I’m not saying that Gallup is right. But what I am saying is that if you’re going to make an argument that the electorate has skewed more conservative than the public at large, you should mention the widely regarded poll that doesn’t really support your point.
Beyond this, they go on to blur the distinctions between Republicans and conservatives to argue that politically the rightward shift in ideology does not result in much benefit to the Republicans. Teixiera asserts that the ideological shift in the public is “completely accounted for” by Republicans and Republican-leaning independents becoming more conservative.
Ultimately, I think they are on to something with this point, but again this is not the assertion they make in the title. It's one thing to claim that the electorate is not moving rightward, quite another to claim that it is but the GOP didn't pick up many votes from it. What’s more, by looking at registered voters rather than actual voters, they are likely overstating even this point. After all, the Pew poll they cite found independent registered voters favoring Democrats over Republicans by 3 points, whereas the final margin among actual voters on Election Day was R +19. What about the “pure independents” who actually show up to vote? Have they moved to the right? Teixiera and Kilgore cannot answer that question because of the constraints of their data.
Actually, I agree with the conclusion that Teixeira and Kilgore are making, even though I think their data and methods are hopelessly flawed. If you look at the historical record, you do not really see much by way of growth in self-identified conservatives after 1994. This makes sense. Ideology is a more stable form of public opinion, as it’s usually built upon core values. The year-to-year changes are probably due to shifts in the opinions of less informed voters who have not developed a full ideological system of thought. So, I’d be willing to cede that much of the bump in self-identified conservatives will fade over time.
But of course, Kilgore and Teixiera want to move from this point to argue against the idea that the midterm elections point to a “natural” majority (Kilgore’s term) for the Republican Party. This is a point I strongly disagree with, and the foundation of my disagreement depends more upon the geographical distribution of public opinion than anything else. Consider the 2000 presidential election, a year in which George W. Bush lost the nationwide popular vote by 0.5 percent yet still won the popular vote in roughly 240 House congressional districts. This is about as many districts as Barack Obama won, even though he won the popular vote by 7.3 percent.
My opinion is that the Democratic Party’s coalition has become too urban for it to sustain itself as a majority coalition in Congress over the long run. Prior to the Depression, the Democrats won when they united the rural South and West with just enough ethnic voters from the big cities. The Democratic super majority that began under Franklin Roosevelt was built upon the South and West, plus massive hauls from the cities. But nowadays the Democrats win the cities, but are much weaker everywhere else. This is important because in our system of government, the distribution of the vote matters. Democrats won the big cities by 65-33 in the 2010 midterm, meaning that their voters were clustered into safely Democratic districts. The Republicans won the suburbs and small towns by smaller margins, meaning that less of their vote was “wasted.” The GOP’s advantage, in other words, is more geographical than ideological.
This trend is only going to become more pronounced after the new district lines are drawn, because, for the first time in half a century, the GOP will dominate the redistricting process. That, combined with the fact that House seats are moving from Democratic strongholds like Massachusetts to Republican ones like Texas, will give the GOP an advantage for the next decade.
In a 50-50 year, I would bet the farm on the Republican Party controlling both the House and (depending upon what seats are up for grabs) the Senate. Now, don’t get me wrong: The Democrats certainly have the votes to force a 50-50 year, which has become the norm over the last several decades. However, their voters are distributed quite inefficiently in the cities and on the coasts, meaning that the Democratic Party wins 190 or so congressional districts by 60-40 or better, but often struggles to cobble together the remaining 30 or so needed for the majority.