One of the strangest of American rituals is the habit of analyzing election outcomes by slicing up the country's demographic pie into little groups and then proclaiming one or another slice to be absolutely crucial to the electoral future of the country. In the cruder media climate of the 1990s, this tended to involve devising cutesy names for your preferred groups. Thus, we had the "angry white men" of 1994, the "soccer moms" of '98, the contrasting "waitress moms" and "office park dads" strategies of 2000, and the "NASCAR dads" of 2002. The Internet and its attendant proliferation of media outlets -- combined with, perhaps, growing laziness on the part of pollsters -- led more recently to a decline in name-quality. With the exception of a fleeting "mortgage moms" bubble this fall, allegedly pivotal demographic groups have tended to go by simpler labels in recent years -- parents of small children, married white women, and the ever-elusive white working class.
2006 has proven no different in the eyes of most commentators. At a post-election forum hosted by the Democratic Leadership Council's affiliated think tank last week, Bruce Reed crowed that after a disastrous showing in 2004, where Democrats won only among high school drop-outs and people with post-graduate degrees, in 2006 the party won or tied in all educational groups. This, he said, proved the wisdom of "macro-targeting the middle class" rather than Karl Rove's fancy "microtargeting" or some kind of base mobilization strategy. Lurking here, however, is a fairly basic fallacy. There's no electoral college for educational sub-categories -- it doesn't really matter whether you "win" or "lose" some specific group. Aggregates determine the winner.
Besides, if you compare the 2004 House exit poll to the 2006 House exit poll it's hard to discern a particular Democratic surge in the middle. Nationwide, Democrats did about five percentage points better with the population as a whole. Breaking it down by the middle educational sub-groups, Democrats saw a six point increase among high school graduates, a four point increase among those with some college, and a three point increase among college graduates. Polls, obviously, have margins of error, and the margin of error for sub-categories gets fairly large. None of these results, in other words, differ meaningfully from overall Democratic performance. And, indeed, the party saw its share of the post-graduate vote go up six percentage points -- again, the same as the national trend. There either was no strategy of targeting the middle class, or else the strategy didn't work.
The same holds if you look at class in terms of income. Democratic performance improved among every group, and improved among every group by about the same margin. The only sub-slice that behaved differently from the others in a meaningful way was the tiny share of voters (five percent or fewer of the total) who earn over $200,000 a year. This group is so small that it's not clear how accurate our information about them is, and nobody's ever suggested that the Democrats did -- or should -- design electoral campaigns aimed at appealing to a tiny wealthy elite.
Nevertheless, the demographic slice-and-dice is a habit the country can't seem to kick. On Saturday, The Washington Post's Allan Cooperman reported that "religious liberals contended that a concerted effort by Democrats since 2004 to appeal to people of faith had worked minor wonders, if not electoral miracles, in races across the country." The trouble, as Kevin Drum points out is that there's no evidence that this is the case. Democratic performance among white evangelicals and among weekly churchgoers did improve, but only by three percentage points. That's slightly less than the party's overall improvement, though the difference isn't necessarily meaningful.
Democrats won, in essence, the old fashioned way. They pounded the GOP for its embrace of a few crude, broadly unpopular things -- massive corruption and an obviously failing war. They pushed a few crude, broadly popular issues like ethics reform, cheaper drugs, and a minimum wage hike. As a result, they got slightly more votes than they previously had from all sorts of people. (The only really dramatic demographic shift stemmed from the Republican decision to try to exploit public fears of immigration; this backfired, and undid recent GOP efforts to woo Hispanic voters.)
Debunking the sub-slice myth is important not just because that myth is annoying. The real problem is that the slice-and-dice approach to politics is often used as a justification for advancing silly substantive policies. The year 2005, for example, saw an outbreak in some quarters of mass panic about Democratic performance among white married mothers of young children. It's hard to think of good ideas designed to appeal to such a specific demographic, so people came up with a bad idea -- demagogic Hollywood-bashing, censorship of video games, and vague promises to "clean up" cable and satellite television. Thankfully, this fad wound up dying down in favor of a campaign waged over broader concerns like national security. And focusing on things that actually matter wound up working just fine.
Going forward, that's the closest thing to a real "lesson" 2006 has to offer. The Republicans, in the first instance, lost the election for themselves, and they lost it by governing the country very poorly. They implemented policies that got lots of people killed, left wages stagnating, and generated tons of corruption. Democrats, as an opposition party, were able to ride to victory largely by pointing this out. Insofar as they gain a greater share of power, the Democratic task starts to become governing in a manner that produces results that improve overall conditions in the country. It's not, after all, as if "security moms" have some special distaste for launching failed wars that "office park dads" don't share, or that the white working class gets mad when the economy goes down the toilet while the black middle class thinks it's great. The baseline political loyalties of different groups vary a great deal, but insofar as things change from year to year, they tend to all change in the same direction. The two parties are fairly evenly matched, and have been for years now. Under the circumstances, big, obvious events -- the Clinton impeachment, 9/11, Iraq, etc. -- and the public response to them make all the difference.
Matthew Yglesias is a Prospect staff writer.
If you enjoyed this article, subscribe to The American Prospect here.
Support independent media with a tax-deductible donation here.