We’re now far enough beyond November 6–and the distractions of the holiday season–to venture some basic judgments about what we learned from the 2012 election cycle. The first I’d offer is not a new lesson, but a confirmed one: For all the endless hype and more-intense-than-ever media coverage of what was often described as a tense and turbulent campaign, the results were almost exactly what you’d expect given the existing party coalitions, a “down” but modestly improving economy, and a significant if not overwhelming Democratic advantage in both strategy and tactics.
The composition of the electorate in 2012 was in almost every respect remarkably similar to that of 2008, despite constant predictions that it would differ, usually in ways undermining Obama’s prospects for victory. The party ID breakdowns were almost identical. So, too, were the racial breakdowns (a slight increase in the percentage of the vote for Hispanics and Asians, and a 2% drop among whites). The percentage of the vote cast by the various by age cohorts barely changed, either; under-30 voters were actually a somewhat higher percentage of the vote in 2012, contradicting expectations from just about everyone other than the Obama campaign itself.
In terms of candidate performance in this stable electorate, Obama generally lost vote share where you’d expect it given the economic situation and the likelihood of 2008/2012 switches among “wrong track” voters: among independents (from 52% to 45%); whites (43% to 39%); college-educated whites (47% to 42%); men (49% to 45%); white men (41% to 35%); and under-30s (66% to 60%). Each and every one of these vote-share losses were cited at various points during the campaign as being incompatible with victory, primarily by analysts who were either using outdated profiles of the electorate, or who assumed (from an assortment of questionable estimates of turnout based on questionable measurements of “enthusiasm”) an electorate closer to 2010’s than 2008’s.
It is impossible, of course, to deduce how much of the 2008-level turnout in Obama’s coalition should be attributed to the campaign’s undoubtedly effective GOTV operation. But there are indications it was pushing an open door. To cite one example: in Indiana, a state that was an intensive focus of the 2008 Obama campaign but written off early in 2012, the under-30 percentage of the vote increased from 19% to 20%, the African-American percentage from 7% to 8%, and the Hispanic vote from 4% to 6%.
In any event, there is little evidence that the many ballyhooed events of the campaign, or the occasional turbulence in the polls, had a residual effect on the outcome. The extremism exhibited by the overall candidate field during the Republican primaries definitely reinforced the president’s efforts to turn the election into a “choice” rather than a straight-forward “referendum.” But primary-season polls showing Romney trailing Obama by large margins–mainly because of relatively low approval ratings from conservatives–turned out to be ephemeral and meaningless. Monthly job creation reports–nearly all of them trumpeted as large developments in the campaign–almost never produced any significant effect on support for the two candidates. The Obama “bump” produced by the two conventions was no more long-lived than the Romney “bump” after the first debate. The endlessly discussed predictions that late undecideds would break decisively to Romney never panned out; Obama’s margin among voters reporting they had decided in October and November (4%) was almost the same as among those deciding earlier. It was all mostly noise, and in fact, the illusion of a tied race or a Romney lead during the last month of the campaign was largely attributable to LV screens that more often than not produced less accurate measurements than the RVs surveys they “corrected.”
Another illusion of the campaign was that battleground states were behaving differently than others because of ad saturation and the heavy deployment of GOTV resources. But most of those perceptions faded at the end. Obama’s margin of victory in Ohio dropped from 5% in 2008 to 3% in 2012: a nearly proportionate decline as compared to national margins. His margin in Virginia fell from 6% to 4%, again closely reflecting the national percentages. In Iowa, for all the talk of the state being razor-close in 2012 or leaning to Romney, Obama won by 6%, as opposed to 9% in 2008–about two points above the national margin in both years. Colorado, considered even more perilous for Obama in 2012, wound up giving him a 5% victory as opposed to 9% in 2008. Given the heavy concentration of ads, GOTV, and campaign events in the battleground states, this is additional evidence that the campaign itself did relatively little to shift votes as compared to the overall strength of the candidates and their parties all along.
This does not, of course, mean that campaign resources were wasted; even that most over-rated of presidential campaign weapons, paid advertising, can become deadly effective if unopposed. And reports of robust early voting by the Obama coalition–which countered months of predictions that younger and minority voters could not possibly match their 2008 “enthusiasm”–may have had a slight but significant effect on the final turnout patterns. But all in all, the evidence supports, as it often has in past campaigns, that the vast media attention devoted to every twist and turn in the candidate competition supplied more entertainment than information.
As a final disclaimer: my minimum high regard for campaign events and activities does not extend to the basic dynamics of Democratic and Republican politics, or to the basic strategies deployed by both sides, which particularly in Obama’s case could be traced well back to the beginning of the cycle and even to his 2008 campaign. The next “lessons learned” post will discuss those factors in some detail.