Hunt for swing voters ignores an even more valuable group of voters

National Mail Voter Registration Form

And so the whole shootin' match comes down to around 4 percent of the voters in six states. [...] Four percent of the presidential vote in Virginia, Florida, Ohio, Iowa, New Mexico, and Colorado is 916,643 people. That's it. The American president will be selected by fewer than half the number of people who paid to get into a Houston Astros home game last year. [...]

That's Democratic strategist Paul Begala writing for the Daily Beast, bemoaning the campaigns' quest for swing voters and the outrageous sums of money that will be spent on convincing those few persuadables in the few states that can tip the election. In fact, according to his math, the $2 billion that will be spent on the election'in his mind, all of which is spent purely to sway those fewer-than-one-million minds'works out to $2,181 per swing voter.

Recent polling suggests that Begala isn't exaggerating too much, though. Swing voters have always been a small segment of the population, but at first glance it seems like there are fewer and fewer of them than ever. Polling back in spring of 2012, a point in the campaign where you might reasonably expect a lot of people to still be undecided, showed the vast majority of votes already locked down. Pew found that only 7 percent were truly undecided and not leaning in one party's direction or the other, while the first day of Gallup's tracking poll this year still found Barack Obama and Mitt Romney already taking over 90 percent of each of their party bases.

While it may be interesting to speculate on why there are so few swing voters any more'certainly the sorting-out of the parties into much clearer ideological and regional camps in the last few decades (with the gradual disappearance of conservative southern Democrats and moderate northeastern Republicans) has helped clear up a lot of people's uncertainty about where they belong, while the growth in news outlets with transparent partisan agendas helps reinforce existing political leans'the more important question becomes whether it's worth spending all that money on them. While no campaign should simply pretend swing voters don't exist and ignore them, eventually you reach a point of diminishing returns when trying to reach them (a point that's got to be somewhat lower than the $2,181 per swing voter cited by Begala), especially when there are other potential sources of votes which are not only potentially more cost-effective to tap but also potentially larger.

Alan Abramowitz, a political scientist from Emory University, is one of the most prominent swing-voter skeptics; he's come out with several articles in the last few months arguing that not only are swing voters are overrated as a voting bloc but that it's a better use of Democrats' time and money to focus on unregistered voters instead. We'll look at both of those arguments, starting with the idea that there are a lot fewer swing voters than we think there are (or at least than the news media encourage us to think there are).

To make this case, Abramowitz looks at a study from the 2008 election between Barack Obama and John McCain. The study, by American National Election Studies, followed a panel of voters for more than a year to see when and how they made up their minds. In the first survey the panel took, they were forced to choose between Obama and McCain; there was no "undecided" or "other" option, but they were asked to indicate whether they were extremely sure, very sure, moderately sure, slightly sure, or not sure at all about their choice. Seventy-five percent were extremely or very sure ... but 25 percent were either moderate or slightly sure or not sure at all. That's a lot of swing voters, right?

However, very few of the voters in that second pool'who you'd think were likely to switch, since their hand had been forced in having to choose someone in the first round'wound up changing their minds. All the mind-changing was basically reversion to the norm; Obama benefited slightly, gaining only an additional percentage point in support along the way. As Abramowitz puts it:

Only 8% of respondents switched candidates between June and November. These switches basically canceled each other out: 9% of McCain supporters switched to Obama while 7% of Obama supporters switched to McCain. Ninety-two percent of respondents ended up voting for the same candidate in November that they supported in June. [...] Nine percent of voters in the swing states switched candidates between June and November compared with 7% of voters in all other states.

The people who were likeliest to switch were, simply, those persons whose initial choice was out of whack with their own party identification. Only 1 percent of Democrats who supported Obama in June wound up voting for McCain, while 4 percent of Republicans who supported McCain in June wound up voting for Obama. On the other hand, 32 percent of Democrats who supported McCain in June wound up voting for Obama, while 39 percent of Republicans who supported Obama in June wound up voting for McCain. In other words, the people who in June seemed likeliest to swing, by the end, wound up not swinging at all, but just coming home to their usual party.

Abramowitz uses the data to break the electorate down into four groups: the stayers (who didn't change between June and November), who make up 92 percent of the electorate, returning partisans (voters who initially planned to swing but reverted to their usual party) at 5 percent, and departing partisans (voters who initially planned to stay loyal but then swung to the other party) at 2 percent. That leaves truly "swinging independents" (the group that the news media would have you believe hold the nation in the balance), who accounted for a total of 1 percent of the electorate. Most self-described "independents," Abramowitz points out, started out with a preference and stuck with it, consistent with many other studies' findings that "independents" actually are partisans, just ones who don't want to get saddled with a partisan label.

Over the fold, we'll talk more about the contention that Democrats should focus less on those few swing voters and more on voter registration and mobilization.