Consumption and (political) identity

So the Iowa caucus is upon is. What about it? Well, drawing parallels between the selling of political candidates and the selling of consumer goods is an old game, dating back at least to the Eisenhower era, and probably earlier. An iteration of this idea that’s gotten more attention lately is the (potential) connection between our consumer choices and our political ideologies. Over the weekend, a Times story about Democratic candidates digging for caucus-goers made passing mention of the Clinton and Obama campaigns relying on “sophisticated voter identification models, using detailed demographic and consumer data.”

In terms of political strategy, the interesting thing about these tactics is they have nothing to do with targeting the so-called “swing voter.” They are about identifying, partly on the basis of consumer behavior, people who are most likely to support your candidate. The effort then goes into making sure such people do so – that they actually go to the polls. It’s not about persuasion, it’s about motivation. (This is doubly important in the bizarre caucus system, of course.)

As an exercise in linking consumption and identity, what’s significant about it is that it only works if that demographic and consumer data really does give a good clue as to who someone will vote for.

Which sounds a little fishy.

Certinaly when it gets reduced to the extreme shorthand version. An example popped up in a recent New Republic piece about the Democratic nomination battle in Iowa, which quoted an unnamed Clinton operative, trying to spin Obama as out of touch with blue-collar workers, referring to him as “the arugula candidate.” Elsewhere there was mention of a pundit saying that, along similar lines, Obama has a “wine track” image that he needs to shed, in favor of a “beer track” image. This was followed by an anecdote suggested Obama has indeed tweaked a regular stump speech anecdote in which usually mentioned a glass of wine, to “a glass of wine or a beer.”

It’s easy to make fun of this as mere pundit-think – surely there is more to you or me than our appetite for arugula, or relative feelings about beer or wine. But when you get past the soundbites it turns out that the efforts to link politics and consumer behavior are pretty sophisticated … or at least complicated.

Another recent piece, in The New Yorker, mentioned the work of a firm called Strategic Telemetry, presently working for the Obama campaign, which “builds profiles of voters that include more than a thousand indicators, long strings of data—everything from income to education to pet ownership—that [the firm’s founder] calls ‘demographic DNA.’”

“The actual combinations that we come up with aren’t really anything that you could put on a bumper sticker,” Strasma told me. “You know, soccer moms or office-park dads. Sometimes people will ask to see the formula, and it comes out to ten thousand pages long.” When the demographic DNA is combined with polling and interviews with Iowa voters, Strasma is able to create the political equivalent of a FICO score—the number that creditors use to determine whether a consumer is a good bet to repay a loan. Strasma’s score tells the campaign of the likelihood that a specific Iowan will support Obama.

Several members of the Bush 2004 campaign team put out a book a year or two ago called Applebee’s America, which deals with similar material: Basically, that campaign’s mining of consumer data to figure out which voters to target. “If you’re a voter living in one of the sixteen states that determined the 2004 election,” the authors wrote, “the Bush team had your name on a spreadsheet with your hobbies and habits, vices and virtues, favorite foods, sports, and vacation venues, and many other facts of your life.”

According to the book, much of the mined data came from a company called Axciom, owner of the “largest collection of consumer data in the U.S.,” drawn from credit card companies, retailers, airlines, and “scores of other places where people do business.” In the election, the book said, Axciom gave (or sold to) the Bush team “a list … showing the stage of life (age, marital status, number of children, etc.) and lifestyles (hunter, biker, home renter, SUV owner, level of religious intrest) of each voter, drawn from a menu of more than four hundred separate categories.”

Again, the fact that the campaign was using this data to identify sympathetic potential voters (who as I understand it were then bombarded with direct mail and other more traditional entreaties to get them to the polls), significantly raises the stakes on the question of whether such demographic profiles are accurate. If a formula based on buying habits identifies the wrong people, a campaign risks motivating hostile voters – the worst possible outcome. And as the book described it, some weird things came up. “Dr Pepper is the only sugared soft drink that has a GOP-leaning consumer base,” the authors wrote. Also: Many republicans watched Will & Grace. They say the Bush team “didn’t know why” some of these patterns existed — and “didn’t care.” They just wanted it to work.

Did it? In the end, they say, their program “was able to predict with 80 to 90 percent certainty whether a person would vote republican, according to postelection surveys conducted by the Bush team.”

Obviously I have no way of knowing whether that’s true, or whether it’s politico bluster. But if it’s even close to true, it’s a pretty interesting statement consumer behavior revealing very surprising things. Interesting enough that I’m guessing the Dems are spending serious money trying to mine that data — and that we’ll see more such efforts in the future.