Political Ads: Overpriced, Inefficient, Essential
The 2012 political cycle is well on its way to being the most money-drenched of all time, as well as the most negative, with spending on TV ads – mostly of the scorched-earth variety – expected to top $3 billion.
That’s serious money, even by today’s lurid standards. But what are the campaigns (and increasingly, outside “Super PACs”) getting for their money? Not a whole lot, if you believe the latest conventional wisdom. “Three months from Election Day, some political strategists already are asking if more TV ads really will make a difference,” the Wall Street Journal noted recently, while NPR posed the question, “Do Negative Ads Make A Difference?” The answer: “Not so much.”
The media are belatedly coming around to something a lot of political scientists have been saying for a while: That the evidence “does not bear out the idea that negative campaigning is an effective means of winning votes”; that attack ads have “a strong negative influence on the viewer’s feeling toward the sponsor but only a slight net negative influence on feelings toward the target,” and have “strong but short-lived effects on voting preferences; that, basically, the claim that TV ads can change minds and votes is “pure speculation,” as George Washington University’s John Sides puts it.
In a sense, this isn’t all that surprising, because, really – who votes based on a TV ad? As The Atlantic’s Clive Crook wrote in June, in the wake of Wisconsin’s notably nasty and ad-heavy recall election, “I find it hard to imagine my own or anybody else’s mind being changed by endless repetition of stupid, annoying TV spots.”
But, in another sense, it’s shocking: $3 billion – and political ads don’t work!?
If that’s true, why do campaigns keep spending time and money on them? One reason, Sides argues, is that political consultants “naturally have an incentive to overstate the effects of ads and thus their own contribution to the election’s outcome.” Another, less cynical, possibility: “Political consultants and candidates are a lot like generals in war,” LSU professor Bob Mann told me. “They fight the battles with the weapons they know.”
But another reason might be that ads actually do work – or work well enough, enough of the time. “You’d be a fool to say [advertising] does or doesn’t work in some blanket fashion,” says Kathleen Hall Jamieson, the director of the Annenberg Public Policy Center at the University Penn. “It depends on the circumstances.”
Which is to say, it depends, most crucially, on which voters you’re targeting. Most voters are just not persuadable, period – they’re committed to one party or the other and virtually nothing is going to budge them. This is true even of so-called “independent” voters, who in fact tend to identify strongly with one party or the other; they’re independents in name only. True “swing” voters, people genuinely capable of changing their minds, and likely to do so, in the run-up to an election, are quite rare – just six per cent of Americans, or less than one-sixteenth of the electorate, according to a recent Washington Post/ABC News poll. Factor out the subset who live in solidly blue or red states (and consequently aren’t in any position to tip the election), and you’re left with about four percent of the voters in six states: 916,643 people, reckons political strategist Paul Begala.
These are the people on whom the election turns. Or, as Stephen Colbert recently quipped, “After two years and billions of dollars, our presidential election is going to come down to a few undecided voters in key swing states. The fate of our country in now in the hands of people who don’t think about what they want until they get right up to the register at McDonald’s.”
Another thing about these people: They’re “rather less knowledgeable about politics, and much more likely to say they follow news and public affairs ‘only now and then’ or ‘hardly at all,'” according to political scientists Larry Bartels and Lynn Vavreck (via the New Yorker‘s Elizabeth Kolbert). A lot of them won’t decide how they’ll vote until election day. Most of us are with Clive Crook: We can’t relate. “To voters who identify strongly with a political party, the undecided voter is almost an alien life form,” Louis Menand wrote in the New Yorker a few years ago. Undecided voters are influenced at the ballot box by all sorts of things, including, says a 2004 Princeton study cited by Menand, the weather.
These 916,643 enigmas are precisely the target of political advertising. “All the advertising dollars are being spent on the 10 percent of swing voters in the 20 percent of states that are battlegrounds,” Vanderbilt University political scientist John Geer told me.
And perhaps for good reason. If a swing voter can be swayed by the weather, why not by a TV ad? Of course, it’s inherently difficult to measure the effects of TV advertising. Whereas the online purchase of a product, say, can often be traced back to the clicking of an ad, the origins of a vote are far more mysterious and complex – especially given that any effect of an ad occurs only among the effects of existing political, economic, and strategic conditions.
Surprisingly, a lot of the research on political ads focuses not on swing voters but on voters in general. In one meta-analysis, for instance, political scientists at Rutgers looked at more than a hundred ad studies and found “no reliable evidence” that advertisements “work.” Not a single one of those studies tested the advertising’s effectiveness specifically on undecided swing voters.
But according to Jamieson, the real world offers some clues. She says that if you look at the data going back to 2000, and compare battleground states, where round-the-clock advertising is the norm, to non-battleground states, where there’s little or no advertising, you find that voters in battlegrounds were more likely to change their minds. The most likely explanation is advertising, she told me, because “nothing else happening out there that has mass exposure.”
In national elections as tight as we’ve become accustomed to, even a tiny target might be worth the massive investment. If only 4 percent – at most – of the voters who see most of the ads are persuadable, no wonder people think ads are ineffective! But in order to “work” – that is, to make a decisive difference in the outcome – advertisements need to persuade only a tiny minority. And although “if you’re a strong Republican, these ads aren’t going to move you very much,” Geer says, by contrast, “the ads can move [swing voters] around.”
So while TV advertising might not be a terribly efficient means of reaching swing voters, it’s probably more efficient than any alternative, since TV still has the broadest reach of any medium. “[Advertising] does have an impact, even if it’s a marginal one,” says Joe Heim, a political science professor at the University of Wisconsin-La Crosse. “And that marginal impact is worth it.”
And even that might be understating things. “Ads now are not just aimed at voters,” says Geer. “They are aimed at journalists.” Whether or not an ad influences TV viewers directly, it has the potential to “shape the narrative of the campaign.” “Probably no more than 1 percent of the American public saw the Swift Boat ad” sliming Democratic Sen. John Kerry’s Vietnam war record, he points out. But come the 2004 election, “80 percent knew the term ‘Swift Boat.’ That’s coming from the news media.”
Another reason why political strategists might not be so crazy after all to pour billions into advertising: Rival campaigns are locked in a kind of arms race. “If a candidate never aired ads, he or she would likely suffer,” Sides concedes. A recent example: When Team Obama aired its anti-Romney singing ad, Romney and Co. felt it had to respond in kind. If not, the media would no doubt have portrayed the Romney campaign as weak and flat-footed.
Campaigns “intuitively believe ads work,” says Jamieson. “And they’re afraid to test the hypothesis that they don’t. If you assume that, you’re taking a huge risk.”