Kevin Drum’s evident confusion regarding how advertising works in political campaigns got me thinking that there has to be a simple way to convey why political scientists generally discount the impact of individual campaign ads on election results. So, in the teaching spirit, let me try to put it in story form, using characters with which most Americans are familiar. (Many thanks to Pete Cahill for providing the illustrative links.) Now, keep in mind that this is a fictional story designed by me to illustrate some basic political science findings. The lessons are real – but the story is not. Got that? (I don’t want to deal with any lawyer types.) Ok, now pay attention.
Let’s start with the typical American voter – Kim Kardashian.
Like most Americans, Kim wants what is best for the country: a good reality show, unlimited clothing allowance, extended hours for dance clubs and, most important, money to pay for this along with all the free publicity she can get.
Now, assume there are two candidates vying for her heart….er….vote. There’s the Republican, professional basketball player Kris Humphries, and the Democrat, recording artist Kanye West. Both make the case that if selected they can get her all the free pub she could ever need and keep her living in the style to which she is accustomed. As evidence, Humphries releases the following campaign ad, a compilation of his on-court highlights.
With a skill set like this, he argues, he will be a perennial all-star who will attract mega-endorsement money and, not insignificantly, a bushel of publicity. Who wouldn’t want to hang with this guy? This is an effective highlight reel, and Humphries makes sure to play it wherever Kim “America” hangs out. The media takes note of this, and agrees that Kris has all the “momentum” in this campaign due to his skilled campaign tactics.
But wait! Kanye is no fool. He puts together his own campaign video highlighting his knowledge of “power”, part of an all-out publicity effort designed to familiarize Kim America with his impressive track record of chart-topping hip-hop hits and music awards . With these awards, of course, comes a very lucrative recording career.
Note two important aspects of these “campaign ads”. First, Kris focuses on his record as a basketball player, and Kanye on his hip-hop accomplishments. That is, neither attempts to create an artificial version of himself by highlighting nonexistent accomplishments – Kris doesn’t pretend to sing and Kanye avoids dunking highlights. There’s a reality out there – their actual career records – that serves as a limiting factor on what they can put in their ads. Second, both campaign spots are designed to activate latent predispositions within Kim America. That is, they don’t try to persuade her to adopt the lifestyle of a convent nun who has taken a vow of poverty. They aren’t trying to change her views – to make her think differently – so much as they are framing their own record in a way that is designed to show how it addresses Kim’s existing attitudes. Kim, they know, craves free pub and a certain lifestyle – and so they sell themselves accordingly. Again, reality – Kim’s preexisting views and needs – constrains what they can do with their campaign advertising.
But our candidates don’t stop there. Kanye decides that in addition to highlighting his own accomplishments, it will help if he can denigrate Kris’. So he puts together a negative ad highlighting Humphries’ failures – here Dwight Howard eats Kris’ lunch with a monster block and turns it into two points for the Magic:
Now let’s add a fourth character to our drama: Kay-Drum. He’s a hip-hop pundit who strongly supports West in this bid for Kim America’s vote. When he sees West’s negative ad, he describes it as “devastating” to Humphries’ chances. The media takes note – Kay-Drum is an expert, after all, and he is quoted everywhere. Poor Kris – he claims that this ad misrepresents his record and besides, the score was 2-2 at the time! Too bad! The media decides Kris is a whiner.
But Humphries has his own attack machine, and he runs an ad revealing that President Obama, of all people, called Kanye West a jackass for interrupting Taylor Swift’s acceptance speech at a video awards show. When Kay-Drum sees this ad, however, rather than changing his views, he instead describes Humphries’ negative ad as “an act of moral depravity” by a “desperate” candidate who knows he can’t win Kim’s heart, er, vote. The media, carefully reporting how the “experts” are responding, decide that Kris’ tactics aren’t working, and that Kanye has run the more effective campaign.
In the end, much to Kay-Drum’s delight, Kim ends up “voting” for Kanye, and they live out their lives as a happy, slightly hefty couple. Humphries, meanwhile, ends up in Brooklyn on a loser team and fades into obscurity. (Reminder: this is a fable designed to teach – these events and characterizations are made up!)
What made the difference? Why did Kim choose Kanye? In the media post-mortem, leading journalists and experts like Kay-Drum make it clear that the more skilled advertising campaign run by Kanye made all the difference. Specifically, many scribes cite the campaign ad in which Dwight Howard “Swiftboated” Kris’ attempted dunk as the turning point in the campaign. It goes down in campaign lore alongside the Daisy and the Willie Horton ads and reinforces the media’s preoccupation with campaign strategy and tactics.
Now here’s the kicker: political scientists – long before any of the campaign ads came out, including the celebrated Howard “Swiftboat” rejection – predicted that Kim would, in the end, choose Kanye over Kris. How did they know this? By ignoring the campaign, and focusing on the fundamentals. First, they started with the simple fact that Kim America’s vote would likely depend on one factor: which candidate could provide her with her the resources necessary to maintain a lavish lifestyle and, not incidentally, receive lots of pub. True, she talked about love and Kanye’s hidden talents and how Kris getting rejected really opened her eyes to his shortcomings, and clothing accessories too, but all of that, basically, was a rationalization of the reasons that really drove her choice: money and face time on TMZ. Political scientists, having already constructed a simple formula based on previous campaigns that estimated the likely career earnings of a hip-hop star versus that of a basketball player, simply added a few variables to account for Kris’ and Kanye’s particular careers (any drug use, love of guns, weak knees, etc.) and came up with their prediction that Kanye was going to easily best Kris in the earnings and free pub categories, and that Kim would choose accordingly.
Note that in constructing this model, they didn’t worry that the campaign advertisements might skew the results. Why were they so confident? Because those ads couldn’t create an alternative universe – they could only frame the existing one. By creating measures of that reality – the earning potential of the two candidates – and assuming both sides would effectively frame their own earning potential, and denigrate the other person’s – political scientists assumed that the independent impact of the ads would largely cancel each other out, and that the final result would reflect the fundamental earning disparity between Kris and Kanye. A key assumption here is that both candidates run highly effective campaigns within the constraints imposed by reality. A second key assumption is that in a high-information environment, with lots of alternative sources by which to evaluate both candidates’ claims, Kim America would recognize the earning disparity between Kris and Kanye. This doesn’t mean she had to understand the economic intricacies of either profession – she just had to be reasonably confident that she knew which person would earn more. And in fact, by hanging out with friends at the clubs, and talking to members of her retinue and Khloe and Bruce and her hairdresser, and hobnobbing with Jay-Z and Beyoncé and Snooki, in my completely hypothetical universe she was able to come to a reasonably well-informed assessment.
And the rest is made-up history. Don’t they look happy? And well fed?
That my friends, is how political science works!
Any questions?
Have your go-to sources for informed political analysis – the ladies of The View – weighed in on the critical “have a beer” test yet? Surely if Joy Behar and Whoopie Goldberg preferred to drink a beer with Kanye over Kris, we would have the makings of a kairotic moment in Kim’s ultimate decision. Suppose then, Elizabeth Hasselback makes an unfortunate slip of the tongue as she expresses a preference to drink a beer with Kris. Since it is well known she prefers the athletic type, her blunder is a manifestation of Kris’ true shortcomings. In fact, the blunder is not only a manifestation of Kris’ shortcomings but of all athletes and those who prefer athletes over rappers in general.
Kay-Drum might be guilty of journalistic malpractice if he failed to include such critical information in a post-mortem. At a minimum he would have a whole lot less to write about both before and after the “big event.” Worse still, without Kay-Drum’s insightful analysis of “devastating” ads or Joy and Whoopie’s beer preferences, Kim America might decide she’s just not feeling it anymore. Maybe she is thinking rapper’s are overrated. Could it be Kim regrets that hasty divorce? Kay-Drum is not sure he can take that chance. 🙂
Mary Sue – Your analysis is so far ahead of mine on the pop culture scale that I will simply sit back and marvel rather than even pretend to muster a response.
But you political scientists now need to take it further. What if we know that Kris and Kanye are jealous types? “If I can’t have her, nobody should.” How would Kris behave knowing that a girl like Kim is extremely likely to prefer Kanye? Wouldn’t he try to short-circuit the entire proceeding by poisoning Kim’s interest in dating at all rather than trying to poison her preference? To me, this is largely what elections are about – both parties have an interest in some voter apathy (just different voters) and (more baldly from one side) voter suppression.
Vijay – In a close election, of course, one can cite any numbers of factors that might be determinative. Generally speaking, we think turn out operations (GOTV) are more consequential than voter suppression tactics. Again, this is one of those forest vs. trees perspective; while the media pays a lot of attention to efforts to enforce voter ID laws, ACORN fraud, etc., political scientists think the marginal impact of these factors is small (although again, in a close election….) Of greater consequence, we think, is the relative ability of the two campaigns to get people to the polls. To a certain extend, however, turnout is endogenous to our prediction model; factors, such as the state of the economy, are good predictors because they presume increase turnout on one side relative to the other.
Professor,
I admit that, as PSCI 104 student of two years ago, I may have forgotten a couple of things now that were on the final exam, so forgive me if I should know the answer to my question. I have kept up with your blog, however, so I feel pretty informed of your stance on campaign effectiveness.
My question is this – would you say that, instead of reading mainstream political blogs every day, I should just wake up a few weeks before the election and look at the economic numbers to understand fully the election, paying attention to absolutely nothing else? With that extra time, I could spend the next few months mounting a grassroots campaign to fire Bobby Valentine – no doubt a better use of my time.
So what, if anything, SHOULD I pay attention to that doesn’t have to do with the economy? If all that matters is the economy, a force which remains outside the control of either presidential candidate, would you say campaigns themselves are essentially useless? Are campaigns only useful in terms of showing you are, at the least, a viable candidate and increasing voter turnout from your respective base? After that, should a candidate merely cross his fingers and hope for the best? As you say, in a close election, perhaps some other minor issues might make a difference, but you don’t seem so convinced.
If all that matters is the broad strokes of the economy, I will happily spend more of my time listening to Kanye West albums and making fun of Kris Humphries. What could be more American anyway?
Will – To be clear, no political scientist that I know – and certainly not me! – believes “campaigns don’t matter”. so, for example, if Obama campaigned all out, but Romney sat on the sidelines, spending no money,running no ads, mounting no get out the vote effort, I am confident Romney would lose. However, when both sides mount effective campaigns, and frame the elections as effectively as possible, given their own records and the underlying fundamentals, the relative impact of either campaign on the outcome is marginal, at best. Your comments, and those coming from lots of readers, convinces me that I need to devote a more complete post to this topic. give me a couple of days. In the meantime, start the “Bring Back Francona” petition drive!
Okay… I’m going to be THAT guy…
For what it’s worth, the manager is also believed to have a marginal impact on baseball seasons – 2ish games. Don’t get me wrong, I MUCH prefer Tito and the way he deals with the media/public, but considering the team has had players miss ~3 years of time on the DL (mostly key players), it’s impressive they’re as close as they are.
Not saying it’s BobbyV being awesome, but I can’t say he’s made things worse.
Vijay – In truth, I’m in complete agreement with you. In fact, I could make the argument that Bobby V. has done a remarkable job with the bullpen in particular, and – considering the injuries, can’t really be blamed for the .500 level performance to date. For that, I point the finger first at the starting pitching…..but it is traditional to blame the manager. It’s a lot like blaming campaign ads when someone loses the presidential race!
Maybe it’s not for you (possibly for a behavioral economist), but I’d be interested to know the political scientist’s position on why fans tend to almost always sympathize with management. Whether it be Celts fans trying to kick Ray Allen on his way out of town for NOT being a greedy ass, or Knicks fans doing the same at Lin even knowing the Knicks kicked him to the curb.
Blaming the manager seems to be the same thing in my mind…
Not “if”, “I’d”…
“However, when both sides mount effective campaigns, and frame the elections as effectively as possible, given their own records and the underlying fundamentals, the relative impact of either campaign on the outcome is marginal, at best.”
OK, but doesn’t it matter how close the fundamentals are? Let’s take the three most recent presidential elections as examples. In Bush-Gore, the fundamentals were very close and the election historically so. That means that every decision that either of the campaigns made could be credited/blamed with the result. Cheney’s debate performance, Gore’s decision not to use Clinton, and obviously anything in Florida all could be credibly cited by pundits as resulting in the Bush Presidency. Of course, when everything makes a difference, the argument quickly devolves to absurdity.
In the Obama-McCain race, the fundamentals predicted a Democratic victory and indeed Obama won handily. Here, almost no tactical decision in the campaign made a difference. Even the Palin nomination, which certainly hurt McCain can’t be blamed for his loss. In elections like this one, I think your point is completely correct, pundits need something to talk about, but it is all blather as none of the events during the campaign made a difference. Only something incredibly significant (a terrorist attack, a major scandal) could have altered the result. This also seems to apply to Kanye-Kris.
In between these two cases is the Bush-Kerry race, which is where Obama-Romney seems to fall. The fundamentals are ambiguous here, I tend to think they favor Obama (slightly improving economy, incumbency, solid personal approval ratings) but a reasonable argument could be made that they favor Romney (particularly the unemployment rate). In a situation like that, admittedly most tactical decisions like the Bain campaign will not change the result but it is conceivable that some will make a difference. For example, while you note that Swift Boating may not have made a difference, I’m willing to bet that the “flip-flop” campaign the Bush folks waged did (abetted by Kerry’s “I voted for the $88 billion before I voted against it.”)
The problem of course is that pundits who get paid to overemphasize small details will treat every tactical decision like the campaign is Bush-Gore, when that was a once-in-a-lifetime contest. Still that doesn’t mean that the campaigns aren’t making some decisions that aren’t important and may affect the final result.
(One final thought . . . even if tactical decisions don’t affect the final result, they may affect the final margin and that could affect close Congressional races. Small consolation to the loser of the presidential race to be sure but not meaningless, especially to the winner)
Stuart – Yes, you are exactly right. In a close election in which the fundamentals show no clear favorite, tactics can make a difference. Many political scientists think this happened in 2000, when Gore ran, in their view, an awful campaign. See my later post on when and why campaigns matter.