Oprah Has the Power to Pick the Next President

According to a study by economists Tim Moore and Craig Garthwaite of UMD, Oprah is omnipotent. Though they aren't the first scholars to note this fact, no one has yet applied her influence to politics. The authors attempt to show that her endorsement swung the Democratic primary to Obama; it was worth between 423,123 and 1,596,995 votes with 95% confidence, and:

"Barack Obama received 278,966 more votes that Hillary Clinton. Given that 423,123, the lower bound of the estimated impact of the endorsement is greater than this difference, the results suggest that Oprah's endorsement was responsible for the difference in the popular vote between Barack Obama and Hillary Clinton."

The study used 'O' magazine subscription data as a proxy for Oprah's influence. The authors found that this variable was significantly positively correlated with both votes cast for Obama and overall turnout (both .95 confidence).

"For example, a 10% change in the county-level circulation of 'Oprah' is associated with an increased vote share for Obama of approximately .2 percentage points... In total, we estimate that the endorsement was responsible for 1,015,559 votes for Obama."

As a point of reference, county wide voting results were more influenced (per individual) by gender, age, and education than 'O' readership. Subscriptions were more predictive than factors like veterans population, urban/suburban mix, or unemployment, however. Holy Shit.

This study reinforces some notions that I had, and drops one huge bombshell. First, everyone knows that Oprah is powerful. She's been named to Time Magazine's list of 100 most influential people more than any other individual in history. She's the highest earning celeb year after year (270 million in 2007). Sales of books in her book club routinely jump 10,000% (!) upon inclusion. The list goes on and on. What I didn't expect, however, was the importance of celebrity endorsements to political outcomes. I doubt most observers would tell you that her endorsement effected the outcome, and that's what is so exciting about studies like this. They give you important information that isn't necessarily intuitive. This particular endorsement by Oprah probably represents the absolute upper bound of the impact of celebrity endorsements. She's the most powerful celebrity today not only in terms of media exposure and income, but also in her ability to drive consumer preferences amongst her fans. This makes her uniquely situated to effect electoral outcomes. Additionally, her political "brand" hasn't been contaminated by previous endorsements or political speech, putting her in a unique position to speak to many voters. I wonder whether other celebrity endorsements will be found to have a significant effect (the authors point to Chuck Norris's endorsement of Mike Huckabee as one possibility). Also, the authors point out that the celebrity endorsement effect may be most pronounced in primaries, where differences are mostly personal rather than ideological. After all, a non-politician celebrity is uniquely positioned to reinforce personal characteristics, but probably is not credible on the policy battleground.

I wouldn't say this paper is overwhelmingly convincing; the authors come up with a model and then think of everything they can to control for, rather than coming up with a particularly elegant framework. That being said, it's definitely good food for thought. If you've read this far, there's a chance that you might actually check out the study. Don't bother with the discussion of model building - it's not really necessary to understand either their methodology or results. Read the into and then skip right to the good stuff.