Rick Perry's Not Always Big on Science, But He Did Use It To Win Some Elections
The scientists structured their research the way drug companies do -- trying different campaign methods and including a control group -- to figure out what voters respond to most strongly. Phone calls from campaign volunteers? Signs on lawns? Candidates actually visiting their town? The way these tests were conducted, and what worked best, are the subjects of a new e-book, Rick Perry and his Eggheads: Inside the Brainiest Political Operation in America, by journalist Sasha Issenberg.
The e-book is actually a chapter from Issenberg's full book on "the new science of winning elections," as he puts it, about how analytical tools like randomized trials are revolutionizing the way campaigns are run. The book, The Victory Lab, is due out in September of 2012.
But given the rest of the country's sudden, fervent interest in our governor, the Perry chapter was released on its own in late August, turning the 31-year-old East Coaster into "an accidental Perry expert," as he put it when we spoke this week.
"This is still a totally exceptional story, what the Perry folks did," said Issenberg, who's written for The New York Times, Philadelphia magazine, the Boston Globe and Slate. "As best I know, there are no other instances of a candidate campaign doing this kind of experimenting the way Perry did it."
And, sure, Issenberg will be the first to admit he's benefited from good timing. "I'd reported and largely written this to work as a book chapter," he explained. "And I had it sitting on my hard drive as he made the decision to run. I think it'd be a lot harder for someone to crack that circle now and get them to talk."
Not that Perry's campaign meant for the experiments to be hush-hush. Carney read a book in 2006 by pioneering election scientists Alan Gerber and Don Green. The book, Get Out the Vote!, tried to answer some of these same questions. As it turned out, in-person visits worked best, mail was much less effective, and phone calls didn't work at all. But both scientists were frustrated that they hadn't had more opportunity to work with real-world campaigns.
So Carney approached them with a deal: Run their experiments on Perry's campaign. When it ended, and only then, they could publish their research wherever they wanted. "This wasn't supposed to be secret forever," Issenberg said.
Carney was mainly bugged by the amount of waste in campaigns, the fact that millions of dollars in fundraising often goes to methods whose efficacy is questionable, like sign planting.
"There is no really good way, short of these randomized experiments, to measure the effect" of these things, Issenberg told Unfair Park. "And the culture of campaigns is not structured to even be skeptical of them -- that's the thing that got to Carney over the years."
But hiring the scientists was still a daring move. First, they were from Yale, of all places.
"No political campaign had ever extended such authority to academic outsiders," Issenberg writes, "and it was all the more surprising because Perry's exuberant anti-elitism was quite specific in its disdain for Ivy League credentials. 'You don't have to have a PhD from Harvard in political science to understand our economics,' he'd say."
Also, Gerber and Green's previous research had always been with liberal groups, like ACORN and the NAACP Legal Defense Fund. But the campaign ended up hiring two GOP-approved election scientists to work with the Yale-ites: Daron Shaw and James Gimpel. Shaw was hired by Karl Rove to consult on George W. Bush's first gubernatorial campaign, while Gimpel was a consultant for the Republican National Committee.
The testing by the quartet of scientists found that the thing that bumped poll numbers the most was when Perry visited a town and gave an in-person stump speech. In every market where he did this, the local media coverage was almost always positive, and his approval ratings went up. Unlike the slight boosts he saw from TV ads, visiting a given market created a lead that Perry held onto week after week. Perry was reelected in 2007, in a year when Republicans lost control of both houses of Congress and many gubernatorial elections nationwide. In his next campaign, the 2010 gubernatorial primary against Kay Bailey Hutchinson, Perry's camp used the same methods and saved an estimated $3 million while beating Hutchison by more than twenty points.
The question of how much you can "universalize" Perry's election results in 2006 and 2010 is still open, Issenberg said. Issues such as the economy and voters' feelings about Obama are probable going to be much bigger factors in a general election.
As Issenberg put it: "If Perry looks like he's outside the ideological mainstream, especially in swing states, it doesn't matter how good their science is."