(The opinions expressed here are those of the author, a columnist for Reuters.)
By Reihan Salam
March 25 (Reuters) - Republicans are very enthusiastic about this year's midterm congressional elections, and it is easy to see why. Obamacare, the president's signature domestic policy legislation, remains unpopular. Turnout during midterm elections skews older and whiter than turnout during presidential elections, and Republicans tend to fare better among older and whiter voters.
And then there is the fact that Democrats are defending a number of Senate seats in states that tend to back Republican presidential candidates. Nate Silver, the editor of FiveThirtyEight, best known for his eerily good job predicting the outcome of last year's presidential election, forecasts that Republicans will retake the Senate.
So can the GOP sit back and relax? Not quite. As Chris Cillizza of the Washington Post reminds us, even if Republicans barely retake the Senate in 2014, the GOP faces a much tougher Senate map in 2016, when the electorate will be younger and more diverse. If Republicans want to achieve ambitious goals like replacing Obamacare and implementing pro-growth tax reform, holding the Senate for two years under a lame-duck president will do them little good.
Republicans need to think long-term. First, the GOP should offer a more compelling domestic policy agenda, as the National Review's Ramesh Ponnuru argues. The second, more prosaic step is to make better use of technological tools. Without a compelling agenda, Republicans won't deserve to win. But even with a compelling agenda, Republicans will have to embrace the technological, experiment-based revolution that has allowed innovative companies like Amazon, Google, and Starbucks to conquer their markets, and which GOP campaigns have been slow to grasp.
Two recent articles describe how conservative political operatives helped Republican David Jolly win a March 11 special election in a Florida congressional district that Barack Obama won in 2008 and in 2012. The first, by Alex Roarty in National Journal, attributed Jolly's success to the efforts of the National Republican Congressional Committee (NRCC). Roarty reports that the NRCC and its allies targeted the voters most essential to victory and hit upon a message - vote Republican or House Minority Nancy Pelosi will once again become House Speaker - that convinced them to come to the polls.
The second article, by Carl Hulse and Ashley Parker of the New York Times, told an entirely different story. Hulse and Parker focused on the role of Americans for Prosperity, a conservative political advocacy group backed by Charles and David Koch, the wealthy libertarian activists who have been vilified by Senate Majority Leader Harry Reid and others on the left as self-dealing plutocrats. According to Hulse and Parker, Americans for Prosperity transformed Jolly's district into a laboratory in which a variety of different approaches were used to impact the outcome of the race. (The Koch brothers' group is keeping the information proprietary, for obvious reasons.)
I don't know which story is closer to the mark. What I can say is that Americans for Prosperity has the better approach for the long term.
To win an election, you have to do a number of things - raise money, register voters, persuade voters to back your candidate, and get voters who back your candidate to turn out and vote. To do these jobs well, you have to develop clear metrics to gauge how effectively you're executing each of them. For example, you don't just want to know how much money you're raising; you also want to know how much money you're spending to raise a given amount of money. A fundraising strategy that works in one month with one donor might not work as well in another month with the same donor. The same basic logic applies with the other jobs as well. Registration efforts that work well with one eligible voter might not work as well with another.
In an ideal world, we wouldn't think about voters as part of broad, diffuse, ill-defined demographics, like "NASCAR dads" or "soccer moms," because these groups are extremely heterogeneous. The problem with thinking in a demographic box is that an Asian-American single mother radiologist from the Bronx might behave like another Asian-American single mother radiologist when it comes to a fundraising appeal, but she will behave like a retired Mexican-American police officer when it comes to voting or, say, sharing campaign messages on Facebook.
In the past, when computing power and data storage were more expensive, breezy generalizations about large groups of people were the best that campaigns could do. Yet generalizations can't tell you much about how real people will respond to stimuli. What campaigns really want is to be constantly learning how people actually behave - to learn not just what people will say in response to a survey, but how that belief translates into what they do. As the cost of computation and storage has collapsed, campaigns have more options. They can, like Americans for Prosperity, turn every campaign into a laboratory. As campaigns run more and more technology-enabled experiments, as they engage in a trial-and-error learning process, they will get better at learning from their mistakes and adapting to new circumstances.
Republicans often lament that Democrats are miles ahead of them when it comes to campaign technology. President Obama's 2008 and 2012 campaigns prided themselves on being state-of-the-art and Democrats often mocked their GOP counterparts for being stuck in the past. Thanks to the efforts of smart political professionals and academic political scientists, Democrats have a big head start on applying trial-and-error thinking to campaigns.
But Americans for Prosperity is demonstrating that conservatives can catch up. By deploying more experiments faster, they can, in theory at least, cram a decade's worth of learning into a much shorter period of time. And that will help the party do better - not just this November, but in many elections to come. (Reihan Salam)
Our Standards: The Thomson Reuters Trust Principles.