Smart. Sharp. Funny. Fearless.
Wednesday, July 18, 2018

Feb. 19 (Bloomberg) — Yes, it’s the middle of the week, but it’s not too late for a little Monday Cranky Blogging, especially when some journalist is being especially obtuse. This time? It’s Josh Kraushaar, who is absolutely certain that Obamacare is going to be a major electoral factor in 2014.

How does he know? Well, he notes that the Affordable Care Act is unpopular, and it’s certainly correct that Obamacare polls badly. While there are months to go until November, it’s also true that the ACA is a bit more unpopular recently than it was earlier.

But then it gets very, very, tricky.

There’s an enormous gap between “Obamacare polls badly” and actual voting outcomes in November 2014. Does “Obamacare polls badly” lead to voters casting ballots for Republicans when they otherwise would have voted for a Democrat? Or to Republican voters voting who otherwise would have stayed home? Or Democratic voters passing by the local polling station instead of voting? It’s not impossible that a single issue, such as Obamacare, will move votes. But it’s unlikely.

What does move votes? Partisanship, of course. Party identification is a major factor in vote choice, even for many of those who think of themselves as voting for a candidate, not a party. Incumbency, certainly. In House elections, challengers are often little known, with incumbents benefiting from their rivals’ anonymity. Presidential approval matters — candidates from the president’s party are helped or harmed depending on how voters believe the person in the Oval Office is doing. Those are the major factors.

So how would an issue such as the ACA factor into voter decisions? In two possible ways. One is that voters who don’t like the law would hold members of Congress directly responsible. Political scientist Gary Jacobson has demonstrated that if voters like (or dislike) something about a candidate, including perhaps the candidate’s issue positions or actions while in office, that it can have an effect beyond party and other factors. A team of political scientists found that voting for health care reform cost House Democrats about six percentage points in 2010, enough (all else being equal) to produce a roughly 25-seat swing to the Republicans. The other way is that the ACA might influence President Barack Obama’s popularity, which, in turn, can influence vote choice.

I’m a longtime skeptic of that 2010 finding — even though the authors are friends of mine whom I respect a lot. Why? Because the study can’t account for what is an important counterfactual: What would have happened had Democrats passed some other signature legislation along partisan lines in 2009 or 2010. In my view, any high-profile Democratic bill at that point would have been vigorously attacked by conservatives, with very similar consequences. Unfortunately, there’s no way to prove that one way or another.