Peter K. Enns & Patrick Wohlfarth, The Swing Justice, J. Pol. (forthcoming 2013).
Whether quantitatively, qualitatively, journalistically, historically, or jurisprudentially, scores of papers have analyzed the “swing justice.” Is there anything left to learn?
Yes. In The Swing Justice, political scientists Peter Enns and Patrick Wohlfarth claim to make two contributions to the existing literature. By my count, it’s more like one-and-a-half, but that’s still a lot for a subject as picked over as this one.
Let me start with the first contribution: The authors show that in a non-trivial fraction of 5-4 decisions, the Court’s swing (median) justice does not cast the decisive vote. I score this a one-half contribution because that observation alone isn’t news. As other writings (including my own) have shown, some medians are far more powerful than others. In part that’s because the strength of the median has less to do with who occupies the “swing” seat and more with the Justices who sit close to the center. When several Justices are bunched together in the middle, medians are weak because either the left or right side of the Court can jump over them to create a coalition. But when medians are ideologically distant from the colleagues closest to them, they can emerge as “super medians”— Justice Clark in the late 1950s or Justice Kennedy today.
So why give Enns and Wohlfarth even a one-half contribution here? Because unlike many (though not all) of us, they don’t assume that there is only one median each Term. Swings, they tell us, can vary from issue to issue and so need to be identified separately for each one. This move doesn’t distinguish between super medians and all others; nor is it altogether clear that it’s necessary (see their note 13). But allowing medians to vary by issue does add some nuance to the analysis. At least it responds to a chronic complaint with the political science literature: that it does not differentiate among legal areas for the purpose of aligning the Justices from left to right.
Having identified the issue-specific medians, as well as the cases in which they did and didn’t cast the pivotal vote, Enns and Wohlfarth develop what I see as the chief contribution of the paper: trying to understand what makes swing Justices tick. It turns out that the swings’ ideology is not a significant predictor of their votes in 5-4 cases—but public opinion is. As the “mood” of Americans grows more liberal, the median is more likely to cast a liberal vote; as it grows more conservative, the median is more likely to cast a conservative vote. This is consistent with existing work, but there are some nice spins here. First, to my knowledge no one has focused explicitly on the influence of public opinion on closely divided (and, often, quite salient) cases; Enns and Wohlfarth fill this gap in fine form. Second, public opinion is not only a driver of the median’s vote; it also helps predict the pivotal vote even when it is not cast by the median. The suggestion here is that in some of the most contentious and important disputes, the “will of the people” plays a crucial role. (Hats off to Barry Friedman.)
There are other interesting findings I could mention—for example, the ability of the Solicitor General to move the median’s vote. The larger point, though, is that The Swing Justice manages to continue the conversation over the importance (or lack thereof) of the median Justice. I can imagine others now (re)joining the discussion—perhaps by evaluating the efficacy of Enns and Wohlfarth’s results for super medians versus all others, or by exploring some of the excellent suggestions for future research outlined in the article’s conclusion.