Does the Eighth Amendment Turn on the Size of Your R-Squared?

Check out the dispute between Justice Stevens and Justices Scalia (and Thomas) in Baze v. Rees. Both sides concur in the Court’s judgment, but Justice Stevens, arguing that the death penalty fails a CBA, notes:


And Justice Scalia, responding to the deterrence point, writes:



Earlier this semester, I taught a class on the empirics of deterrence (with a focus on the death penalty). It strikes me that Justice Scalia has the better of the argument here, if he is read to say that knowing whether the death penalty deters (or not) sounds like the kind of question that is answerable with data, but probably is not, at least for now. As John Donohue and Justin Wolfers, who Justice Stevens somewhat ironically relied upon, argued:


Further, Supreme Court Justices don’t have the training or staff necessary to sort through competing empirical studies and reach a definitive conclusion. (Justice Stevens’ weighting of studies notwithstanding). And even were the Court to appoint a “special statistics master,” can a constitutional question of this magnitude turn on econometric rabbit-holes?

You may also like...

3 Responses

  1. A.J. Sutter says:

    Apropos of “can a constitutional question of this magnitude turn on econometric rabbit-holes?,” wave after wave of clerks will come from law school with econometrics and other L&E doctrine freshly in their brains. When you teach them L&E-style analysis do you also teach them when to shut it down? What kind of discrimination do you teach students in applying economics to law?

    My impression from journal articles and from L&E textbooks like Cooter & Ulen is that lawyers are far more credulous about the power of economics than economists even. (I’ve seen much more extensive critiques of the Coase theorem in environmental econ texts, e.g.) Does it literally take a life-and-death case to make L&E’s possible drawbacks evident?

  2. Alice Ristroph says:

    I tend to agree that the empirical evidence doesn’t tell us whether the death penalty deters, increases, or has no effect on homicides. But I think Stevens would agree with that, too. Stevens’s concurrence doesn’t claim that the evidence establishes no deterrence. Instead, he says the evidence fails to establish deterrence. Absence of evidence is not evidence of absence…

  3. Richard Bourke says:

    @Alice: Correct about Stevens’ argument; if no evidence is _statistically significant_ then no evidence exists (either way).

    It is the age old question – do you execute someone that you are 90% sure are guilty, or do you let off those you are certain 10% certain of being innocent?

    Personally I would rather 10 guilty escape death if it saves one innocent.