This post is part of a series exploring the ideas in an important new article by Professor Aaron-Andrew Bruhl (Univ. Houston), entitled Measuring Circuit Splits: A Cautionary Note. One important concept Bruhl explains is that of parallel review. Journalists and academic writers often describe particular federal circuit courts as being the "most-reversed" or "most-affirmed" by the Supreme Court, but the numbers used to arrive at such rankings are often dubious, in part for the reason that they tend to count only direct review decisions by the Supreme Court, ignoring parallel review. The Supreme Court may reverse the Ninth Circuit when it receives an appeal from that court (the basis of most direct review tallies), but this ignores the fact that the Court in that decision is simultaneously reversing and affirming some other circuits that have split on that issue; and the fact that the Court may reject (or endorse) the Ninth Circuit's holdings when reviewing appeals from other circuits. As Bruhl puts it:
Consider a hypothetical Supreme Court that resolves three cases, one each from the Sixth, Seventh, and Ninth Circuits. The Supreme Court affirms in the case from the Seventh Circuit and reverses in the other two, for a reversal rate of 67%. In truth, however, the Supreme Court has reviewed not just three decisions but probably several times that number, because each of those three cases on the docket presented legal questions that had divided the lower courts for years. Suppose the legal question in each case can be simplified to a binary choice between two options such as X or not-X . . . A few important differences between direct and parallel review now become apparent. For one, the full measure of the Supreme Court’s reversal rate can differ significantly from the rate on direct review.
Bruhl next explains that parallel review, though ignored by many researchers, "is probably the more appropriate measure for most questions one would want to study." Yet counting the cases for parallel review is easier said than done - Bruhl demonstrates that the main databases of Supreme Court decisions do not always flag reversals or affirmances of circuit courts accurately or completely. Here are problems to keep in mind:
First, the Supreme Court does not always list all the circuits that have taken sides on the question before the Court. Merely conducting a word search for "Fifth Circuit" in Supreme Court decisions will reveal only the decisions that use those words, but will not cull results where the Court rejected (or affirmed) a position taken by, say, the Fifth Circuit, without acknowledgement, when the case itself was an appeal from another circuit. Worse, sometimes the Supreme Court does not reveal that it is resolving a circuit split at all, even when an issue has generated many conflicting decisions stretching back for years.
Second, the breakdown of the split is not always clear, especially when the question at issue is nonbinary. To give an example from some of my own previous articles (here, here, and here), individual circuits use different nomenclature for the same rule regarding the entrapment defense (objective rule, ALI rule, MPC rule, due process rule are all interchangeable), or for the same permutation of entrapment (e.g., "vicarious" or "indirect" entrapment). Lower courts vary in using the phrases "Nondelegation Doctrine" and "Delegation Doctrine" to describe the same legal rule, or in using the terms "severance" or "bifurcation" to describe the same procedural device. Moreover, two circuits may have directly opposite, mutually-exclusive rules for a legal issue, but another circuit may use both approaches combined into a two-step rule (as when state courts use both the objective and subjective tests for the entrapment defense each time the defense comes up).
More posts will follow exploring Bruhl's arguments. Stay tuned.