Last month, a House Judiciary subcommittee held a hearing on whether Congress should create new lower-court judgeships. In its most recent recommendation, the Judicial Conference of the United States sought five new appellate judgeships, all in the Ninth Circuit. By contrast, in her testimony Duke law professor Marin K. Levy suggested that the increase in the appellate caseload of approximately 20% since 1990 (when Congress last created new federal appellate judgeships) warrants adding a much higher number of new judgeships.
Let me explain why I am very skeptical of Levy’s caseload benchmark.
For starters, I’ll note that total filings in the federal appellate courts fell 9.9% from 2011 to 2020, according to data from the Administrative Office of the U.S. Courts. (See this year-by-year chart.) In order to assess whether this trend will continue, it would be useful to understand why it has occurred. If this trend does continue, we might soon need fewer judges, not more.
Second, Levy’s comparison of filings from 1990 to now assumes that a filing is a filing is a filing—that is, that on average a volume of X filings in 1990 would impose the same burden as a volume of X filings now. That’s a sensible, and perhaps inevitable, starting-point assumption. But is it accurate? Or is it reasonable to explore whether the composition of a court’s caseload has changed over thirty years, perhaps giving it a lighter average case (but perhaps, of course, giving it a heavier one)?
Third, judicial productivity ought to have increased dramatically over the past thirty years, as a result of technological advances. When I was a Ninth Circuit law clerk thirty-five years ago, my fellow law clerks and I all used typewriters, and I doubt that any judge on the court prior to Judge Alex Kozinski used a word processor. It’s much, much easier to compose judicial opinions and to work from home or on travel than it used to be. Is it unreasonable to expect the average judge to be at least twice as productive as a result?
Fourth, the metric of filings per judgeship is really a metric of filing per authorized active judgeship and thus ignores the important contribution that senior judges play. There are various reasons—increased lifespans, judges taking senior status at a younger average age—why there might be more senior judges freely adopting a larger caseload. One data point I’ve looked at indicates that senior judges now account for the equivalent of 55 active judgeships, versus only about 35 some two decades ago.* Given that there are only 179 authorized active judgeships, it seems a large mistake to exclude senior judges from the calculus. (That said, I recognize that it’s also fair to consider whether too much is being asked of senior judges.)
* Senior judges accounted for 22.8% of the case participations in 2020, and active judges accounted for 74.1%. (Visiting judges accounted for the remaining 3.1%.) The ratio between the two numbers is roughly the same as the ratio between 55 and 179. I did a similar calculation for 1997 data.