Lemley & Miller: Judges Who Sit by Designation Less Likely To Be Reversed by Fed. Cir.

I'm thrilled to be here! Lisa’s list of non-faculty presenters at IPSC prompted me to check out a nifty new paper by Mark Lemley and Shawn Miller, demonstrating that the Federal Circuit reverses claim constructions far less frequently when the district judge below previously sat by designation in a claim construction appeal. Lemley and Miller examined each claim construction appeal decided in the last 12.5 years, coding for (1) whether a claim construction was reversed and (2) whether the district judge previously sat by designation on a Federal Circuit panel reviewing claim construction. The Federal Circuit found error about half as frequently in cases where the district judge previously participated in a claim construction appeal compared to cases where the district judge had never sat by designation (15.7% v. 33.1%, see Table 1 on page 11). This result was robust to controls for judicial experience, whether the appeal was decided after Phillips v. AWH Corp., and the district of origin, among others. The authors argue “the most likely explanation” for their result “is not a learning effect but a consequence of the personal relationships district judges develop with appellate judges while sitting at the court” (p. 28) – the effect appears to persist even for the (few) judges who sat by designation but never heard a claim construction appeal.

I wouldn't be surprised if this study catches the eye of many a reversal-weary district judge. If Lemley and Miller’s result stands up to scrutiny and this relationship is, indeed, causal, a judge with a significant patent docket could save a lot of net time with a weeklong trip to D.C.

The paper also raises some fun questions for future research. The authors themselves plan to tackle whether the effect holds in the regional circuits, noting that regional appellate judges may already know many of the district judges within their circuit. I’m also curious whether the effect holds for district judges who dissented on their circuit panels. And I wonder if there’s any way to distinguish between “informal deference” through formation of trust versus a reluctance to harm a personal relationship or hurt someone’s feelings by reversing an acquaintance. Whatever is to come, the paper presents a cool finding and is worth a read.