My amicus brief in the Multi-Color Corporation bankruptcy seems to have touched a nerve, with some interesting responses from both the debtor and Judge Kaplan. I’ll note that this is not the first time something has seemed amiss with New Jersey venue, and it’s not even the only pending case with strange venue.
I want to respond to the debtor’s claims about case distribution, to Judge Kaplan’s comments. In a separate post I’m going to discuss venue in the Eddie Bauer’s bankruptcy (which is with Judge Meisel).
Tell me you don’t understand statistics without telling me you don’t understand statistics.
The debtor has responded, claiming that my own data refute my thesis because there are chapter 11 case assignments distributed across the Jersey judges. To this end, the debtor cites a bunch of large chapter 11 cases that went to other Jersey judges. Let’s be clear: I do not claim that all the cases go to Judge Kaplan. Only that a disproportionate number of them go to him. Citing the fact that some cases went to other judges is not a serious response.
The debtor also claims that I offer “no mechanism by which any party could engineer a particular judge assignment.” That’s true. I don’t know how the cases end up with Judge Kaplan, and I don’t want to speculate. (Maybe the debtor’s counsel has some idea. That’s what I would want to ask debtor’s counsel if I were a Third Circuit Judge hearing oral argument…)
But I will note this. The New Jersey Complex Case Procedures require that the court be given three days notice prior to the filing of the bankruptcy petition with any matters requiring upfront attention flagged in advance. Perhaps it’s just one way communication, but that would surprise me. More importantly, it implies a pre-filing assignment of the case to a judge: there is no reason to tell the Clerk of the Court what matters are going to be coming in hot, but that’s something that a judge would want to know.
What I can’t tell from the Complex Case Procedures is what sort of communication might occur with debtor’s counsel. Might the debtor’s counsel learn of the case assignment prior to the filing? If so, then if the debtor doesn’t like the assignment, it just files in Houston or perhaps Wilmington. Or maybe the debtor delays on filing and sees if a different Jersey judge is assigned. Again, I don’t know if this is what happens. I am not arguing about a particular causal mechanism. All I can say is that Judge Kaplan is getting large case assignments at a rate exceeding two distributions from the expected mean, which is a fancy way of saying that it is “unusual.”
Judge Kaplan’s Statistical Disputes: Even if Correct, They Are Immaterial
Judge Kaplan himself has responded in Bloomberg with some quibbles with my New Jersey case counting. Specifically, he notes that depending on the time period, there weren’t 9 bankruptcy judges in the district, but 7 or 8. He also observes that it’s reasonable for a chapter 22 to go back to the same judge, so Rite-Aid 2.0 shouldn’t count.
I’m not sure about the chapter 22 point—if one judge got it wrong on the plan feasibility question, maybe a different judge ought to have a crack at it—but his point about the number of judges is certainly fair. So let’s see the impact of adjusting for both Judge Kaplan’s criticisms:
If we reduce the Kaplan case count and total case count both by 1 to account for Rite Aid 2.0, he was still assigned 48% of the cases, rather than 50%. And if we go from a random 1 in 9 to a random 1 in 7, that moves the random distribution from 11% to 14%.
So now we’ve gone from 50% against a random distribution of 11% to 48% against a random distribution of 14%. Either way, the outcomes are WAY off (>2 standard deviations from the mean) from an expected random distribution.
Bottom line is that there is something unusual going on with New Jersey case assignments. I’m not in a position to know why and maybe there’s a perfectly innocent explanation, but having this sort of case distribution explains why some parties are so concerned about the abuse of venue.