Dr. Ganglmair, how were “patent trolls” able to abuse the legal system in the Western District of Texas?
Bernhard Ganglmair: In the U.S. judiciary litigants are able to choose from a number of courts to bring their cases – subject to some limitations. They will file in a district (the “forum”) where they expect favorable treatment by the district’s judges. An extreme form of this type of “forum shopping” is a strategy called “judge-shopping,” and it is apparently what had happened in the Western District of Texas (WDTX) before random case assignment was introduced in July 2022.
Litigants were able to select one particular judge almost for sure by going to the Waco Division of the WDTX. The only judge in that division, Judge Albright, had a reputation of being friendly to patent owners: when he presided over a case, the probability of patent invalidations was lower, and trial schedules were faster compared to those by the average judge in the same district. He was extremely popular among patent owners, and in 2021, this judge presided over 25 percent of all new patent infringement cases in U.S. federal courts.
What happened after “random case assignment” was introduced?
Bernhard Ganglmair: Our analysis of cases filed before and after the court’s rules were amended shows: Judge Albright was assigned 100 percent of new patent cases in the WDTX’s Waco Division before July 2022. After that date, when random case assignment was introduced, his share fell to a monthly average of 11 percent.
As a consequence, total patent infringement case filings in the WDTX dropped by around 70 percent. Some patent enforcers left the district altogether. Others filed fewer cases. We document a particularly large drop in case filings by companies that do not actually use the patented technologies in the product market – non-practicing entities (NPEs) or so-called “patent trolls.” That’s important as suits of trolls can seriously harm innovations, for instance, when patents are used against startups.
The case filings by owners of high-value patents, which give a high economic benefit to the owner, or low-quality patents also dropped more than others. The reason: With a judge that is less friendly to patent owners, plaintiffs face a higher risk of seeing their patents invalidated, putting an end to a potential income stream from licensing revenues. Such an outcome is particularly costly for owners of high-value patents or owners of low-quality patents that are more likely to be found invalid by the court. We interpret our findings as evidence that judge shopping was particularly prevalent for these types of patent owners. They filed these cases only because of Judge Albright. Once they knew they would face an “average” judge, they decided against the case.
How great is the risk of “judge shopping” in the U.S.?
Bernhard Ganglmair: The danger of “judge shopping” is certainly very real. It was only in March 2024 that the Judicial Conference of the United States, which sets policies for federal courts, recommended districts to change their assignment practices in cases that have a national relevance. The recommendation was to randomly assign cases among all judges in a district to avoid the scenario in which the presiding judge would be known to the plaintiff at the time of the filing – as it was the case in the Waco division of the WDTX. Our research confirms the relevance and effectiveness of this recommendation by using a local policy change in line with the Conference’s recommendations.
Is the judiciary in Europe immune to strategic court selection?
Bernhard Ganglmair: Not at all. “Forum shopping” (and its cousin, judge shopping) is a real concern in Europe, too. For instance, a concern of potential forum shopping has been raised in the context of the Unitary Patent. The Unified Patent Court is the litigation forum at the European level. It has several divisions scattered across member states. So there is a risk that patent enforcers will try to file in the country that seems most friendly to them.
We also see judge (or forum) shopping in the context of standard-essential patent litigation. In recent years, German judges have started to deviate from a prior decision by the European Court of Justice, which is making their courts more friendly to patent owners. This type of “bias” could open the door to strategic court selection if unchecked. The lesson learned from our research is that randomization of case assignment will curb the influence of individual judges and make specific courts less attractive to trolls trying to game the system. This may reduce the overall volume of infringement cases.