Earlier this month, the Fifth Circuit granted a mandamus petition in In re Clarke, No. 24-50079, 2024 U.S. App. LEXIS 5099 (5th Cir. Mar. 1, 2024), effectively reversing a Texas court’s decision to transfer an Administrative Procedure Act (APA) case to a D.C. court under 28 U.S.C. § 1404(a). This is another in a recent line of mandamus decisions clarifying the law around transfers for convenience. The Fifth Circuit’s renewed interest in this area springs partly from its recognition that issuing few such decisions over the last fifteen years has generated conflicting outcomes among Texas district courts and the Federal Circuit, which both apply Fifth Circuit law to the § 1404(a) motions that have become ubiquitous in Texas patent litigation. See In re TikTok, Inc., 85 F.4th 352, 367 (5th Cir. 2023).

The key takeaways from the Clarke opinion are:

• The local-interest transfer factor turns on the location of the events giving rise to suit and succeeding on this factor may depend on how those events are characterized. In the context of patent infringement, the Federal Circuit has already characterized the relevant events as including the design, development, and testing of accused products. See, e.g., In re Apple Inc., 979 F.3d 1332, 1345 (Fed. Cir. 2020). It is unclear whether In re Clarke permits a re-evaluation of the Federal Circuit’s characterization or consideration of other interests and how nationally “diffuse” those interests may be.
• The weight accorded to the docket-congestion transfer factor continues to erode. Although a district court’s evaluation of its own docket efficiency is entitled to deference and much more weight than docket statistics, the docket-congestion factor is speculative due partly to a transferor court’s inability to reliably gauge the congestion of the transferee court. This factor cannot be dispositive.

The Clarke plaintiffs sued the Commodities Future Trading Commission (CFTC), an independent federal agency regulating the U.S. derivatives market, for violating the APA by ordering that a marketplace for trading on predicted outcomes of political events be closed and all pending contracts liquidated. In re Clarke, 2024 U.S. App. LEXIS 5099, at *3. The plaintiffs, including users of the trading platform living and working in Austin, Texas, filed suit in the Austin Division of the Western District of Texas. Id. The CFTC moved to transfer the action to the U.S. District Court for the District of Columbia, where corporate plaintiffs actually operating the marketplace resided. Id.

Courts in the Fifth Circuit consider the following factors in evaluating whether transfer under § 1404(a) is warranted.

The private-interest factors are:

• the relative ease of access to sources of proof;
• the availability of compulsory process to secure the attendance of witnesses;
• the cost of attendance for willing witnesses; and
• all other practical problems that make trial of a case easy, expeditious, and inexpensive.

The public-interest factors are:

• the administrative difficulties flowing from court congestion;
• the local interest in having localized interests decided at home;
• the familiarity of the forum with the law that will govern the case; and
• the avoidance of unnecessary problems of conflict of laws or in the application of foreign law.

In the Clarke district court, Judge David Ezra granted CFTC’s motion to transfer for three principal reasons. First, congestion in the Austin Division favored transfer. Id. at *7–8. Second, the District of Columbia has a greater local interest in the case because the CFTC and two corporate plaintiffs that operate the marketplace are based there. Id. at *9. And third, all the private interest factors (like the convenience of willing witnesses and ease of access to proof) were neutral because the case is likely to be resolved on the administrative record without resort to witness testimony. Id. at *15–16. On mandamus, the Fifth Circuit deemed the transfer opinion “patent error,” finding issues with the trial court’s treatment of each factor.

Docket Congestion. Although it held that Judge Ezra did not clearly abuse his discretion in finding that the Austin Division’s congestion favored transfer, id. at *7–8, the appellate panel also stated that “it would be a stretch to say that court congestion ‘favors D.D.C.’ and not just transfer ‘somewhere else,’” id. at *19–20. The Fifth Circuit clarified that comparisons between the transferor and transferee courts’ time-to-trial statistics and weighted-case counts “carry little weight” and must give way to a transferor court’s evaluation of its own “docket efficiency.” Id. at *8 (quoting TikTok, 85 F.4th at 363 n.12). And here, Judge Ezra had properly evaluated Austin’s docket efficiency in view of the court’s inability to timely rule on the transfer motion or an earlier preliminary injunction motion. Id. at *7–8. Yet the Clarke panel also opined that a transferor court’s estimation of a transferee court’s congestion is an “uninformed guess.” Id. at *20–21.

Local Interests. The Fifth Circuit determined that Judge Ezra erred in focusing on the parties instead of the events giving rise to suit. Id. at *12 (“That point bears repeating: We focus on the events—not the parties.”). The panel went further, clarifying that “the local-interest inquiry is concerned with the interest of non-party citizens in adjudicating the case.” Id. In doing so, the panel distanced itself from another of its recent mandamus decisions that identified the location of witnesses and the plaintiff’s residence as important considerations under this factor. Id. (recontextualizing Def. Distributed v. Bruck, 30 F.4th 414, 435 (5th Cir. 2022)). In addition, the panel reaffirmed that a “completely diffuse interest”—an interest applying to every judicial district—“cannot affect the local-interest determination.” Id. at *10. This is consistent with the Federal Circuit’s longstanding position that “the sale of an accused product offered nationwide does not give rise to a substantial interest in any single venue.” In re Hoffmann-La Roche Inc., 587 F.3d 1333, 1338 (Fed. Cir. 2009).

Applying this guidance, the Clarke panel found this factor neutral for two reasons. First, the events giving rise to suit occurred in Austin, where individuals using the trading platform reside and were “harmed.” 2024 U.S. App. LEXIS 5099, at *14–15. But they also occurred in D.C., where the marketplace was developed and where its operators reside. Id. Second, if the plaintiff prevailed, the district court must set aside the CFTC’s action, which “affects persons in all judicial districts equally.” Id. at *15.

Notably, the Fifth Circuit did not address or give any weight to the location of the CFTC’s places of business, from which the CFTC presumably evaluated the marketplace before ordering it be liquidated—which could be characterized as events giving rise to suit. Nor did the panel address or consider the location of non-parties similarly situated to the Austin-based plaintiffs, like marketplace users in the D.C. area.

Private Interest Factors. The Clarke panel found that Judge Ezra erred in finding all the private interest factors neutral based on mere speculation that this case would be resolved on the administrative record. Id. at *16–17. That had already been disproven by the CFTC’s pressing for dismissal on threshold issues that could require evidentiary hearings and testimony. Id. Instead of remanding for the district court to reconsider the private interest factors in the first instance, the panel determined that the third private interest factor—the cost of attendance for willing witnesses—“strongly favors the Western District of Texas” because plaintiffs had identified an Austin resident as a primary source of testimony on a threshold issue, whereas the CFTC had not identified any relevant witnesses. Id. at *18–19.

In short, Clarke provides some insight for Texas patent litigants into relevant considerations for the docket-congestion and local-interest factors but is unlikely to mark a drastic shift in how Texas district courts or the Federal Circuit approach the § 1404(a) inquiry.

[View source.]