concurring in part and dissenting in part:
I am pleased to concur in Sections I, IIA and B of Judge Smith’s opinion, the latter of which hold that petitioners have shown a clear and indisputable right to relief from the district court’s patently erroneous interpretation of their Lexecon waivers and that mandamus relief would be “appropriate under the circumstances” of this case. I dissent, however, from Section IIC, which concludes that a pending appeal of one set of bellwether cases provides an adequate remedy at law rendering mandamus relief unavailable.
A few additional facts about this litigation need to be stressed. Not only are there presently over 9,300 cases pending in the Northern District of Texas for pretrial proceedings, but they'comprise cases that were transferred from “home” districts' of filing by the MDL panel and others that were “direct filed” in the Northern District without undergoing the formal transfer pr'ocess. All of the New York cases set for trial in the instant “bellwether” case'were “direct filed.” But for the possibility of a “global waiver” of personal' jurisdiction, the Northern District had no claim to personal jurisdiction over the cases: none of the plaintiffs’ surgeries occurred in Texas; the plaintiffs aren’t Texas residents; and neither general nor specific jurisdiction exists over the petitioners for purposes of these disputes. For that reason, the district court relied solely on the “global waiver” and extended it from a waiver of venue for pretrial purposes ■ only and for two bellwether trials to waiver of personal jurisdiction in all of the thousands of cases. Petitioners are being forced to trial over their objections to personal jurisdiction.
' By comparison,' a scholarly opinion from the Southern District of Texas in an MDL case resulted in dismissal of a nonresident defendant against which there was a “direct filed” case by a nonresident plaintiff. In re Heartland Payment Sys., Inc. Customer Data Sec. Breach Litig., No. CIV.A. H-10-171, 2011 WL 1232352 (S.D. Tex. Mar. 31, 2011). The court first noted that the defendant’s agreement to transfer for purposes of pretrial proceedings was not inconsistent with and did not waive its personal jurisdiction challenge. 2011 WL 1232352 at *5-6. Finding no waiver, the court then decided that it lacked personal jurisdiction over the non-consenting defendant based on Heartland’s lack of minimum or relevant contacts with the Southern District of Texas. 2011 WL 1232352 at *6-10. In re Heartland is contrary to this case, in that the district court here ignored the limitations on the parties’ agreed order for direct filing. That order expressly states that, “[u]pon completion of all pretrial proceedings applicable to a case filed directly in the [MDL court], this Court may transfer the case, pursuant to 28 U.S.C. Section 1404, to a court of appropriate jurisdiction for trial.” (emphasis added). Moreover, the inclusion of any action in the MDL proceeding under the order’s auspices “shall not constitute a determination by this Court that jurisdiction or venue is proper in this District.” The petitioners’'failure to waive personal jurisdiction, as the panel majority hold, means that the district court is asserting “reach out and touch someone” power wholly at odds with its Constitutional and statutory limitations. See Point Landing, Inc. v. Omni Capital Int'l, Ltd., 795 F.2d 415, 423 (5th Cir. 1986) (en banc) (“[federal courts] must ground their personal jurisdiction on a federal statute or rule. As courts of limited jurisdiction, the federal courts pos*357sess no warrant to create jurisdictional law of their own”) (internal citation and quotations omitted).
This is where I begin to differ with the majority’s holding that petitioners have an adequate remedy in a pending appeal of the third bellwether trial arising from California plaintiffs, some of whom direct-filed in the Northern District. I agree that mandamus relief should only be granted in unique situations and must be used “sparingly and with utmost, care.” In re Chevron U.S.A., Inc., 109 F.3d 1016, 1021 (5th Cir. 1997) (Jones, J. concurring). But this is not simply a case in which reasonable minds can differ over the court’s finding of waiver. Nor is the irreparable harm alleged by •the petitioners simply the cost of trying and possibly appealing the judgment from another set of bellwether' plaintiffs, although that will be more millions of dollars. See In re Lloyd’s Register North America, Inc., 780 F.3d 283, 295 (5th Cir. 2015) (when the issues in mandamus' have significant ramifications, the pendency of another appeal in this court ’strengthens the case for mandamus relief).
The Supreme Court has stated • that “[t]hese hurdles [to the issuance of mandamus relief], however demanding, are not insuperable.” Cheney v. U.S. Dist. Court for D.C., 542 U.S. 367, 381, 124 S.Ct. 2576, 2587, 159 L.Ed.2d 459 (2004) (granting writ). There are numerous ongoing ramifications of the court’s erroneous decision that harm not only these petitioners but, importantly, the plaintiffs in these 9,000 + cases. . •
First, when a court plainly acts in excess of its jurisdiction, mandamus may issue to prevent the usurpation of power. Abelesz v. OTP Bank, 692 F.3d 638, 653 (7th Cir. 2012) (“If there were a colorable - argument supporting the district court’s - exercise of jurisdiction, we would view this case differently. The overwhelming clarity of this issue, however, calls for use of the extraordinary writ of mandamus to confine the district court to the proper exercise of its jurisdiction”); see also In re Impact Absorbent Techs., Inc., 106 F.3d 400 (6th Cir. 1996) (“[although uncorrectable damage may not result if petitioners are forced to wait for a remedy on direct appeal, the clearly erroneous nature of the district court’s order calls for a more immediate remedy.”); see also Holub Indust., Inc. v. Wyche, 290 F.2d 852, 856 (4th Cir. 1961). Ordinary appeal can be an inadequate remedy if the “challenged assumption or denial of jurisdiction” is “so plainly wrong as to indicaté failure to comprehend or refusal to be guided by unambiguous provisions of a statute or settled common law doctrine.” Am. Airlines v. Forman, 204 F.2d 230, 232 (3rd Cir. 1953).
Second, the lack of personal jurisdiction is not the only problem here, because petitioners’ issues necessarily embrace improper venue. Where there is no personal jurisdiction over the petitioners, the trial is in an improper venue. Moreover, petitioners note they never consented to waive venue for any proceedings beyond the pre.trial stage of these cases. The majority’s focus on the alleged sufficient remedy for lack of jurisdiction in the pending appeal overlooks that where venue is improper, this court has held that appeal is an inadequate remedy rendering the grant of mandamus relief mandatory. In re Volkswagen of Am., Inc., 545 F.3d 304, 319 (5th Cir. 2008) (en banc) (“an appeal will provide no remedy for a patently erroneous failure to transfer venue”); In re Ford Motor Co., 591 F.3d 406, 416 (5th Cir. 2009) (same).
Third, the appellate remedy is inadequate when the error the parties are subjected- to will be repeated. “Mandamus is particularly appropriate here because of the potential for the trial court to repeat” its error. Journal Pub. Co. v. Mechem, 801 *358F.2d 1233, 1237 (10th Cir. 1986) (writ granted). The district court has misinterpreted the petitioners’ case-specific waivers to hold that it may potentially try all 9,000+ cases in the Northern District of Texas, no matter their lack of any connection with this state, much less its district. As confirmed by plaintiffs’ counsel in oral argument, that is what the court means by a “global waiver.” Even though a curiously timed late-breaking order of the trial court states that no further bellwether trials will occur, this does not signal a retreat from its holding that a “global waiver” occurred. Nor does the order retreat from the court’s repeated refusals to slow down its processes while the appeal of jurisdiction in the third bellwether set of cases is taking place. Repetition, or surely threatened repetition, of its error, manifestly deserves correction at the earliest moment.
Fourth, the appellate remedy is inadequate where the court’s error is likely to affect cases beyond those before us. “Writs of mandamus are supervisory in nature and are particularly appropriate when the issues also have an importance beyond the immediate case.” In re Volkswagen, 545 F.3d at 319. Such effects are foreordained where the court is managing thousands of cases and the usual object of MDL management, especially with bellwether trials, is to incentivize rational settlements. Even though the third and fourth grouped bellwether cases have been tried without proper venue or jurisdiction, and the results from an unauthorized forum may be skewed in numerous ways, their impact on the parties’ settlement approaches cannot be denied. The pressure to settle will be ongoing until final resolution of an appeal in this court or even in the Supreme Court. That a court’s patent errors can compound into unjust settlements is well recognized and has led courts to find mandamus a necessary remedy even where a possibility of ultimate appellate relief exists. Abelesz, 692 F.3d at 652-53. The undeniable pressure on defendants to settle is a reality in these alleged mass tort cases.
Fifth, the appellate process in both the third and fourth grouped bellwether cases will extend for several years. Briefing will not even begin until the end of September 2017 in the third such case. One may reasonably predict that this court’s processes, even without en banc review, would not conclude until the end of 2018 at best. The certiorari process would extend for many months thereafter. Concomitantly, the parties may be pursuing appeal in the fourth grouped bellwether cases, the subject of this petition, leading to various potential interactions between the two appeals in this court. Thus, it is unlikely that the “sufficient” appellate process will have concluded before 2019 or 2020. Suppose the petitioners, who have allegedly denied any intention to settle, pursue this lengthy process? In the meantime, what happens to the 9,000+ plaintiffs in this litigation that began in 2011?
The majority’s decision to deny relief overlooks the impact on the plaintiffs. If the district court lacked jurisdiction over these direct-filing plaintiffs’ cases, as our panel majority concludes, they will receive a take-nothing judgment nearly a decade after their suits were filed and will have to start all over—if they have the stomach for it. For the remaining thousands, the goal of the bellwether process will have been perverted by unreliable judgments, delayed by the appeals, and undermined when those judgments are reversed. Allowing the court’s conduct of trials outside its jurisdiction to spawn such unpredictability and unfairness will harm petitioners or plaintiffs and most likely both. Such an outcome belies the goals of efficiency, economy, fairness, and predictability for *359which the MDL system supposedly exists. See 28 U.S.C. § 1407(a) (“[Transfers shall be made by the [MDL panel] upon its determination that transfers for such [pretrial] proceedings ... will promote the just and efficient conduct of such actions”).
A final word about Judge Costa’s separate opinion. First, we do not owe “double deference” to the district court even if (and Judge Smith’s opinion takes no firm position on this) the question of waiver of personal jurisdiction should be reviewed for abuse of discretion on appeal. Judge Smith’s opinion emphasizes that the “mandatory language [of Section 1407] creates a powerful presumption in favor of remand ...hence the “clear and unequivocal” standard for a Lexecon waiver. The record, viewed in its proper context, does support the wholesale abrogation of petitioners’ remand rights and assumption of jurisdiction that the district court did not have. Second, mandamus relief can be justified when courts must address new or evolving legal issues. See, e.g., United States v. Horn, 29 F.3d 754, 770 (1st Cir. 1994); see also 16 Charles Alan Wright & Arthur R. Miller, Fed. Prac. & Proc. Juris. Section 3934.1 (3d Ed. 2004 & Supp. 2017). Finally, for my own part, I am not concerned about the “centralizing power” inherent in MDL proceedings when they are confined, as the statute says, to pretrial matters or to bellwether cases to which both parties have expressly and unequivocally consented. I am even less concerned about the “centralized power” of federal appellate judges when the downside of withholding such power, as I have shown, is the district court’s abuse of its limited jurisdiction, waste of trial resources, huge expenditure of legal resources on utterly unnecessary jurisdictional wrangling, and ultimately, the forfeiture of public confidence in a system that can achieve neither efficiency nor economy in handling mass tort cases.
The issuance of a writ in this unique case, a “bellwether” for thousands more, would do what the writ was intended to do—confine the district court to a lawful exercise of its prescribed jurisdiction. See Cheney, 542 U.S. at 380, 124 S.Ct. at 2576. I respectfully dissent from the denial of relief ordering the district judge not to try the instant grouped cases over which he lacks personal jurisdiction.'