Rescuing multidistrict litigation from the altar of expediency

Carter G. Phillips Gene C. Schaerr*' Anil K Abraham**

I. INTRODUCTION

One of the more disturbing developments in our judicial system in the wake of mass tort and other complex litigation is the willingness of courts to depart from clear and unbending procedural requirements-thereby sacrificing key structural protections embodied in those requirements-in the name of judicial economy or efficiency. The notion that judicial process must be compromised because of the perceived exigencies of expansive litigation is well illustrated by the widespread misinterpretation of 28 U.S.C. 1407 as allowing so-called "self-transfers" in cases involving multidistrict litigation (MDL). This practice is lawless, as Judge Alex Kozinski has conclusively demonstrated in a recent dissent.1 But the practice nevertheless has been perpetuated among the lower federal courts in the erroneous belief that it promotes judicial economy. This concession to perceived expediency at the expense of judicial process, however, has not gone unnoticed. The United States Supreme Court has granted certiorari to review the legality of self-transfers in Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach (In re American Continental Corp. /Lincoln Savings & Loan Securities Litigation).2

The legal issue presented in the Lexecon case is the proper interpretation of 28 U.S.C. 1407, a question that has already been examined in a scholarly fashion both by Judge Kozinski in his Ninth Circuit dissent and in numerous articles mentioned in that opinion.3 For the most part, these arguments do not bear repeating here, except to provide a framework for the issues that have not been as thoroughly considered, viz., the policy implications of permitting self-transfers in cases consolidated under the MDL statute. Analysis of those policy concerns reveals plainly that Congress was correct to insist that consolidated cases be returned to their transferor courts for trial. That practice best serves the interests of plaintiffs, defendants, and the federal courts. The policy debate over these competing transfer schemes should inform both the Court's judgment in deciding the legal question (at least marginally) and Congress's judgment about whether to amend the statute (assuming that the Supreme Court decides to reverse the Ninth Circuit's decision, as we believe it will). Accordingly, this Article will provide the first attempt to analyze the policy issues underlying self-transfer in the context of the Lexecon case.

When Congress enacted the MDL statute, it exercised its prerogative to make difficult policy choices and struck a specific balance between concerns of judicial efficiency and the rights of litigants. In so doing, Congress authorized the Judicial Panel on Multidistrict Litigation (JPML) to transfer related cases pending in district courts around the nation to a single district court, permitting a unitary disposition of common pretrial issues. The statute directs that the cases be remanded to their transferor courts once the pretrial issues have been resolved.

Section 1407 thus advances judicial efficiency, but at the cost of partially compromising the right of litigants to individualized adjudication of their claims. In enacting this provision, Congress has determined, correctly in our view, that such a compromise does not impose too great a cost in return for the benefits generated by the compromise-i.e., that the more efficient use of pretrial judicial resources outweighs the harm to litigants. In other words, Congress has considered the policy implications of transfers in the MDL context and, in the very text of 1407, has struck what it deemed to be the appropriate balance between judicial efficiency and the rights of individual litigants.

Unsatisfied with Congress's judgment in this policy matter, the federal courts charged with managing multidistrict litigation have refused to apply the statutory scheme as Congress wrote it. Instead, the courts have arrogated to themselves the congressional prerogative to make difficult, complex policy choices. Rather than remanding 1407 cases to their districts of origin once the pretrial issues have been resolved, as expressly mandated in the text of 1407, MDL courts have chosen to use other federal transfer statutes-28 U.S.C. 1404 and 1406-to retain permanent jurisdiction of cases temporarily transferred to them under

1407. This self-transfer procedure has been carelessly rubberstamped by the federal courts of appeals with almost no analysis.4 Section II of this Article briefly elucidates the statutory scheme and succinctly explains why the courts' adoption of the self-transfer procedure is invalid as a matter of statutory construction.

In adopting this procedure, these courts not only have departed from the plain language of these transfer statutes, as well as the contemporary congressional understanding of them, but also have struck a sharply different balance between the rights of litigants and the concern for judicial efficiency. This balancing of interests is impermissible, not only because it is not the balance chosen and enacted into law by Congress, but also because it simply is not sound judicial policy. To the contrary, the balance improperly adopted by the courts actually subverts sound judicial policy and has inflicted considerable harm upon the right of individual litigants-defendants as well as plaintiffs-to individualized adjudication of their claims.

Section III of this Article demonstrates that sound judicial policy is undermined by the application of the self-transfer mechanism because it introduces multiple anomalies into the adjudication of multidistrict litigation. In transferring cases to themselves under 28 U.S.C. 1404(a), for example, judges presiding over MDL cases routinely disregard the operative factors identified in that statute-the convenience of individual parties and witnesses "in the interest of justice"-and rely instead upon such nonstatutory factors as the interests of parties and witnesses as a group and the perceived needs of the judicial system as a whole. Moreover, the consolidation of trial proceedings that results from self-transfers undermines the integrity of the entire trial process, to the detriment of plaintiffs, defendants, and juries alike. Self-transfers also undermine the protections afforded to defendants (as well as plaintiffs) by the federal venue rules contained in 28 U.S.C. 1391. Perhaps worst of all from the standpoint of judicial administration, the existence of the selftransfer mechanism encourages the most perverse kinds of forum shopping. This includes the well-known "file-and-transfer" strategy by which plaintiffs sometimes file a lawsuit in one jurisdiction to obtain the benefit of favorable substantive law, but then seek to have their cases transferred to another jurisdiction in search of a procedurally more favorable forum.

In addition, section IV of this Article argues that the alleged efficiency benefits often cited in support of self-transfer are illusory. Most of the efficiencies of MDL consolidation arise at the pretrial stage, especially with respect to the management of discovery. Efficiencies at the trial stage are comparatively rare and insubstantial. Where such benefits do exist, they can easily be achieved through the mechanism Congress created, namely

1404(a). Accordingly, self-transfers create few efficiency benefits to weigh in the balance against their enormous costs.

In short, the policy judgment embodied in the plain language of 1407 is entitled to respect from the federal courts not only because it is Congress's judgment, but also because it is a sensible one as a matter of sound judicial administration-far more sensible than the self-transfer procedure created out of whole cloth by the lower federal courts.

II. SELF-TRANSFERS ARE FORBIDDEN BY 28 U.S.C. 1407 AND VIOLATE THE PLAIN LANGUAGE OF 28 U.S.C. 1404

As a matter of statutory interpretation, self-transfer is independently foreclosed both by 28 U.S.C. 1407 and by 28 U.S.C. 1404.

A. Section 1407

The MDL statute specifically directs that cases transferred to an MDL district court be remanded before trial to the districts from which they were transferred:

When civil actions involving one or more common questions of fact are pending in different districts, such actions may be transferred to any district for coordinated or consolidated pretrial proceedings.... Each action so transferred shall be remanded by the panel at or before the conclusion of such pretrial proceedings to the district from which it was transferred unless it shall have been previously terminated....5

Therefore, the plain language of the statute forbids a transferee court from usurping trial jurisdiction over a 1407-transferred case by means of self-transfer.s Even proponents of the self-transfer procedure have conceded that the procedure "evade[s] the specific admonition of section 1407."7 In fact, one judge who used the procedure admitted: "I must recognize candidly that there is nothing in the language of 28 U.S.C. 1407(a) or 1404(a) which directly allows, or even suggests, that the transferee judge has the power to transfer cases to his district, or any district, for purposes of trial."8

The statute's legislative history confirms that the language means what it says. Indeed, Congress explicitly rejected the idea of allowing multidistrict litigation to be consolidated for trial.9 For example, in the House of Representatives report on the bill that eventually became 1407, the Committee on the Judiciary stated that "[t]he proposed statute affects only the pretrial stages in multidistrict litigation" and "would not affect the place of trial in any case."lo The Judiciary Committee thought it wise to limit the bill to pretrial matters because that was the extent of its advisory committee's historical experience with multidistrict litigation.ll The Judiciary Committee also felt that "trial in the originating district is generally preferable from the standpoint of the parties and witnesses, and from the standpoint of the courts it may be impracticable to have all cases in mass litigation tried in one district."12 In fact, the Judiciary Committee flatly stated: "The subsection [1407(a)] requires that transferred cases be remanded to the originating district at the close of coordinated pretrial proceedings. The bill does not, therefore, include the trial of cases in the consolidated proceedings."13 Thus, it is beyond doubt that Congress meant what it said when it mandated that 1407 cases be remanded to the transferor district prior to trial.

Rather than heed the plain language of the statute, as confirmed by the powerful legislative history, the federal courts routinely have chosen to defy it. The core of this defiance is Rule 14(b) of the JPML, which expressly contemplates self-transfer: Each transferred action that has not been terminated in the transferee district court shall be remanded by the Panel to the transferor district for trial, unless ordered transferred by the transferee judge to the transferee or other district under 28 U.S.C. 1404(a) or 28 U.S.C. 1406. In the event that the transferee judge so transfers an action under 28 U.S.C.

1404(a) or 1406, no further action of the Panel shall be necessary to authorize further proceedings including trial.l4

The JPML thus apparently believes that self-transfer is within the authority of transferee district courts.l5 Yet the language in Rule 14(b) recognizing the availability of self-transfer flies in the face of the congressional directive in 1407 to remand transferred cases back to their districts of origin before trial. Because the rule violates the plain language of 1407, the rule also contravenes its enabling statute.l6 The rule is therefore invalid and cannot be used as a justification for the self-transfer procedure.

B. Section 1404

The plain language of 28 U.S.C. 1404(a) provides an independent basis for declaring self-transfers ultra vires. That statute contemplates that cases will be transferred from one district court to another, not from one district court to itself. The statute states, in pertinent part: "For the convenience of parties and witnesses, in the interest of justice, a district court may transfer any civil action to any other district or division where it might have been brought.""7 The natural reading of this language compels the conclusion that the "other district or division" is a district or division different than the one in which the district judge executing the transfer sits-i.e., other than the district or division of the MDL transferee judge.ls Therefore, 1404(a) does not authorize a district court to transfer a case to itself.

Indeed, the self-transfer procedure twists 1404(a) into unrecognizable form. Section 1404(a) is ordinarily used to relinquish jurisdiction in favor of another forum. In the self-transfer context, "[t]ransferee courts improperly use 1404(a) to acquire, rather than relinquish, jurisdiction."ls In fact, "[a] judge granting a 1404(a) motion always divests himself of jurisdiction over the transferred case"-except when the judge is granting a selftransfer.20 In short, "[a] transferee court using 1404(a) is not transferring a case to another district. It is simply proclaiming trial jurisdiction over the case."zl Thus, even if the plain text of

1407 were not enough to demonstrate conclusively the illegitimacy of self-transfer, the text of 1404(a) provides a second, independent basis for declaring the self-transfer procedure invalid.

III. SELF-TRANSFER SUBVERTS SEVERAL IMPORTANT VALUES UNDERLYING THE FEDERAL JUDICIAL SYSTEM

Even disregarding the plain language of 1407 and 1404(a), strong policy reasons exist for invalidating the selftransfer procedure. Among other things, the self-transfer procedure: (a) fundamentally distorts the traditional approach to analyzing transfers under 1404; (b) undermines the integrity of the trial process; undermines the protections afforded by the federal venue rules; and (d) promotes forum shopping. While policy arguments can never trump the plain language of a statute, they can (and in this case do) confirm that there is no tension between the text of the statute and congressional intent.

A. Self-Transfer Fundamentally Distorts Traditional 1404

Analysis

The judicially created self-transfer procedure leads to two fundamental distortions in the analysis traditionally and appropriately used by courts in determining whether to grant a transfer under Section 1404. Those distortions are: (1) subordination of individual interests to the group interest; and (2) subordination of the interests of all litigants to concerns about judicial efficiency.

1. Subordination of individual interests to group interest

By its terms, 1404(a) permits a transfer only "[f]or the convenience of parties and witnesses, in the interest of justice." When this standard is applied in the context for which it was designed-namely, by a district judge in the forum where the action was originally brought-the focus will naturally, and necessarily, be on the convenience of the parties and witnesses in that individual case, not on the convenience of parties and witnesses in some other, unrelated case or cases.

In contrast, by allowing judges to apply 1404(a) to an entire set of consolidated cases, the self-transfer procedure often leads courts to look at the overall convenience of an entire group of litigants, rather than the convenience of each individual litigant, or even the litigants and witnesses in each individual case. As one court stated:

[O]ne must recognize that this is not a typical section 1404(a) situation. The court is not considering the transfer of one case from one district to another but rather the transfer and consolidation of 32 cases filed in twelve districts into one district for trial. Thus, instead of looking to the individual convenience of each party and each witness, the court must look to the overall convenience of all parties and witnesses.22

Even if a judge correctly estimates that overall convenience is "optimized" by a self-transfer, the judge's analysis generally fails to address, in any serious way, whether any individual is adversely affected.23 Indeed, "It]he concept of `convenience of parties and witnesses' takes on an entirely different meaning when a 1404(a) motion, encompassing tens or hundreds of cases, is brought before a 1407 transferee court. In such a setting, the interests of the individual litigants are subordinated to the collective good."24

This problem is both underscored and exacerbated by the fact that the same case may be found both inappropriate and appropriate for a 1404(a) transfer. That is, prior to consolidation under 1407, the district court in which the case was filed may deny a 1404(a) transfer as inappropriate. Then, after a 1407 transfer, the transferee district court may deem the same case appropriate for a 1404(a) transfer. The JPML considers the fact that the original trial court denied a 1404(a) transfer as a factor weighing against a 1407 transfer, but does not consider it to be dispositive.25 The 1407 transferee court, considering issues of judicial economy and collective convenience, could thus decide to grant a 1404(a) self-transfer after the original trial court, looking only to the 1404(a) criteria of convenience to the parties and the interest of justice, had rejected a 1404(a) transfer.26

The extent of this differential treatment between MDL and non-MDL cases should not be underestimated:

"In the local court a case is more likely to be treated as an individual matter with appropriate consideration given to the convenience of the specific parties and local witnesses. On the other hand, if the motion is before a distant transferee court, that court may be more inclined to view the entire litigation as a complex whole rather than weigh the convenience of the parties and witnesses in the particular case."27

2. Subordination of interests of all litigants to judicial efficiency concerns

In addition, the self-transfer procedure often leads transferee courts to import into their analysis another criterion that is nowhere mentioned in the text of 1404(a): the convenience of the trial judge. Ordinarily, this type of judicial economy concern is not taken into consideration in determining whether a transfer under 1404(a) is warranted in a particular case. To the contrary, traditional 1404(a) analysis is driven by the twin concerns of party and witness convenience, not inconvenience of the judge.28

By interbreeding the 1404(a) and 1407 analyses, the MDL transferee courts have conflated the two legal tests. Rather than applying the 1404(a) test independently of 1407 concerns-which still would be improper for MDL transferee courts because of the plain language of the latter provision-these courts appear to have imported the concept of judicial inconvenience into the 1404(a) analysis. Indeed, it may be more accurate to say that judicial convenience and other judicial economy considerations have become the dominant criteria in 1404(a) analysis in the MDL context. This unfortunate development has not gone unnoticed.29

In an ordinary transfer situation, it is apparent why judicial efficiency would not present a significant concern. The district court faced with a typical 1404(a) motion is focused on only one case and only one set of litigants whose rights are at issue in the transfer motion. That single case, at least typically, is no different from any other case on the district court's docket in size or scope. Whether the judge must try this case in addition to all the others on the docket, or whether a transferee court would have to add the case to its docket, will raise only a marginal judicial efficiency concern. As a result, 1404(a), which governs ordinary transfers, is not concerned with judicial convenience or efficiency.

On the other hand, in MDL self-transfer situations, notions of judicial efficiency have become the driving concern.ao Indeed, it was the concern of judicial efficiency that prompted the development of MDL procedures in the first place.3l It should come as no surprise, then, that when district judges improperly import

1404(a) into the MDL context, they also erroneously import a concern for judicial efficiency into 1404(a) analysis.

One court of appeals has succumbed to this tendency to intermingle the two legal standards. The court reviewed a set of ten consolidated antitrust cases in the early 1980s and held that the district court had not abused its discretion in self-transferring the cases.32 In so doing, the court of appeals rejected the wellestablished, individualized approach to 1404(a) transfers.

Although the district court did not expressly quantify the interest in plaintiffs' convenience, it concluded that plaintiffs' convenience was outweighed by the comparative economy of trying one action in the Eastern District rather than several actions in the [plaintiff] states' home districts. Given the complexity of the proceedings, we consider this conclusion to be reasonable. Moreover, we also believe that in an appropriate case the court may properly consider whether judicial efficiency would be served by enabling the judge who has presided over the pretrial phase of a multi-district proceeding, and thereby has become familiar with the parties and the issues, to try the actions himself rather than to return them to one or more district judges who must acquaint themselves with the cases' complexities.33

In fact, the court went so far as to quote with approval language from Pfizer, Inc. v. Lord"4 that self-consciously elevated the convenience of the judge to the level of a 1404(a) statutory criterion:

"While the convenience of the judge is, of course, not normally a factor to be weighed in considering a section 1404(a) motion, in the unusual circumstances of this case we do not feel that it was an abuse of discretion to give some consideration to this problem in balancing the various interests."35

While courts that incorporate judicial convenience into the 1404(a) analysis profess merely to be reading 1407 and 1404(a) together,36 they are in fact creating a new statutory scheme by (1) disregarding 1407's prohibition on self-transfer and (2) treating the two statutes as if they were written contemporaneously and with multidistrict litigation in mind.37 By importing 1407's concern about judicial efficiency into 1404(a) analysis, courts improperly subordinate the interests of the litigants to the perceived convenience of the judiciary. Even those who approve of self-transfer acknowledge that it sacrifices individual rights to what is perceived as judicial economy.38

In sum, even if it were proper to apply 1404(a) tc multidistrict litigation-and it is not, for all the reasons just mentioned-such an application of 1404(a) does not give court; license to alter the legal test established under 1404(a) juris prudence by adding an efficiency component to the test. How ever, that is the inevitable result of allowing self-transfers.

B. Self-Transfer Undermines the Integrity of the Trial Process, to the Detriment of Both Plaintiffs and Defendants

To be sure, in enacting 1407, Congress chose to allow judicial efficiency to outweigh individual convenience, but only for pretrial rulings and hearings.39 One reason Congress gave for its decision not to expand 1407 transfers to include trial proceedings was that "trial in the originating district is generally preferable from the standpoint of the parties and witnesses."4

Congress no doubt came to this conclusion in part because of the Coordinating Committee for Multiple Litigation's representations to it during the hearings on House Bill 8276: "The major innovation proposed is transfer solely for pre-trial purposes. The statute's objectives of eliminating conflict and duplication and of assuring efficient and economical pre-trial proceedings would thus be achieved without losing the benefits of local trials in the appropriate districts."41 Chief Judge William H. Becker, a member of the Coordinating Committee, submitted a prepared statement to the Senate while testifying on Senate Bill 3815,42 stating the point even more directly:

In massive multi-district litigation, transfer for pretrial purposes only is often desirable for many reasons. These reasons include (1) the economy and efficiency of trying local issues, such as damages to individual parties, in the local district wherein the local witnesses and documents are found; (2) the desirability, and often the necessity, of employing local lawyers to process the local issues; (3) the inability of one or a few transferee districts to try fully hundreds or thousands of claims for relief as distinguished from ability to conduct pretrial of hundreds or thousands of claims involving one or more common questions of fact, not local in scope.49

Litigants' experience with self-transfers shows that both Chief Judge Becker and Congress were right to be concerned on that score.

Obviously, a decision by an MDL court to self-transfer for trial purposes will often mean that the parties and the witnesses must travel to a distant city for the trial. That is certainly one problem that Congress had in mind when it concluded that "trial in the originating district is generally preferable."" Beyond such inconvenience, the consolidation that results from a self-transfer often undermines the integrity of the trial process in at least four concrete ways.4

First, such consolidation of trial-related issues frequently causes litigants to lose control of individual cases.' Thus, consolidation often deprives an individual litigant of the ability to craft and present its own case to the jury in the most effective manner.47 Instead, differences in individual cases tend to be ignored.48 This is plainly contrary to Congress's intent. In fact, Congress had been assured by the Coordinating Committee for Multiple Litigation-the body that had helped develop 1407 as a response to the courts' experience with the electrical equipment antitrust cases-that "[p)roposed 1407 would maximize the litigant's traditional privileges of selecting where, when, and how to enforce his substantive rights or assert his defenses."49

Second, consolidated trials often result in prejudice to those defendants (or plaintiffs, for that matter) whose cases are stronger than those of the other parties aligned on the same side of the case. For example, a defendant with a particularly strong defense may suffer from having its case tried with those of other defendants with weaker defenses. The jury may be confused by the multiple defenses and may be prejudiced against the former defendant by virtue of its apparent association with the latter defendants.50

Third, regardless of such differences among the cases, consolidated trials often require a jury to sift through and understand far more evidence than they would if the cases were tried one-byone.51 The problem of "jury overload"-particularly in complex financial or antitrust cases-has been frequently noted by courts and commentators.52 Consolidating numerous cases for trial exacerbates the risk that a jury will not be able to understand and digest the evidence and will instead reach a decision based on secondary, or even improper, factors. This overload will inevitably create unfair prejudice to some parties and give an unfair advantage to others. Once again, the legislative history suggests that Congress understood this risk and sought to avoid it by requiring remand at the close of pretrial proceedings.53

Fourth, even if a jury is able to understand all the evidence, the sheer length of a consolidated MDL trial creates its own problems. Better-educated jurors, whose jobs might allow them to sit on a one- or two-week trial, will often be excused from service if it appears that the trial is likely to drag on for months.54 This will have an obvious, and often dispositive, effect on the make-up of the jury.55 Moreover, the remaining jurors may become angry at having to devote a large amount of time to a trial-anger that rarely falls evenly on all the parties.

In short, the consolidated trials that generally result from self-transfer raise a host of concerns about the integrity of the trial process. These concerns amply confirm Congress's well-considered judgment that "trial in the originating district is generally preferable."6 Congress has made that judgment and rejected the idea of allowing consolidation of MDL cases for trial-whether by self-transfer or otherwise. The courts have no business overriding Congress's command.

C. Self-Transfer Undermines the Protections Afforded Litigants by the Federal Venue Rules

Another detrimental effect of this judicially created selftransfer scheme is that it causes litigants to lose the protections accorded them by the federal venue rules.57 The venue rules restrict the available forums in which a plaintiff can file suit, thus ensuring that the defendant will face the lawsuit in a forum that has some connection either to the substance of the lawsuit or to the defendant himself, and is not completely unrelated to these factors.58 In addition to protecting the defendant from unfamiliar forums, the venue rules also offer the plaintiff some protection by according the plaintiff a measure of discretion in choosing where to file suit.59

When conducting a transfer under either 28 U.S.C. 1404(a) or 28 U.S.C. 1406(a), a district court can transfer the case only to a forum in which venue is proper.60When the JPML transfers a case pursuant to 28 U.S.C. 1407, it is not restricted by the standard venue limitations and is free to choose any district court as the MDL transferee forum.sl As a result, many MDL courts have cases in front of them that could not originally have been brought in the MDL forum, especially in instances where cases from all over the nation are transferred to one MDL court. In other words, proper venue for trial purposes does not lie in many of these MDL transferee courts for many of the cases transferred to these courts for coordinated or consolidated pretrial treatment pursuant to 1407.62

Accordingly, it is-or at least should be-far from a simple matter for MDL courts to execute self-transfers in these cases. Any self-transfer accomplished under 1404(a) or 1406(a) must satisfy the venue restrictions contained in those statutes. However, when MDL transferee courts conduct self-transfers, it is all too easy to disregard venue limitations to varying degrees, even though the courts should be required to comply with these limitations by the very statutes they use to execute the self-transfers.63

Moreover, a party that has already been brought before an MDL court may be reluctant to stand on its rights under the venue rules for fear of alienating the MDL judge, whose rulings on pretrial matters will often not be complete when the decision on a self-transfer is briefed and argued. Litigants should not be forced to choose between alienating the MDL judge and enjoying the protections granted them by the federal venue rules. Yet, the self-transfer procedure often creates just that kind of dilemma.

D. Self-Transfer Promotes Forum Shopping

Finally, the self-transfer mechanism inevitably leads to forum shopping. As the universe of proper venues is stretched by self-transfer, more forums become available for the purpose of litigating the merits of individual cases. As more forums become available, more opportunities to forum shop arise, and with more opportunities to forum shop come greater incentives to take advantage of these opportunities.

The Lexecon case, now pending before the United States Supreme Court, provides an excellent example of this forum shopping dynamic. The litigation began in the Northern District of Illinois when Lexecon Inc., a law-and-economics consulting company, filed a defamation lawsuit against two law firms, Milberg Weiss Bershad Hynes & Lerach (Milberg Weiss) and Cotchett, Illston & Pitre (Cotchett). United States District Judge James B. Zagel handled the initial pretrial matters in the case, and he entered several rulings against the respondent law firms.64The respondents then moved, pursuant to 28 U.S.C. 1407, that the JPML transfer the case to the District of Arizona for consolidation with other cases involving Lincoln Savings and Loan, cases that formed part of the basis for Lexecon's lawsuit.65

The Arizona district judge who had presided over the Lincoln Savings litigation recused himself.gs Despite this turn of events, the respondents asked that the case be transferred to any judge in Arizona-a request that on its face raises the specter of forum shopping." The overarching reason for conducting the 1407 transfer-at least putatively-was to allow consolidation of the defamation suit with other cases related to the Lincoln Savings litigation so that a court familiar with all of the issues could handle the pretrial matters in the defamation suit. However, achievement of that goal was significantly impaired by the recusal of the original judge, an event that critically undermined the rationale for conducting the transfer.

Regardless, the respondents pressed their transfer motion. The only reasonable explanations for their behavior are that (a) they preferred the District of Arizona enough to request a transfer there under any circumstances, (b) they disliked the Northern District of Illinois enough to ask for a transfer to another state, or (c) both. This is a classic example of forum shopping.

An even more egregious possibility is that self-transfer will allow a plaintiff to "double forum shop" or engage in a "file-andtransfer ploy."ss A plaintiff could file a case in a district whose substantive law was favorable, and then request transfer to another district pursuant to 1407 if that would produce some procedural advantage (e.g., procuring a more favorable jury pool). Once the 1407 transfer is completed, the plaintiff could file a motion for self-transfer under 1404(a) and end up with the substantive law of the original forum as the rule of decision in addition to any procedural benefits gained by having the case transferred to the MDL forum.

The clearest example of the "file-and-transfer ploy" occurs in diversity cases. Under Ferens v. John Deere Co.,69 the law of the transferor court applies in diversity cases after a 1404(a) transfer, even if it is the plaintiff who moves for a change of venue.70 "`Cases commenced in other districts are treated as if they are pending in those other districts whether transferred to [the MDL] court for pretrial purposes under the multidistrict litigation statute, 28 U.S.C. 1407, or transferred for trial for the convenience of witnesses, 28 U.S.C. 1404."'71

In sum, a plaintiff seeking to "file-and-transfer" can file suit in an inconvenient forum that has favorable substantive law. The plaintiff can then request a 1407 transfer to an MDL court in an advantageous forum. Finally, the plaintiff can move for a self-transfer under 1404(a) and complete the forum shopping gambit, having secured both an advantageous forum in which to try the case and the substantively favorable law of the transferor district. Absent the option of self-transfer, this ploy would be far less attractive to an enterprising litigant, because it could not obtain the benefit of a more favorable jury.

Even though some limited forum shopping is appropriate and authorized by Congress,72 double forum shopping is not. It should never be tolerated.73 Self-transfer encourages this fileand-transfer tactic by injecting jury- and other trial-related considerations into the litigants' decision. By allowing a plaintiff to "have [its] cake and eat it too," self-transfer encourages the worst kind of forum shopping.74

IV. CONGRESS CORRECTLY CONCLUDED THAT CONSOLIDATING PROCEEDINGS FOR TRIAL WOULD NOT PROMOTE JUDICIAL EFFICIENCY

Proponents of self-transfer will undoubtedly contend that any anomalies created by the procedure are justified on efficiency grounds. However, it is not at all evident that consolidating cases for trial will achieve the putative benefits claimed by the proponents of self-transfer.

Trials are much less amenable to consolidation than are most pretrial proceedings. For example, both motions to dismiss and motions for summary judgment are decided as a matter of law, and "Congress may have felt that federal judges are fungible for purposes of resolving legal questions."75 The reasoning underlying the decisions on such motions should therefore apply to any case with analogous facts. Also, consolidating discovery proceedings makes sense: separate discovery proceedings can often result in unnecessary duplication of effort because each proceeding seeks disclosure of the same materials. The same cannot be said of trial matters, which vary from trial to trial depending on factual differences and varying litigation strategies.

Indeed, in trial proceedings, factual differences between cases can often be the basis for reaching opposite results in otherwise similar cases. Such differences, however, are generally exposed through careful examination and cross-examination of witnesses and other trial procedures. Development and presentation of such case-specific facts, moreover, can generally be accomplished no more efficiently in a single, omnibus trial than in individual trials.77

For similar reasons, there are few, if any, benefits to be gained from consolidating rulings on evidentiary issues. Evidentiary issues are often fact-bound, and a full airing of the relevant facts is often necessary to achieve sensible rulings on such matters. This process is generally not amenable to consolidation, but rather demands individualized attention.

Moreover, it is no simple matter even to administer a consolidated trial for cases brought from all over the country, as one district court recently recognized in the "Sarabond Products" case. In that multidistrict litigation, seventeen cases originally had been filed in five states with different product liability standards.7s At the outset, each representative case-of which there would have to have been at least five, one for each product liability standard-would have required a separate choice-of-law analysis in order to identify the law applicable to each representative case. Alternatively, a single jury would have had to keep separate in its deliberations seventeen sets of relevant facts judged by five different legal standards. Indeed, merely fashioning jury instructions and special interrogatories would have posed a monumental task for the parties and the court. Recognizing that, at best, a consolidated trial would have mired the court in a hopelessly complicated endeavor, the court declined to consolidate the cases for trial.79

Given the practical difficulties of consolidating cases for trial, it is easy to appreciate the wisdom of Congress's judgment, expressed in the plain language of 1407, that MDL cases be remanded to their transferor districts for trial proceedings. But even when there are efficiencies to be gained in consolidating trial proceedings, self-transfer is not the only way to achieve them. The federal transfer scheme already includes a mechanism that permits the lower courts to consolidate litigation for trial purposes without running afoul of 1407's remand directive. The device Congress has put in place for achieving this end is none other than 1404(a) itself. Under the federal transfer scheme enacted by Congress, any 1407 case considered a candidate for a consolidated trial would first be remanded to the transferor court pursuant to the compulsory language of 1407. Assuming venue requirements were satisfied, the transferor court could then entertain a 1404(a) motion to transfer the case back to the MDL forum for a consolidated trial, based on its assessment of the proper 1404(a) criteria: party and witness convenience in the interest of justice. The transferor court, unhindered by the pressure of resolving multiple related cases like the one in the transfer motion, could provide the parties with the individualized attention to party and witness convenience demanded by 1404(a).

If the transfer motion is well founded and there are indeed significant efficiencies to be gained from a consolidated trial, there is every reason to believe that the transferor court will grant the transfer motion in the interest of justice. At the very least, there is no reason to believe that the transferor court would be biased against granting the transfer. In the event the motion is granted, the resulting transfer would achieve precisely the same result as an improper self-transfer, but without defying the plain language of any statute or promoting the kinds of anomalies described above.

The mandatory remand has an added structural protection for litigants that is utterly vitiated by self-transfer. As Judge Kozinski observed in his Lexecon dissent, some MDL judges develop "proprietary feelings" toward the cases transferred to them under 1407.7 The reason may be something as innocent as a desire on the part of MDL judges to finish what they have started. Whatever the reason, the fact remains that MDL judges do develop those feelings. In deciding whether to grant a selftransfer under 1404(a), this dynamic tips the scales in favor of granting a self-transfer and thereby disrupts the careful balancing required by the transfer statute. On the other hand, following the remand directive of 1407 avoids this problem by putting the transfer decision in the hands of the transferor judge. Unhindered by any "proprietary feelings," these judges can undertake a dispassionate analysis under 1404(a) to determine whether a transfer back to the MDL forum is indeed in the interest of justice.

In short, the federal transfer provisions, as written, permit courts to reap all the putative advantages of the self-transfer mechanism without imposing on litigants any of the disadvantages of that procedure

V. CONCLUSION

Self-transfers subvert sound judicial administration by fundamentally altering the criteria traditionally applied to transfer requests under Section 1404(a), by undermining the trial process and the protections of the federal venue rules, and by encouraging forum shopping. Moreover, any genuine efficiency benefits arising from self-transfers can be achieved without self-transfer by resort to the scheme Congress enacted-namely, consideration of a 1404(a) motion by the transferor court after the case is remanded. The Supreme Court would do the judicial system and the litigants who depend on that system a great service by declaring in the Lexecon case that the self-transfer procedure is not only unauthorized, but indeed prohibited by law. Once the Court has reached that conclusion, Congress would do well to reject the inevitable pleas to reinstate that procedure through amendment.

[Author Affiliation]

* Managing Partner, Sidley & Austin, Washington, D.C. B.A., Ohio State University, 1973; MA, Northwestern University, 1975; J.D., Northwestern University, 1977. Law Clerk to Hon. Robert A. Sprecher, United States Court of Appeals for the Seventh Circuit, 1977-78; Chief Justice Warren E. Burger, United States Supreme Court, 1978-79. Assistant to the Solicitor General, United States Department of Justice, 1981-84.

[Author Affiliation]

** Partner, Sidley & Austin, Washington, D.C. BA., Brigham Young University, 1981; M.A., Yale University, 1985; J.D., Yale University, 1985; M. Phil., Yale University, 1986. Law Clerk to Hon. Kenneth W. Starr, United States Court of Appeals for the District of Columbia Circuit, 1985-86; Hon. Warren E. Burger and Antonin Scalia, United States Supreme Court, 1986-87.

[Author Affiliation]

*** Associate, Sidley & Austin, Washington, D.C. A.B., Dartmouth College, 1991; J.D., Yale University, 1995. Law Clerk to Hon. Jerry E. Smith, United States Court of Appeals for the Fifth Circuit, 1995-96.