A key goal of any civil-justice system is to allocate cases to courts in sensible and efficient ways. Most systems exhibit allocative rules that rigidly and substantially limit party choice among forums. Not so in the United States. This paper details the unique landscape of forum shopping in American courts along three dimensions: vertical shopping between federal and state court, horizontal shopping among states, and individual shopping for particular judges. It describes the legal, structural, and cultural foundations that enable and even encourage forum shopping in the United States, especially as contrasted with other countries. It then explains and assesses its persistence in American litigation culture today.
One of the most basic arrangements of form and structure in the public civil-justice system is the mechanism for allocating a case to a court. One might think that proper case allocation would be both easy and uncontroversial. A few centuries ago, when social and commercial relationships were local, that was, perhaps, largely true. Say that, in New York City, painter Bob wants to sue his neighbor, shopkeeper John, for failing to pay for a paint job that Bob performed for John’s shop. Probably everyone—Bob, John, New York, you, me—would all agree that their dispute properly belongs in a local New York court competent to hear civil contract claims. Even without the system’s assistance or direction, Bob likely would file his lawsuit in the right court.
Things are much more complicated today. Nations, societies, commerce, and even individuals are no longer local. Nor are they simple. Some nations have large, segmented territories with diverse ecosystems of courts and judges. Social communities are a complicated web of diffuse, integrated, and segmented networks. Commercial relationships involve intangible communications and services, artificial entities, and international transactions. Individuals travel freely, over great distances, often interacting with strangers. Today, the easy lawsuit of Bob v. John might still occur, but only among a variety of far more complex lawsuits. Say a German company contracts with an Austrian company to construct an airport in Iraq, with the construction insured by London insurers underwritten by a U.S. company based on the state of New Hampshire. The airport is later found to have corrosion damage, and a dispute arises about whether the insurers are liable for coverage. Where should that lawsuit be heard? The answer isn’t easy.
Further, the modern litigation system’s sophistication and, frankly, costliness means that even Bob v. John may not be so simple today. Perhaps painter Bob knows that shopkeeper John’s business is strapped for cash, and so Bob strategically decides to sue John in California to impose cost and burden on John to defend in a faraway court. That doesn’t make much sense from any perspective, except one: Bob’s desire to put pressure on John to win.
To resolve difficult questions of forum, and to inhibit the kind of forum shopping that Bob might consider, civil-justice systems have designed legal mechanisms to direct the allocation of cases to their proper forums. In general, the public system tries to identify the appropriate forum, in light of party and system interests, along different dimensions, such as geography, case type, party type, and relief sought. A proper allocation of a case to an appropriate forum furthers values shared between the system and the parties: efficiency, cost control, and results that are accurate, consistent, and predictable. A misallocation can lead to inconvenience, cost, delay, and inaccuracy. Although all public court systems have allocation mechanisms grounded in these values, the mechanisms they use, and the relative rigidity with which they impose them, vary considerably.
The continental system of case allocation, for example, relies on a system of top-down codified rules based on the presumption that there is one natural forum for the dispute. This approach has ancient roots, stemming from the Roman and Canon law’s goal of imposing order by arranging things “in their suitable and appropriate places.” This principle became part of legal thought in Western Europe, supplemented in the Enlightenment Age by a commitment to scientific rationalism and geometric precision. At the same time, the Enlightenment brought ideas of fundamental fairness to the legal system. Civil turmoil at the time of the French Revolution afforded the opportunity to impose those ideas in a new civil-justice system of judicial jurisdiction and competence that reflected coherent structure, methodical precision, and detailed codification.
A key feature of the modernization of the civil-law court system was to reform the ancien régime of disparate and overlapping first-order tribunals that had proliferated to excess and that spawned inefficiencies, confusion, and systemic tensions. Case-allocation conflicts among these tribunals were resolved by resort to the King for transfer to the appropriate tribunal, but that system proved ineffective in resolving conflicts, and it fomented social and political resentment. Reformation of the system focused on the principle of the “natural judge,” the one right court for the case, as set by fairly rigid allocative rules of jurisdiction and competence. As part of the commitment to that principle, civil-law systems promoted both territorial limits and subject-matter specialization. Ultimately, modernization of civil-law systems eliminated much party choice in selecting the forum.
The common-law tradition in the UK developed its case-allocation rules in a different fashion, but it has ultimately arrived at a modern system that constrains party choice with rigidity similar to the continental system. Early on, English courts developed a fairly clear distinction between law and equity, and, as a small country with centuries of relatively stable unification under the monarch, case-allocation conflicts presented less challenges than on the continent. Notice to the defendant in a civil lawsuit had been an animating principle since the Magna Carta, and so territorial competence of a court was founded less on rational connections between the forum and the case or parties and more on the appearance of the defendant through civil arrest or court summons. A particular court had jurisdiction over the parties not because of any abstract natural law but because of the physical assertion of power.
Modernization made significant steps through the County Courts Act of 1846, the Common Law Procedure Act of 1852, and the Judicature Act of 1873. These reforms merged law and equity, established a structure of generalist courts with minimal subject-matter specifications and with expanded court discretion to issue summons beyond territorial borders. Since then, however, the UK has gravitated toward significant subject-matter specialization. Today, the High Court of England and Wales is divided into a Family Court, a Business and Property Court (divided into separate specialist courts or “lists” of Admiralty; Business; Commercial; Competition; Financial; Insolvency and Companies; Intellectual Property; Property, Trusts, and Probate; Revenue; and Technology and Construction), and the Queen’s Bench (which hears most other kinds of civil lawsuits). In addition, the High Court’s first-instance jurisdiction is limited to personal-injury claims of at least 50,000 pounds and other claims of at least 100,000 pounds; claims for under those amounts are relegated to the County Court, which has different tracks for different amounts in controversy, with tailored sets of procedural rules. For cases filed in the wrong court, the UK enables intra-system transfer of the case to the right court, though transfer is discretionary and takes into account party concerns. These modernized features of the UK civil-justice system mean that although parties have some choice in selecting the court, the system tends to funnel cases to the court with appropriate subject-matter competence. As for forum selection along geographic dimensions, the relatively small size of England and Wales, with the gravitational center of London, means that geographic considerations are of only minor concerns.
The United States, as in so many other areas of the law, remains an outlier. American courts exhibit very little subject-matter specialization compared to other countries. Court divisions based on the nature of the relief sought are minimal. Territorial restrictions, imported initially from the common law’s focus on service of process, have developed in uniquely American ways to permit, in most cases, multiple geographic forums with personal jurisdiction over the parties. The upshot is that plaintiffs generally have multiple lawful forums from which to choose, and, even after the plaintiff selects a forum, the defendant often has an opportunity to transfer the case to a different forum. And the U.S. civil system generally defers to the parties’ preferences in each instance. The law often even encourages parties to exercise forum choice. On top of all these forum options structured by the law is the reality, known to the parties, that the heterogeneity of forum choices matters to the outcome of the dispute.
With such options for party influence over forum selection comes forum shopping, which reflects the reality that a plaintiff who must initially choose between available forums is likely to choose one favorable to the plaintiff’s interests. Forum shopping is a term loaded with negative connotations. That negativity is deserved in some circumstances, such as in Bob’s choice to sue John in California for reasons of inconvenience to John. But there is an undercurrent in American legal culture of prideful acceptance of forum shopping that taps into a deference to party wishes, the unique structure of U.S. constitutional federalism, and the peculiarities of American adversarialism.
In this paper, I detail the unique landscape of forum shopping in American courts. I begin by describing the legal, structural, and cultural foundations that enable, encourage, or discourage forum shopping in the United States. Then, I offer some observations and reflections on American forum shopping by explaining and assessing its persistence in American litigation culture today.
Descriptively, forum shopping in the United States can be divided into three main dimensions: vertical (federal or state court), horizontal (which state), and individual (which judge). I discuss each dimension below.
Vertical forum shopping is mostly a binary choice: federal court or state court. In U.S. legal parlance, the vertical choice of forum is referred to as subject-matter jurisdiction, even though the subject-matter of the case has only partial impact on the vertical division. Crucial to vertical forum shopping is the fact that the United States is a federal nation that treats the states as quasi-independent sovereigns, with their own court systems, laws, and case-allocation regimes.
States are entitled to structure their state courts as they wish, but most have adopted a system of generalist first-instance courts that are able to hear all cases not expressly allocated elsewhere. Some specialized state courts do exist, often involving family-law matters or tax matters. But the vast bulk of civil cases are heard by ordinary state courts of general jurisdiction. Importantly, the source of law is generally irrelevant to the availability of forum. Thus, a California state court can hear claims arising under New York law, federal law, or even British law, unless that law gives some other forum exclusive jurisdiction to hear the claim.
U.S. federal courts can hear all civil matters arising under federal law, and, except in a few areas like antitrust, intellectual property, and securities, federal law grants the federal and state courts concurrent jurisdiction to hear federal civil claims. In other words, if a plaintiff wants to sue her employer for discrimination on the basis of race under Title VII of the federal Civil Rights Act, she may choose to bring her case in federal court or state court. In addition, the Constitution authorizes, and Congress has granted, concurrent jurisdiction to the federal courts over state claims that involve more than $75,000 and that are either between citizens of different states or between U.S. citizens and citizens of other nations. Thus, for Bob’s case against John, if Bob happens to reside in New Jersey but John resides in New York, and if Bob’s claim is for more than $75,000, then Bob can choose to sue John in either federal or state court. Finally, in preference for joining related claims together in one lawsuit, federal law authorizes concurrent jurisdiction over cases with a federal-law claim and a related state-law claim, even if the state-law claim is between same-state citizens and for less than $75,000. Thus, even if Bob and John were residents of the same state and Bob’s paint-job claim was for less than $75,000, Bob could still sue John in either federal or state court if Bob were able to join that state-law claim with some related federal-law claim he had against John.
This regime may seem, at first blush, strange. Why should federal courts be authorized to hear state-law claims? After all, state courts are available to hear those claims and, presumably, are more efficient and accurate in resolving them. It makes more sense for Congress to give federal courts authorization to hear federal-law claims, but why give them only concurrent jurisdiction, such that state courts could hear those claims as well? And why, of all things, let the plaintiff get to choose, with unfettered discretion, among these forums?
The reasons have to do with the federalist structure and history of the United States. At the time of the Declaration of Independence, state courts had existed for many decades; meanwhile, neither federal courts nor federal laws existed. When the federal courts were created, in 1789, most Americans assumed that the state courts would continue to be the primary tribunals for resolving all civil disputes, including any disputes arising under federal law. In fact, Congress did not even grant the lower federal courts general jurisdiction to hear federal-law claims until 1875, after the American Civil War. (I should note that the Judiciary Act of 1789 did give the U.S. Supreme Court appellate jurisdiction over all state supreme courts to review questions of federal law that they had decided.) So giving state first-instance courts concurrent jurisdiction over federal-law claims was seen as deference to the state courts’ long, established history, with error correction and protection of federal-law uniformity maintained by the appellate jurisdiction of the U.S. Supreme Court. And rather than potentially insult the states by ousting their jurisdiction over federal-law claims—effectively implying that they could not adequately adjudicate those kinds of cases—Congress granted concurrent jurisdiction and offloaded the forum choice to the parties.
Federal diversity jurisdiction over state-law claims between parties from different states, by contrast, was granted to the lower federal courts in 1789. The rationale was to promote interstate harmony by offering an ostensibly neutral federal forum to alleviate any appearance of bias that a state court might have in favor of one of its own residents and against an out-of-state resident. The amount-in-controversy limit helps ensure that only cases of significant magnitude, as a proxy for the risk of bias, will be eligible for concurrent federal diversity jurisdiction. In cases eligible for diversity jurisdiction, giving the plaintiff the choice makes some sense because the plaintiff, who generally will be the out-of-state party—is in the best position to determine whether the risk of state bias warrants the federal forum in a particular case.
In all cases of concurrent jurisdiction under either diversity jurisdiction or federal-question jurisdiction, if the plaintiff chooses federal court, then the plaintiff’s choice is final. If, however, the plaintiff chooses state court, then the defendant gets a choice. Congress has statutorily granted defendants a right to transfer (in U.S. parlance, to “remove”) a concurrent-jurisdiction case from state court to federal court. For federal-question jurisdiction, the idea is that the system will allow the misallocation of a federal claim to a state court only if both parties agree. For diversity jurisdiction, the idea is that if the defendant is the out-of-state party, then the defendant needs a mechanism to choose to invoke diversity jurisdiction, which is specifically designed to protect the out-of-state party from the appearance of the risk of bias of state court.
So the upshot is that parties have a lot of choice in forum shopping between federal and state courts—the law even sanctions that choice. Forum shopping based on the justifications for concurrent jurisdiction—efficiency, accuracy, and the alleviation of court bias—are benign, even warranted. As it happens, though, state and federal courts have a number of other differences that induce forum shopping quite apart from, and for less noble reasons than, the justifications assumed by Congress. And the evidence suggests that parties exercise their forum choices with those other differences in mind.
Until 1938, for example, federal courts hearing nonfederal claims under diversity jurisdiction were authorized to develop their own bodies of law of tort, contract, commercial paper, and other areas of general common law. Starting in the late 1800s, that regime became a boon to corporations. As entities that repeatedly found themselves litigating against individuals from different states, corporations embarked on a concerted effort to use the federal courts hearing cases on diversity jurisdiction to develop bodies of substantive law favorable to business interests. By the 1930s, the lower federal courts were seen as far more pro-corporate than the state courts.
The famous case of Erie Railroad Co. v. Tompkins put an end to that regime in 1938 by abrogating a century’s worth of federal common law and disavowing the authority of federal courts to develop it. And, in the subsequent case of Klaxon Co. v. Stentor Electric Manufacturing Co., the Court held that a federal court must apply the horizontal choice-of-law principles that a state court would apply. Today, under those cases, the substantive law generally depends upon the claim, not upon the forum.
Though Erie dramatically curbed vertical forum shopping based on the choice of substantive law, the year of 1938 marked another important development: the adoption of the Federal Rules of Civil Procedure. For the first time, the federal courts had a uniform set of procedures applicable only to them. And because the state courts maintained their own rules of procedure, the adoption of the Federal Rules generated a new basis for forum shopping: choice of procedure. And this choice-generating factor applies in both diversity cases and federal-question cases.
Procedure shopping through forum shopping is widespread and can generate startling results. In one case, called Shady Grove, a New York healthcare provider and an insured individual sued Allstate Insurance Co. for filing to pay or deny their insurance claim within the 30-day period prescribed under New York law. They sought the statutory interest that had accrued on the late payment. The total interest amount owed in their case was quite small—only around $500—so, to make litigation economically viable, the plaintiffs decided to sue as a class of all persons and entities to whom Allstate owed interest payments under New York law. The problem was that, in New York state court, New York law did not allow claims for statutory interest to be maintained as a class action. But Allstate was not a New York company, and so the plaintiffs filed their class action in federal court under diversity jurisdiction. And, in federal court, Federal Rule 23 allowed class actions for statutory interest. By choosing federal court, the plaintiffs were able to turn a $500 individual claim into a class action seeking more than $5 million, and the Supreme Court upheld both the forum choice and the application of Rule 23.
As Shady Grove shows, both plaintiffs and defendants can vertically shop for either federal procedure or state procedure in a variety of cases. And the law encourages them to do so.
In addition to vertical forum shopping between federal and state court, parties can horizontally forum shop for a federal or state court in a particular state. The United States is a physically large and diverse territory, making geographic forum selection important regarding considerations of party convenience and system efficiency. In addition, each state can organize, fund, and set procedures for its courts as it pleases, resulting in differences among state courts.
For cases filed in state court, the choice of state is limited only by the doctrines of personal jurisdiction and forum non conveniens. Personal jurisdiction is widely understood today as being set primarily by the Due Process Clause of the Fourteenth Amendment to the U.S. Constitution. States can give their courts less territorial authority than what the Fourteenth Amendment permits, but few have constrained their courts very much. The Fourteenth Amendment allows a court to assert personal jurisdiction over a party who (1) resides in the state, (2) is served while present in the state, (3) consents to personal jurisdiction in the state, or (4) has sufficient minimum contacts with the state such that the exercise of personal jurisdiction is consistent with traditional notions of fair play and substantial justice. The fourth category includes the state in which the party is “at home,” which for an individual is generally the state of domicile and for a corporation is generally the state of incorporation and the state where the headquarters office is located. The fourth category also includes any state that has a sufficient link between the party and the claim, such as if the injury occurred in the forum state and the party purposefully conducted activities in the state.
Importantly, these possibilities are in the disjunctive. Thus, the following state’s courts will have personal jurisdiction over a defendant corporation: (1) any state where the defendant has consented to personal jurisdiction, (2) the state of the defendant’s incorporation, (3) the state of the defendant’s headquarters, and (4) any other state with a sufficient nexus with the defendant and the claim, such as where the defendant’s state-related activities caused harm. That is much more limited than the old “doing business” personal jurisdiction once thought to apply to most national corporations, but it still permits plenty of options for plaintiffs to forum shop among states. In particular, the use of contracts and potentially business-registration statutes to extract consent to personal jurisdiction can proliferate the number of states with personal jurisdiction.
In federal court, the forum-shopping opportunities are even wider. Generally, federal courts exercise the same scope of personal jurisdiction as state courts, and the venue statute adds little in the way of constraint. However, Congress has granted federal district courts nationwide personal jurisdiction in specific instances, such as in antitrust and securities cases. In addition, the federal courts have nationwide personal jurisdiction when the suit arises under federal law, the defendant isn’t subject to any state court’s personal jurisdiction, and the defendant has sufficient minimum contacts with the United States as a whole. For example, a foreign company that does substantial business with the United States as a whole but does not target any one state in particular could, in a case involving a federal claim for injury caused by the defendant in the United States, be subject to every federal court’s personal jurisdiction, even those located in states other than where the injury occurred. Importantly, for actions against foreign residents, venue is proper in any judicial district.
The larger the number of available states with personal jurisdiction, the more the opportunities of plaintiffs to forum shop among states. Why would plaintiffs do so? The answer is multifaceted. Different state courts have different procedures, so forum shopping by state court can be clandestine procedure shopping. Regardless of court, plaintiffs might shop for juries; California juries, say, might be perceived to be more claim-friendly than Iowa juries. Plaintiffs might forum shop as a clandestine way of shopping for the applicable substantive law, which could vary depending upon where the case is filed. More nefariously, plaintiffs might shop for a forum that, geographically, is highly inconvenient for the defendant, like our friend Bob. Less nefariously, plaintiffs might shop for a forum that is most convenient for them. Plaintiffs might also select their home state for the hope of a generalized home-state advantage. Finally, plaintiffs might shop to join cases together to gain cost savings and economies of scale.
A key illustration of this kind of horizontal forum shopping is Keeton v. Hustler Magazine, Inc., in which a resident of New York wanted to bring a libel suit against Hustler Magazine, an Ohio corporation. The plaintiff was outside the applicable statute of limitations in New York, so she sued in New Hampshire, whose “unusually long” limitations period would allow the claim, and she argued that New Hampshire had personal jurisdiction over the magazine because the magazine sold and distributed thousands of copies in New Hampshire. The Supreme Court upheld New Hampshire’s personal jurisdiction over the magazine based on the magazine’s minimum contacts with the state. Further, the Court allowed the plaintiff to recover damages in that New Hampshire suit based on injuries the plaintiff suffered from distribution of the libelous publication in all fifty states.
Forum shopping by state also can occur before and after the plaintiff files the case. Before the filing, parties can agree, in prelitigation contracts, to restrict or permit a forum for any future dispute, and the Supreme Court has upheld these kinds of agreements. After filing, a state court can dismiss a case on forum non conveniens grounds so that it can be refiled in a more appropriate court. Although state doctrines of forum non conveniens vary considerably, they tend to take into consideration similar factors of party convenience and public interests. In federal court, the venue-transfer statute allows either party or the court sua sponte to transfer the case to another federal court—including one in a different state—for the convenience of the parties and witnesses and in the interests of justice, and the federal version of forum non conveniens exists for when a more appropriate forum is abroad.
These front- and back-side mechanisms generally help to control the plaintiff’s ability to forum shop at filing, but, occasionally, they create their own forum-shopping opportunities for plaintiffs. In Ferens v. John Deere Co., a Pennsylvania plaintiff was injured in Pennsylvania by a combine machine he had purchased from John Deere, a Delaware corporation. The injury gave the plaintiff both tort and contract claims under Pennsylvania law. The plaintiff filed, in federal court in Pennsylvania, a diversity case on his contract claims, but, by that time, the Pennsylvania statute of limitations on his tort claims had expired. Accordingly, the plaintiff filed a tort action in a Mississippi federal court because the applicable Mississippi statute of limitations was unusually long. Then, the plaintiff sought to use the venue-transfer statute to transfer the Mississippi case to Pennsylvania, where it could be consolidated with his Pennsylvania case. In essence, the plaintiff hoped to capture the Mississippi statute of limitations for his tort claim, while having Pennsylvania law otherwise apply to both claims, and have them both litigated together in his home state. The Supreme Court allowed this rank forum shopping by the plaintiff.
U.S. judges are, predominantly, generalists. Federal courts and state courts have relatively few defined specializations in comparison to judicial systems abroad. U.S. judicial education is undifferentiated from attorney education; in fact, judges often are selected from the ranks of attorneys who have been in practice for several years. Accordingly, one might think that the ability to choose a particular judge is of little importance.
Yet meaningful differentiation among judges does exist. Judges, after all, are human, with their own predispositions. And American law is designed to give judges quite a bit of discretion in interpreting and applying the law, and even in making law in the common-law tradition. All judges—from first-instance state-court judges to the U.S. Supreme Court justices—exercise the power of judicial review under a system that some have described as “unique” across the globe. And because judges are usually elected or term appointed (as in most state courts) or are politically appointed for life (as in most federal courts), judges tend to hold, and are understood to hold, partisan political views. Thus, some judges have become known for favoring certain kinds of claims or parties.
Usually, individual judicial views are diffuse, and most courts’ rules for assigning judges use randomized methods to insulate judicial assignment from party influence. Accordingly, forum shopping for a judge with a particular view or personality or predilection is often too speculative to be worthwhile. But there are exceptions. Many rules provide for nonrandom assignment of cases to a judge who already is handling highly related cases because the efficiencies of having highly related cases heard by a single judge outweigh the appearance of unfairness that might otherwise attend to nonrandom assignment.
More troubling is when plaintiff-side forum shopping capitalizes on certain court locales known to be staffed by a single judge or a small group of judges with largely homogenous views. Two examples come from Texas. One involves patent cases filed in the Eastern District of Texas. Until 2017, courts followed precedent from the Federal Circuit—a specialized federal appellate court for hearing patent cases—that, in patent cases, both personal jurisdiction and venue under a special patent-venue statute were appropriate wherever the patent infringer purposefully and regularly distributed infringing products. Those cases essentially meant that patent-infringement plaintiffs could choose to file in any federal court where the infringing products were distributed.
Substantive patent law is uniform federal statutory law—more uniform than most federal laws because the Federal Circuit provides circuit uniformity across the nation. So the substantive law wasn’t a factor in forum choice. And, usually, federal procedures are uniform as well. However, the Eastern District of Texas, a mostly rural area not known for patent innovation, began to interpret the Federal Rules, and to exercise judicial discretion under them, in pro-plaintiff ways. Commentators have argued that Eastern District judges did so in a concerted effort to benefit their locality through reputational enhancement and increased commercial activity. Regardless of motivations, the Eastern District of Texas became a magnet for patent cases, and the judges there enhanced that condition by routinely denying or delaying defendants’ motions to transfer patent cases to a different district. In 2015, for example, forty-four percent of all patent-infringement filings were in the Eastern District of Texas. The Eastern District had a case-assignment system that tended to route certain cases in predictable ways, and most patent cases were routed to a single district judge, Rodney Gilstrap, who sat in the small rural town of Marshall, Texas (population 25,000), and who was known for his extremely pro-plaintiff procedures and plaintiff-friendly juries. From January 2014 to June 2016, Judge Gilstrap heard around a quarter of all patent cases filed in the United States.
The Supreme Court finally put an end to much of this forum shopping in TC Heartland LLC v. Kraft Foods Group Brands LLC, when it held that the patent-venue statute was not as permissive as lower courts and parties had previously assumed. The decision dramatically reduced the patent docket of the Eastern District of Texas, but the solution just proves the lure and power of forum shopping.
The other example involves other federal districts in Texas. Although federal districts constitute a unified court, they often are further divided into geographical “divisions.” And sometimes those divisions have only a single judge assigned to them. The Northern District of Texas, for example, has seven divisions, five of which have only a single judge hearing all or most cases. Judicial assignments are confined by division, so a case filed in one of those divisions of the Northern District of Texas will almost certainly be assigned to a single judge known in advance. The state of Texas, represented by its Republican attorney general, recently has exploited this nonrandom judicial-assignment practice when filing suits challenging the Biden Administration’s progressive policies. As one scholar, Professor Steve Vladeck, has written:
Texas has exclusively filed suits challenging federal policies . . . in divisions where all or nearly all cases are assigned to judges appointed by Republican presidents. In fact, almost without exception, Texas has filed its cases in districts presided over entirely or almost entirely by judges appointed by President Trump.
Professor Vladeck continued:
This case is a perfect example of Texas shopping for judges . . . . Texas filed this particular suit in the Victoria Division of the Southern District of Texas. Victoria is neither Texas’s state capital . . . nor does it have any particular connection to the . . . issue in this case. Texas’s Amended Complaint in the case does not contain a single allegation tying Victoria to its lawsuit, claiming only that venue lies “because the State of Texas is a resident of this judicial district, and a substantial part of the events or omissions giving rise to Plaintiffs’ claims occurred in this District.”
So why did Texas file this lawsuit in Victoria? Although the Southern District of Texas has nineteen authorized judgeships—the fifth-most of any district in the country—the Victoria Division has exactly one district judge to whom new civil cases are assigned: Judge Drew B. Tipton, appointed in 2020 by President Trump. This lawsuit is one of five that Texas has filed against the federal government in the Victoria Division which was then assigned to Judge Tipton. By filing this case in Victoria, Texas was able to select not just the location for its lawsuit, but the specific federal judge who would decide this case: a judge Texas likely believed would [rule in its favor]—and who in fact did so, even as other courts have rejected similar challenges.
So, judge shopping, though rare, does happen, and apparently less rarely in Texas.
The descriptions above set out both motive and opportunity for forum shopping in the United States. It is no wonder, then, that forum shopping by American lawyers has been described as “a national legal pastime,” with entire textbooks being devoted to the practice. Like much of American procedure, U.S. forum shopping is exceptionalist among the world’s nations; in the main, American civil litigation presents a greater range of potential forum shopping by both plaintiffs and defendants. Further, U.S. law often encourages forum shopping. I’ll offer some observations as to why that is, about the unique ways in which the American legal system attempts to control forum shopping, and on some lingering concerns with U.S. forum shopping.
The answer to why the U.S. system offers such broad opportunities for forum shopping has both structural and cultural components. Structurally, the constitutional, federal structure of the U.S. government imposes some case-allocation principles that frame forum selection. The Due Process Clauses, for example, set outer bounds of courts’ personal jurisdiction that cannot be altered by Congress or the states. For the most part, the states permit their courts to exercise personal jurisdiction to constitutional limits, and the federal courts generally exercise personal jurisdiction to state constitutional limits, so horizontal forum-shopping opportunities have largely been set by the Constitution, as interpreted by the Supreme Court, which provide a relatively broad scope of horizontal forum choice as compared to other countries. Exacerbating horizontal forum shopping is the U.S. system’s commitment to generalist, rather than specialized, judges, which eliminates opportunities, prevalent in other countries, to funnel cases to specific courts. Finally, the very existence of the U.S. Constitution, as the founding compact among the states, exacerbates the difficulty of achieving coordinated agreement among the fifty states. In contrast to Europe, which has a history of collaborative convention-making, it would be quite extraordinary for the U.S. states to adopt something akin to the Brussels Convention or its Recast.
Vertical forum shopping has structural components based on the Constitution and judicial federalism. Article III grants concurrent jurisdiction over diversity and federal-question cases to federal courts. One of the reasons for concurrent, rather than exclusive, federal jurisdiction was “as a hedge against a feared ‘consolidation’ of federal judicial power.” The Framers also posited that concurrent jurisdiction would create a market of federal and state courts competing, in a race to the top, for the people’s affection. That commitment to shared power continued in the first Congress, which granted lower federal courts concurrent diversity jurisdiction but no general federal-question jurisdiction. Thus, although Article III does not mandate concurrent jurisdiction, and although Congress has provided for exclusive federal-court jurisdiction in discrete areas of federal law, tradition establishes a “deeply rooted presumption in favor of concurrent state court jurisdiction” as part of a longstanding commitment to federalism. This tradition has long justified concurrent jurisdiction between federal and state courts, along with its concomitant vertical forum shopping.
The second dimension explaining U.S. forum shopping is grounded in American legal culture. American laissez faire, egalitarianism, and rugged individualism have contributed to a legal culture that is highly adversarial in both individual cases and in general industry groups representing plaintiff, defense, or corporate litigation interests. The entire U.S. legal system is structured around persistent commitments to litigant autonomy and zealous advocacy. Part of that commitment is grounded in the assumption that the parties know what is best for their litigation in comparison to the government. And part of that commitment is grounded in the framing that, in the battle that is civil litigation, forum shopping is a strategic weapon no less legitimate than any other. Forum shopping—at least when not “abuse[d]”—could even be considered a “duty” under the ethics rules. The result is that most forum shopping is generally accepted as legitimate advocacy, while some commentators even defend forum shopping as normatively desirable in some forms. And despite sporadic expressions of contempt, courts largely condone, even encourage, forum shopping as an accepted practice.
These features of the American legal system help explain why forum shopping is widespread. They also help explain why America’s approach to controlling forum shopping is also exceptionalist. In contrast to rigid, top-down controls favored in Europe, like exclusive jurisdiction, multijurisdictional conventions, and specialist courts, U.S. controls on forum shopping—when they exist—tend to be of two varieties. The first variety posits that the answer to forum shopping is through market-based disincentives. The second variety posits that the answer to forum shopping is more forum shopping.
Market-based disincentives are prevalent forms of U.S. control. After all, the very visualization of jurisdictional overlap as a market is an acutely American vision that dovetails with other American values of individualism and free markets. Some market controls on forum shopping—such as the inefficiencies, costs, delays, and risks of inaccuracy felt by parties that select inconvenient or inappropriate forums—operate as natural controls. But others are imposed controls. Erie, for example, was a case designed to create a market-based disincentive to vertical forum shopping by unifying the substantive law to be applied under any choice of forum. The Supreme Court’s recent restrictive turn in personal jurisdiction will reduce—but usually not eliminate—the plaintiff’s market options. And the risk of back-end enforcement problems helps motivate the plaintiff to shop horizontally for a legitimate forum. The U.S. system relies on these market controls to cabin some forum shopping.
Where excessive forum shopping persists, the U.S. system’s answer is to promote more forum shopping. Are plaintiffs excessively or unfairly forum shopping for certain pro-plaintiff state courts? Give defendants the ability to remove the case to federal court or to force refiling elsewhere on grounds of forum non conveniens. Are plaintiffs forum shopping for certain pro-plaintiff federal courts? Give defendants the ability to transfer the case to a different federal court. Do these controls still give plaintiffs too much of a forum-selection advantage? Allow defendants to forum shop by initiating a declaratory-judgment action in their preferred forum. Worried about the extra cost of all this forum shopping? Allow the parties to jointly forum shop through a forum-selection clause or an arbitration clause in a prelitigation agreement.
It is not clear how successful these market- and party-driven controls are at curbing forum shopping, at least the kind of forum shopping that is normatively undesirable. Treating courts like a marketplace for justice ideally could create a race to the top for efficient and fair civil justice, but some evidence suggests that courts and systems are motivated to skew toward partiality to market themselves to one side or the other in an effort to attract cases. Arguably, the Eastern District of Texas advertised excessively pro-plaintiff procedures and reputations to attract initial filings by plaintiffs, while Delaware markets its courts as business-friendly and business-knowledgeable in an effort to induce business to incorporate there. Forum-selection clauses are also complicated. In circumstances of equal bargaining power, forum preselection can be beneficial for its certainty and for enabling a non-zero-sum agreement to a forum that is likely to reduce inefficiencies and costs for both parties. In circumstances of disparate bargaining power, however, the normative desirability of forum-selection clauses is far less clear.
Perhaps the greatest threat of U.S. forum shopping lies in the middle of the perfect storm of universal judicial review, public litigation, and nationwide injunctions. The U.S. theory of litigation accepts a significant portion of the government’s regulatory responsibility, resulting in many public-rights claims—discrimination, antitrust, and the like—being litigated by private litigants and remedied by the courts. Federal courts, including first-instance courts, have assumed the authority to issue nationwide, also called “universal,” injunctions that purport to control conduct wherever it would occur, even if outside of the court’s district borders. These conditions incentivize extreme forum shopping in high-profile, consequential public litigation by encouraging plaintiffs to race to the most partisan judge, wherever located, most likely to award a nationwide injunction. Such a practice oversteps boundaries, stifles the useful percolation of legal issues among multiple lower courts, undermines judicial legitimacy, and erodes the rule of law. The likely solution, consistent with the market-based solutions favored by U.S. tradition, is not to inhibit forum shopping but rather to reduce the availability of nationwide injunctions, an idea that some justices of the Supreme Court seem intrigued by. It remains to be seen how the United States will tackle this problem.
In the end, U.S. forum shopping is broad in scope, widespread in practice, and quintessentially American in design. How forum shopping will evolve, especially as litigation continues to go global, will be watched with interest by scholars from the U.S. and from elsewhere around the world.
