Article

“Contractually Valid” Forum Selection Clauses

Introduction

On December 3, 2013, the United States Supreme Court issued its opinion in Atlantic Marine Construction Company v. United States District Court.1 The case resolved a longstanding circuit split relating to the procedure for enforcing a forum selection clause in federal court.2 The answer, a unanimous Court held, depends on the identity of the court named in the clause.3 When the chosen court is a federal court in a different federal district, the matter is governed by the 28 U.S.C. § 1404(a).4 When the chosen court is a state court or a foreign court, the matter is governed by the doctrine of forum non conveniens.5 The decision was generally well-received by scholars and has been widely cited by the lower federal courts.6 It did not, however, address the most complex and vexing issue relating to forum selection clauses: the issue of when a forum selection clause is valid in the first place.

The Court’s decision not to engage with this issue was purposeful. A footnote buried deep in the opinion stated that the Court’s “analysis presupposes a contractually valid forum-selection clause.”7 In making this presupposition, the Court avoided having to grapple with the thorny question of what, exactly, makes a clause “contractually valid.”8 Over the past eight years, federal judges and their law clerks have devoted countless hours to answering this question.9 They have pored over state contract law.10 They have reviewed the Supreme Court’s seminal decisions in The Bremen and Carnival Cruise.11 They have grappled with mind-bogglingly complicated choice-of-law issues. They have considered whether special venue provisions in federal statutes trump forum selection clauses. And they have researched whether the federal courts are obliged to follow state statutes invalidating forum selection clauses. None of these questions have easy answers. Taken together, they form a dense knot of doctrine.12

This Article seeks to untangle the knot. Its first goal is to supply an answer to the question of when a forum selection clause is “contractually valid” as that term is used in Atlantic Marine. This inquiry is complex. Indeed, it is so complex that a federal court must undertake three separate inquiries to answer it. First, the court must determine whether a forum selection clause is valid.13 If a clause appears in a contract that is not supported by consideration, for example, the clause is not valid. The inquiry into validity is complicated, however, by the fact that the courts have devised special rules of contract law that apply exclusively to forum selection clauses.14 Some of these rules relate to fraud.15 Others relate to the rights and obligations of third parties.16 Still others relate to contract termination.17 As a first step in assessing whether a clause is valid, the court must apply these rules to determine whether the parties have entered into a binding agreement to resolve their disputes in a particular forum.

Second, the court must interpret the clause to determine its meaning.18 If a party is seeking to dismiss or transfer a case on the basis of a forum selection clause, for example, it must persuade the court that the clause selects the court of the chosen jurisdiction to the exclusion of all others.19 That party must also convince the court that the clause is broad enough to cover the claims asserted.20 If a clause is non-exclusive, or if it is too narrow to cover the claims, then it is not “contractually valid” as that term is used in Atlantic Marine.

Third, the court must determine whether the clause is enforceable. There are, broadly speaking, two reasons why a forum selection clause might not be enforceable. First, a clause may be unenforceable because it is contrary to public policy.21 Second, a clause may be unenforceable because it is unreasonable.22 Although public policy and reasonableness are famously elastic concepts, the courts have over the years identified a number of criteria that they consistently apply to determine whether a clause is unenforceable on one of these grounds. If a clause is unreasonable or contrary to public policy, then it is not “contractually valid” as that term is used in Atlantic Marine.

When a forum selection clause passes each of the tests set forth above—when it is valid, exclusive, applicable, and enforceable—then the court must apply the framework set forth in Atlantic Marine to decide whether to grant the defendant’s motion to transfer or dismiss.23 In almost all cases, this motion will be granted.24 When a clause is contractually valid, the Supreme Court has stated that the case should be transferred or dismissed absent some “extraordinary circumstances unrelated to the convenience of the parties.”25 In practice, therefore, the court’s determination as to whether a forum selection clause is contractually valid at step one of the analytical framework will almost always prove outcome determinative at step two.26 A passing reference to a “contractually valid” clause in a footnote thus becomes the hinge upon which the entire inquiry pivots.

The Article’s second goal is to provide an empirical assessment of federal practice with respect to clause enforcement post-Atlantic Marine. To this end, it draws upon an original, hand-collected dataset of 658 published and unpublished federal cases decided between 2014 and 2020 where one party challenged the enforceability of the forum selection clause. These cases reveal that federal courts uphold forum selection clauses roughly eighty-eight percent of the time when the court considers the issue of whether a clause is enforceable. The cases also demonstrate significant variation across states. Federal courts sitting in Florida enforced these clauses ninety-six percent of the time. Federal courts sitting in California, by comparison, enforced these clauses only eighty percent of the time. These findings make it possible, for the first time, to assess how rules relating to clause enforceability play out across the federal court system post–Atlantic Marine.

The Article’s third and final contribution to the literature is to suggest a number of reforms. Over the past fifty years, the federal courts have liberalized the rules for enforcing forum selection clauses to the point that these provisions are today given effect in the overwhelming majority of cases. It is no exaggeration to state that these provisions operate as battering rams that smash their way to the chosen court in all but the most extraordinary cases. This trend clearly benefits large corporations who have the leverage to draft take-it-or-leave-it agreements that mandate litigation occur in a forum that favors them. This trend just as clearly disfavors consumers, employees, and other individuals who lack the power to push back on clauses that require them to travel great distances to bring a lawsuit. In an attempt to rebalance the scales, the Article advances proposals for reform. Each of these proposals may be realized without significant modifications to existing doctrine, without the intervention of the U.S. Supreme Court, and without the enactment of new federal legislation. Instead, they call for the lower federal courts to take a step back, take a hard look at the legal regime they have created, and make slightly different choices at the margins. Although each of these reforms may seem small, their cumulative effect would be to create a fairer and more equitable enforcement regime for forum selection clauses in federal court.

The Article proceeds as follows. Part I discusses whether a forum selection clause is valid. Part II canvasses the law relating to the proper interpretation of forum selection clauses. Part III considers the complex question of clause enforceability. It provides a by-the-numbers empirical account of how the federal courts resolved enforceability cases between 2014 and 2020. Part IV makes the case for a number of pragmatic reforms which would, if adopted, make it easier for resisting parties to persuade the court that a forum selection clause is not “contractually valid” as that term is used in Atlantic Marine.

Validity

As every first-year law student learns, determining whether a valid contract exists can be a complicated undertaking. One must first inquire as to whether the contract was properly formed via offer, acceptance, and consideration.27 One must then inquire as to whether there exists a viable contract defense such as minority, fraud, mutual mistake, unconscionability, lack of capacity, or a failure to comply with the statute of frauds.28 If a contract was never properly formed, or if the defendant can establish the existence of a viable defense, the contract is invalid. In such circumstances, it does not matter whether the putative agreement has been breached or whether the plaintiff has suffered damages. An invalid contract is an unenforceable contract.

The same is true for forum selection clauses. An invalid clause has no legal effect.29 As a threshold question, therefore, a court called upon to dismiss or transfer a case on the basis of a forum selection clause must determine whether the clause is valid as a matter of contract law. This inquiry into validity is distinct and separate from the inquiry into how the contract should be interpreted or whether the clause is enforceable. Issues relating to interpretation are addressed in Part II. Issues relating to enforceability are addressed in Part III. This Part is focused exclusively on the question of whether a forum selection clause is valid.

Choice of Law

When a federal court is called upon to determine whether a forum selection clause is valid, it must first decide whether to apply state or federal law. Although the U.S. Supreme Court has held that federal law governs the enforceability of forum selection clauses when suit is brought in federal court, it has never held that federal law governs the question of whether the clause is valid.30 It follows that questions of clause validity should be resolved under state law.31

The next logical question is which state’s contract law to apply. In cases where the contract contains an enforceable choice-of-law clause, the federal courts should generally apply the law of the state named in the choice-of-law clause to determine whether the forum selection clause is valid.32 In cases where the contract omits a choice-of-law clause, the courts should perform a choice-of-law analysis and then apply the contract law of the appropriate state.33

Contract Law

Over the past decade, the federal courts have held that forum selection clauses were invalid on a number of occasions. In some cases, a clause was deemed invalid because the plaintiff never signed the agreement containing the clause.34 In other cases, courts have held that the clause never became a part of the contract after conducting a battle-of-the-forms analysis under Uniform Commercial Code Section 2-207.35 In each of these instances, the courts applied rules of contract law to determine the validity of the forum selection clause.

In other instances, the courts have modified these rules to account for the unique attributes of forum selection clauses. There are three scenarios where the courts routinely apply these modified rules. The first scenario arises when the resisting party argues that he was fraudulently induced into signing the contract containing the forum selection clause. The second scenario arises when an enforcing party argues that it is entitled to take advantage of the forum selection clause by virtue of its status as a third-party beneficiary to the agreement. The third and final scenario arises when the resisting party argues that the forum selection clause is not binding because the agreement has been terminated. Each of these situations is examined below.

Fraud

The courts have long recognized that fraud constitutes a valid defense to a breach of contract claim.36 If one party persuades another to sign a contract on the basis of an intentional misrepresentation, the agreement is invalid as a matter of law. This rule operates differently, however, when it comes to forum selection clauses. It is not enough for the resisting party to show that the contract was induced by fraud. Instead, the resisting party must show that the clause itself was induced by fraud.37

The origins of this rule may be traced back to Scherk v. Alberto-Culver Co., a case decided by the U.S. Supreme Court in 1974.38 In Scherk, the Court stated that an arbitration clause could only be invalidated on the basis of fraud if the clause itself was procured by fraud.39 The resisting party must show, in other words, that the parties specifically discussed the arbitration clause at the time of contracting and that the resisting party agreed to write it into the agreement due to fraudulent representations made by the other party.40 The purpose of this rule was to make it easier to enforce arbitration clauses by making it more difficult for the resisting party to escape the clause by arguing the entire contract was the product of fraud.41

This rule soon migrated from arbitration clauses to forum selection clauses.42 The migration produced a line of doctrine that makes it that much harder to persuade a court that a forum selection clause is invalid on the basis of fraud. A recent review of more than a thousand state and federal cases turned up only a handful of instances where the resisting party succeeded in convincing a court that a clause was invalid due to fraud.43 On the other side of the ledger, there were dozens of cases where the courts rejected the resisting party’s argument because there was no showing that the clause itself (separate and apart from the contract) was procured by fraud.44

The upshot is that the traditional fraud defense operates differently in the context of forum selection clauses. If the defendant argues that a price term, an indemnification provision, or an express warranty is invalid because the contract was induced by fraud, it need only show that the contract as a whole was the product of the plaintiff’s misrepresentations. If the defendant argues that the forum selection clause was induced by fraud, by comparison, it must prove that the clause itself was fraudulently induced.

Rights and Obligations of Third Parties

Under ordinary circumstances, a non-party to a contract may assert rights under that agreement as a third-party beneficiary only if she comes forward with evidence showing a “clear and definite intent” on the part of the contracting parties to confer an “enforceable benefit” upon her.45 Thereafter, the beneficiary may go to court to enforce the agreement notwithstanding the fact that she did not sign it. In the context of forum selection clauses, corporate executives, affiliates, and subsidiaries sometimes argue that they are third-party beneficiaries to forum selection clauses written into contract signed by a particular company.46 These claims rarely succeed.47 It is no easy task to persuade a court that the corporate parties to an agreement “clearly and definitely” intended to confer an “enforceable benefit” upon their executives, affiliates, and subsidiaries.

This state of affairs has prompted the courts to make it easier for non-parties to take advantage of forum selection clauses.48 Imagine a scenario where a parent company and its subsidiary are sued in Texas. Both defendants move to transfer the case to New York pursuant to a forum selection clause in the contract. If the subsidiary is not able to partake of the forum selection clause, then the suit against the subsidiary will proceed in Texas while the suit against the parent company will be transferred to New York. This is inefficient and a waste of judicial resources.

To avoid this outcome, the courts have reworked the doctrine of third-party beneficiaries to make it easier for non-parties to take advantage of forum selection clauses.49 Many courts now apply a new doctrinal test—the closely-related-and-foreseeable test—to determine whether a non-signatory is covered by a forum selection clause.50 Under this new test, there is no need to show that the non-signatory is a third party beneficiary of the agreement.51 One need only establish that the person seeking to take advantage of the forum selection clause is so “closely related” to the signatory that it was “foreseeable” that they would be covered under the clause.52 This is a less demanding standard that makes it more likely that litigation involving related parties will proceed in the same forum. The closely-related-and-foreseeable test constitutes another deviation from the ordinary rules of contract law that is applied exclusively in the context of forum selection clauses.53

Termination or Cancellation

When a contract is terminated or cancelled, the future obligations of both parties to the agreement are ordinarily extinguished.54 When a consultant terminates an agreement with a client, for example, the consultant is no longer obligated to provide future services to that client and the client need not pay the consultant for future work. So long as the party terminating or cancelling the agreement does so in a manner consistent with the terms of the contract, that contract will not bind the parties going forward.

This rule, like the ones discussed above, operates differently when applied to forum selection clauses. The courts have generally held that a forum selection clause will continue to bind the parties to an agreement even after that agreement has been terminated or cancelled.55 Even though the parties have decided to extinguish most of their contractual obligations to one another, the courts have reasoned, they probably wanted to preserve the provisions in the agreement relating to dispute resolution.56 Terminating an agreement is therefore not enough to cast off the obligations imposed by a forum selection clause. There must be a specific agreement terminating the clause itself before it will cease to have any legal effect.

*    *    *

The first step in determining whether a forum selection clause is contractually valid is to ascertain whether that clause is valid as a matter of contract law. In most cases, this inquiry merely requires the court to apply traditional common law rules to make sure that a contract was properly formed. In a few cases, however, the courts have modified these rules to account for the unique attributes of forum selection clauses. If the clause is deemed valid, the court must then seek to ascertain the precise meaning of the words and phrases in the clause to determine whether the clause is exclusive and applicable. This issue is addressed in the next Part.

Interpretation

Questions sometimes arise relating to the intended meaning of the forum selection clause. The relative brevity of these clauses means that the same interpretive questions tend to occur and reoccur. Over time, the courts have developed a number of interpretive rules—canons of construction—that seek to resolve these interpretive questions in a manner that aligns with the expectations of most contract users.

This Part discusses two of these interpretive rules.57 After a brief discussion of choice of law, it surveys the rules that the courts apply to determine whether a clause is exclusive or nonexclusive. It then analyzes the rules relating to scope.

Choice of Law

Most federal courts have held that state law governs the proper interpretation of a forum selection clause.58 Only the Ninth Circuit has held that the interpretation of a forum selection clause is an issue governed by federal law.59 The position taken by the Ninth Circuit is difficult to defend. Although the U.S. Supreme Court has suggested that federal law governs the question of whether a forum selection clause is enforceable—an issue addressed in Part III—that Court has never held that federal law governs the question of how to interpret a forum selection clause.60 Accordingly, a federal court should apply state law to interpret the clause. In deciding which state’s law to apply, it should follow the same choice-of-law rules discussed above in the discussion of validity.61

Exclusivity

One interpretive issue that frequently arises in the context of a forum selection clause is whether the clause is exclusive or nonexclusive.62 An exclusive clause provides that a dispute must be resolved in the chosen court and nowhere else.63 A nonexclusive clause consents to jurisdiction or venue in the courts of the chosen jurisdiction but does not foreclose the possibility that disputes may be resolved elsewhere.64 Exclusive clauses are also known as mandatory forum selection clauses.65 Nonexclusive clauses are also known as permissive forum selection clauses.66

To determine whether a forum selection clause is mandatory or permissive, the courts will look to see whether the clause contains so-called “language of exclusivity” that expresses an intent to litigate in the chosen courts and nowhere else.67 If a clause states that a dispute “must” be resolved in the chosen court, for example, or that a dispute may “only” be resolved in the chosen court, or that a dispute “shall” be resolved in the chosen court, then it will be deemed exclusive.68 Similarly, if a clause states that the chosen court shall have “exclusive” or “sole” jurisdiction to hear the case, the clause will be deemed to possess the necessary language of exclusivity.69

If a clause is nonexclusive, the federal courts are markedly less likely to dismiss or transfer the case. This is because a nonexclusive clause lacks the all-important language of exclusivity. If a clause merely states that the parties “consent” or “submit” to jurisdiction or venue in the chosen court, for example, then the clause is likely to be deemed nonexclusive.70 Similarly, if a clause states that the chosen court shall have “nonexclusive” jurisdiction, then the clause will not be interpreted to preclude litigation elsewhere.71 Although these clauses facilitate the bringing of a suit in the chosen jurisdiction, they cannot compel courts to refrain from hearing cases over which they otherwise possess jurisdiction. When parties bicker about the intended meaning of a forum selection clause, a common interpretive dispute relates to whether the clause is exclusive or nonexclusive. Under Atlantic Marine, only exclusive clauses are “contractually valid.”

Scope

When the contracting parties write a forum selection clause into their contract, there can be little doubt that they are selecting a forum in which to resolve future claims for breach of contract. What happens, however, when one of the parties asserts a non-contract claim against the other? Are tort and statutory claims covered by the forum selection clause? Or does the clause only cover contract claims? This interpretive issue is regularly presented to the federal courts.72

The easiest cases are those where the clause contains language suggesting that it covers noncontractual claims. If a clause states that any dispute “relating to” or “arising in connection with” the agreement must be resolved in the chosen court, for example, then all courts agree that it is broad enough to cover tort and statutory claims that have a connection to the contract.73 If a clause does not contain such language, however, then the federal courts will apply one of several canons of construction that assign a presumptive meaning to it.74 Unfortunately, there is considerable diversity of practice when it comes to these interpretive rules.

Some federal courts take the position that “tort and statutory claims are never covered by a generic forum selection clause” because such “claims do not originate in the contract.”75 Since these claims “originate in the common law of tort” or the relevant statute, these courts reason, they are not covered by the clause.76 This narrow approach has been rejected by other courts on the theory that it is not in keeping with the likely intent of most contracting parties.77 If the parties took the time to write a forum selection clause into their agreement, these courts reason, it seems implausible that they would never want that clause to apply to related tort or statutory claim under any circumstances.78

Other federal courts have held that non-contractual claims come within the ambit of a generic forum selection clause if they “arise out of the same operative facts as a parallel” claim for breach of contract.79 These courts ask whether the same facts pled in support of the contract claim could also be pled in support of the tort or statutory claim.80 If so, then these courts will hold that the non-contract claim is covered by the clause.81

Still other federal courts take the position that non-contractual claims are covered by generic forum selection clauses when it is necessary to refer back to the contract in order to resolve these claims.82 If a non-contractual claim cannot be resolved without first interpreting or construing the contract, for example, these courts hold that the claims are covered by a generic forum selection clause.83 Similarly, if a non-contractual claim cannot be resolved without first determining whether the defendant is in compliance with the contract, these courts generally hold that the claims fall within the ambit of the clause.84 Some courts have also held that a non-contractual claim is covered by a generic forum selection clause where the viability of the claim ultimately depends on the existence of a contractual relationship between the parties.85

Finally, some federal courts have adopted a hybrid approach that combines one or more of the interpretive rules set forth above.86 The Eighth Circuit, for example, has taken the position that non-contractual claims are covered by a generic forum selection clause when the claim requires interpretation of the agreement or the claim ultimately depends on the existence of a contractual relationship or the claim involves the same operative facts as a parallel claim for breach of contract.87 This hybrid approach makes it more likely that a generic clause will be given a broad scope because it gives the party arguing that the clause applies to non-contract claims multiple bites at the interpretive apple.88 If a clause is not broad enough to cover the claims at issue, then it is not “contractually valid” as that term is used in Atlantic Marine.

*    *    *

The second step in determining whether a forum selection clause is subject to the analytical framework set forth in Atlantic Marine is to interpret the clause. A clause is only “contractually valid” when is it is exclusive and broad enough to cover the claims asserted. If a clause satisfies these criteria, then the courts will proceed to consider whether it is enforceable. The issue of enforceability is explored in the next Part.

Enforceability

The U.S. courts have long struggled to determine precisely when forum selection clauses should be enforced. Prior to 1972, most U.S. jurisdictions held that forum selection clauses were per se unenforceable.89 The courts in this era were concerned that the routine enforcement of forum selection clauses would divert cases to jurisdictions where the defendant’s personal, social, or political standing could affect the outcome of the case.90 This outcome, in the eyes of many courts, would serve “to bring the administration of justice into disrepute.”91 These courts also objected to forum selection clauses because they were viewed as attempts to use private agreements to “oust” the courts of jurisdiction that was otherwise granted to them by law.92 In the view of the U.S. Supreme Court, such “agreements in advance to oust the courts of the jurisdiction conferred by law are illegal and void.”93 In the face of such decisions, contract drafters rarely bothered to write forum selection clauses into their agreements prior to 1950.94

In the second half of the twentieth century, however, the judicial hostility to forum selection clauses began to wane. In 1968, the Model Choice of Forum Act was approved by the National Conference of Commissioners on Uniform State Laws.95 This Act took the position that forum selection clauses were presumptively enforceable.96 In 1971, the Second Restatement of Conflict of Laws similarly took the position that forum selection clauses were enforceable so long as they were “fair and reasonable.”97 The most momentous shift, however, occurred when the Supreme Court decided The Bremen v. Zapata Off-Shore Co. in 1972.98

In The Bremen, the Court overturned more than a century of precedents relating to the enforceability of forum selection clauses.99 The Court held that forum selection clauses were presumptively enforceable as a matter of federal admiralty law.100 It further held that the party seeking to avoid the clause bore a “heavy burden of proof” to convince a court that it should not be enforced.101 The Court also announced two exceptions to the rule of presumptive enforceability.102 First, it held that forum selection clauses were not enforceable when they were contrary to public policy.103 Second, it held that forum selection clauses were not enforceable when they were unreasonable.104 The enforcement framework set forth in The Bremen is now widely used by federal courts in the United States to determine a forum selection clause should be given effect for reasons having nothing to do with its validity.

This Part provides an account of the doctrinal rules that determine when a forum selection clause is enforceable under the framework established by The Bremen. It first addresses the question of what law the federal courts should apply to evaluate the enforceability of a forum selection clause. It then surveys recent case law in an attempt to determine when a clause is likely to be deemed unreasonable or contrary to public policy.

Choice of Law

Although federal courts apply state law to issues of validity and interpretation, they generally apply federal law to assess whether a clause is enforceable. They cite Stewart Org., Inc. v. Ricoh, a case decided by the U.S. Supreme Court in 1988, to justify this approach.105 In Stewart, an Alabama company entered into a dealership agreement with a company headquartered in New Jersey.106 That agreement contained a forum selection clause providing that any disputes had to be resolved in New York.107 When the business relationship soured, the Alabama company sued the New Jersey company in federal court in Alabama.108 The New Jersey company invoked the forum selection clause and moved to transfer the case to New York pursuant to 28 U.S.C. § 1404(a).109 The Alabama company sought to defeat the transfer motion by arguing that Alabama public policy—as articulated by the state’s courts as a matter of state common law—provided that the forum selection clause was unenforceable.110

The decision rendered by the Eleventh Circuit in Stewart squarely addressed the issue of whether the enforceability of a forum selection clause was governed by state or federal law.111 That court held that federal law should be applied.112 On appeal, the Supreme Court came at the question from a different angle. Instead of analyzing the law that governed the issue of clause enforceability, the Court instead analyzed the law that governed the motion to transfer.113 Having framed the issue in this manner, it should come as little surprise that the Court concluded that a motion to transfer a case from one federal court to another was governed by the federal transfer statute, 28 U.S.C. § 1404(a), rather than the common law of Alabama.114 In analyzing whether a transfer should occur, the Court held that a forum selection clause should “receive neither dispositive consideration . . . nor no consideration . . . but rather the consideration for which Congress provided in § 1404(a).”115 Since Section 1404(a) does not contain any express reference to forum selection clauses, these instructions provided limited guidance to lower courts.

Nevertheless, a majority of federal courts of appeal have read Stewart to mean that federal law governs the issue of enforceability at step one of the Atlantic Marine analysis.116 As a rule, these courts look to the test laid down in The Bremen­ to determine whether a clause is enforceable. In practice, this means that a federal court will generally decline to enforce a forum selection clause when it is contrary to public policy or unreasonable.

Public Policy

In The Bremen, the Supreme Court held that “[a] contractual choice-of-forum clause should be held unenforceable if enforcement would contravene a strong public policy of the forum in which suit is brought, whether declared by statute or by judicial decision.”117 In practice, courts rarely rely on judicial decisions as a basis for declining to enforce forum selection clauses. Where there is a statute on the books stating that these clauses shall not be given effect, however, the courts can and do invoke public policy as a basis for non-enforcement.118 Some of these statutes are federal laws enacted by Congress. Others are state statutes enacted by a state legislature. We begin with the federal statutes.

Federal Public Policy

The federal courts rely on three types of federal statutes to invalidate forum selection clauses on public policy grounds. First, if a federal statute contains a special venue provision that requires or allows a suit to be brought in a particular place, a forum selection clause selecting the courts of another place may be deemed contrary to public policy. Second, if a federal statute contains language stating that the rights conferred by the statute may not be waived, and if enforcement of the forum selection clause is likely to lead to the waiver of these rights, then enforcing the forum selection clause may be deemed contrary to public policy. Third, and finally, the courts have held that forum selection clauses are not enforceable in bankruptcy court when enforcement would interfere with the policy goal of channeling claims against a debtor into a single forum.

Special Venue Provisions

There are no federal statutes that specifically direct the courts not to enforce forum selection clauses. There are, however, a great many federal statutes that contain special venue provisions. The Federal Employee Liability Act (“FELA”), for example, provides that “an action may be brought in . . . the district of the residence of the defendant, or in which the cause of action arose, or in which the defendant shall be doing business at the time of commencing such action.”119 The Carmack Amendment states that “[a] civil action under this section may be brought against the carrier alleged to have caused the loss or damage, in the judicial district in which such loss or damage is alleged to have occurred.”120 The Miller Act provides that “[a] civil action brought under this subsection must be brought . . . in the United States District Court for any district in which the contract was to be performed and executed, regardless of the amount in controversy.”121 Each of these statutes contains clear rules about the venue where a lawsuit may be brought. The question is whether these special venue provisions render unenforceable forum selection clauses stating that a suit must be brought in a different court.

The courts have answered this question differently for different statutes. With respect to FELA claims, for example, the U.S. Supreme Court has held that a contractual provision that purports to limit the ability of a plaintiff to choose a forum is void as against public policy.122 With respect to the Carmack Amendment, the lower federal courts have similarly held that the special venue provisions in that Act preempt any forum selection clause that purports to limit where a suit may be brought.123 When a plaintiff asserts a claim under FELA or the Carmack Amendment, therefore, the federal courts will generally decline to enforce a forum selection clause selecting a court other than the one where the lawsuit was filed by pointing to the special venue provisions in the applicable act.

The courts have reached different conclusions with respect to claims arising under other federal statutes. With respect to the Employee Retirement Income Security Act of 1974 (“ERISA”), for example, most courts have held that that statute’s special venue provisions do not invalidate forum selection clauses written into plan documents.124 Most courts have similarly held that the venue provisions in the Miller Act do not render clauses in construction contracts unenforceable.125 There is disagreement among the courts as to the effect of the venue provisions in Title VII of the Civil Rights Act of 1964.126 The courts have held that the decision by Congress in 2008 to delete the special venue provisions in the Jones Act means that forum selection clauses are now enforceable in cases brought under that statute.127 Finally, courts have held that the special venue provisions in the American with Disabilities Act do not preclude the enforcement of forum selection clauses requiring suits arising under that statute to be litigated somewhere else.128

Anti-Waiver Provisions

The federal courts will also sometimes refuse to enforce foreign forum selection clauses on public policy grounds when the applicable statute contains an anti-waiver provision stating that the rights conferred by the statute may not be waived via contract.

The Securities Act of 1933 and the Securities Exchange Act of 1934, for example, both contain anti-waiver provisions.129 If the contracting parties were to write an express provision into their agreement for the purchase or sale of securities waiving the protections provided by these Acts, that provision would be voided by the anti-waiver provisions.130 If the parties were to write a foreign choice-of-law clause selecting the laws of a foreign nation to govern their contract, and if the foreign laws lacked investor protections that were equivalent to those provided by federal securities law, then the choice-of-law clause would likewise be voided by the anti-waiver provisions. Lastly, and most importantly for our purposes, if the parties were to write a forum selection clause selecting the courts of a foreign nation into their contract, and if a U.S. court believed that these foreign courts were likely to apply a foreign law that did not provide investor protections equivalent to those provided by federal securities law, then the foreign forum selection clause would also be voided by the anti-waiver provision.131 In this way, an anti-waiver provision may lead to the non-enforcement of a forum selection clause.

In a series of cases decided in the 1990s, the federal courts of appeal were asked to decide whether the securities laws of England provided equivalent protection to those of the United States when Lloyd’s of London sought to enforce English forum selection clauses against U.S. plaintiffs.132 In each of these cases, the court held that the forum selection clause at issue was enforceable because the securities laws of England were similar to the securities laws of the United States.133 These cases also made clear, however, that if the choice-of-law clause and the forum selection clause had operated in tandem to select a foreign law that provided substantially less protection to investors, the forum selection clause would have been unenforceable because enforcement would have led to the waiver of non-waivable rights under federal securities laws.134

In cases where the foreign court seems likely to apply U.S. law—or a foreign law that confers equivalent legal protections—the U.S. court will generally enforce the foreign forum selection clause notwithstanding an anti-waiver provision.135 In cases where the foreign court seems likely to apply a foreign law that would result in the loss of non-waivable rights, by contrast, the U.S. court will generally invoke the anti-waiver provision and refuse to enforce a foreign forum selection clause on public policy grounds.136

Bankruptcy

Bankruptcy is another area where the federal courts sometimes decline to give effect to forum selection clauses on public policy grounds. The intuition underlying these cases is straightforward. The purpose of bankruptcy law is to bring all of the debtor’s creditors together into a single forum for purposes of determining how to allocate the assets in the bankruptcy estate.137 If the bankruptcy court were to enforce forum selection clauses redirecting certain claims to other courts, then this purpose would be thwarted.138 Accordingly, most federal courts have held that the public policy in favor of centralized resolution of claims provides a valid public policy reason not to enforce forum selection clauses in bankruptcy cases.139

State Public Policy

Over the past fifty years, state legislatures across the United States have enacted more than two hundred statutes that purport to invalidate forum selection clauses when written into various types of contracts.140 Depending on the state, a forum selection clause may be invalid when written into a child-support contract, a consumer contract, a consumer credit agreement, a consumer lease, a construction contract, an employment agreement, a foreclosure agreement, a franchise agreement, a high-cost home loan agreement, an insurance agreement, a payday lending agreement, a sales representative agreement, a student-loan agreement, a structured settlement agreement, a timeshare agreement, or a wholesaler agreement.141 Several states have enacted statutes directing their courts not to enforce foreign forum selection clauses when enforcement will lead to the waiver of certain constitutional rights.142 And at least one state—Idaho—has enacted a statute invalidating all forum selection clauses that choose the courts of any other state.143

This tsunami of state statutes mandating non-enforcement of forum selection clauses has led to a divergence in federal practice. Some federal courts hold that state statutes declaring forum selection clauses void on public policy grounds are dispositive on the issue of enforceability and decline to enforce the clause. Other federal courts have held that state statutes are merely one factor to consider in assessing the issue of enforceability as a matter of federal law. These courts generally enforce a clause even in the face of an invalidating state statute.

Approach #1: State Statute Is Dispositive

Some federal courts have held that the existence of an invalidating state statute is dispositive with respect to the issue of clause enforceability. These courts will decline to enforce a forum selection clause as a matter of federal law when a state statute directs state courts sitting in the same state not to enforce it.144

The Ninth Circuit recently adopted this approach in Gemini Technologies, Inc. v. Smith & Wesson Corporation.145 In that case, an Idaho company sued a Massachusetts company in federal court in Idaho.146 The Massachusetts company moved to dismiss the case on the basis of forum non conveniens, citing a forum selection clause in their agreement selecting the state courts of Delaware.147 The Idaho company sought to defeat the motion to dismiss by invoking an Idaho state statute invalidating all outbound forum selection clauses.148 The Idaho company lost at trial and appealed the case to the Ninth Circuit. 149

The Ninth Circuit began its analysis by observing that the analytical framework set forth in Atlantic Marine only applied when a forum selection clause was “contractually valid.”150 It then looked to the test set forth in The Bremen to determine whether the clause at issue in the case was enforceable.151 In the court’s words:

[The plaintiff] has identified an Idaho statute that clearly states a strong public policy. Idaho Code § 29-110(1) provides: “Every stipulation or condition in a contract, by which any party thereto is restricted from enforcing his rights under the contract in Idaho tribunals . . . is void as it is against the public policy of Idaho.” . . . [S]atisfaction of Bremen’s public policy factor continues to suffice to render a forum-selection clause unenforceable. Bremen held that “[a] contractual choice-of-forum clause should be held unenforceable if enforcement would contravene a strong public policy of the forum in which suit is brought, whether declared by statute or by judicial decision.” We have found nothing in Atlantic Marine that compels a different rule.152

This approach treats the state statute as dispositive on the issue of clause enforceability at step one of the Atlantic Marine analysis. It does not address the question of whether the case should be transferred under Section 1404(a) or dismissed under forum non conveniens at step two of that analysis.153 The Ninth Circuit remanded the case for the lower court to consider whether the case should be dismissed under the usual forum non conveniens analysis that applies absent a contractually valid forum selection clause.154

Approach #2: State Statute Is One of Several Factors

Other federal courts reject the notion that state statutes—standing alone—provide a conclusive answer to the enforceability question.155 These courts treat state public policy as merely one factor to consider in determining whether a forum selection clause should be enforced as a matter of federal law.156

In Gita Sports Ltd v. SG Sensortechnik GmbH & Co. KG, for example, a federal court in North Carolina was asked to enforce a forum selection clause that stated that all disputes had to be resolved in Germany.157 The U.S.-based plaintiff argued that the clause was unenforceable because the North Carolina legislature had enacted a statute directing courts not to enforce forum selection clauses.158 In evaluating this argument, the court did not view state public policy as dispositive.159 In the court’s words:

Choice of forum . . . provisions may be found unreasonable if (1) their formation was induced by fraud or overreaching; (2) the complaining party “will for all practical purposes be deprived of his day in court” because of the grave inconvenience or unfairness of the selected forum; (3) the fundamental unfairness of the chosen law may deprive the plaintiff of a remedy; or (4) their enforcement would contravene a strong public policy of the forum state.160

The court acknowledged that enforcing the German forum selection clause would be contrary to North Carolina public policy.161 It ultimately concluded, however, that this fact was not dispositive because the contract had not been induced by fraud or overreaching, the complaining party would not be deprived of its day in court, and the chosen law was not fundamentally unfair.162 Since three of the four factors in the balancing weighed in favor of enforcement, the court held that the clause was enforceable notwithstanding the North Carolina statute.163

There are, broadly speaking, three problems with this approach. First, there is nothing in The Bremen to suggest that a balancing test should be used to evaluate enforceability. In that case, the Court clearly stated that a forum selection clause was unenforceable if it was contrary to the public policy of the forum.164 The Court also stated that a clause was unenforceable if the clause would deprive the resisting party of its day in court, if it was induced by fraud or overreaching, or if there was another basis for deeming the clause unreasonable.165 Each of these bases for non-enforcement is separate and independent from all the others. To require a showing that a clause is both contrary to public policy and unreasonable is to misread the test laid down in The Bremen.

Second, to the extent that a balancing test frequently leads the federal courts to disregard statutes that are binding on state courts that sit in the same jurisdiction, it creates an Erie problem.166 In Erie Railroad Company v. Tompkins, the Supreme Court held that the lower federal courts should seek to avoid creating situations where litigants were encouraged to forum shop between state and federal courts in the same state.167 If a state has enacted a statute invalidating forum selection clauses, the state courts sitting in that jurisdiction will enforce the statute as written. If the federal courts in that jurisdiction apply a balancing test that routinely leads to the clause being enforced, defendants will have a strong incentive to remove the suit to federal court so as to take advantage of a more favorable federal rule.168 This approach thus encourages precisely the sort of forum shopping that Erie sought to discourage.169

Third, the use of a balancing test to determine enforceability is needlessly complicated. Regardless of how the enforceability inquiry is resolved, the court must then apply a second balancing test to decide whether the case should be transferred or dismissed under Section 1404 or a forum non conveniens analysis.170 To stack one balancing test on top of another balancing test creates needless uncertainty and greatly complicates the work of the court.

Reasonableness

The Supreme Court held in The Bremen that forum selection clauses are unenforceable when “enforcement is shown by the resisting party to be ‘unreasonable’ under the circumstances.”171 In defining an “unreasonable” clause, the Court imposed a “heavy burden” on the party resisting enforcement. Mere inconvenience, the Court held, was not enough. Instead, a clause was only unreasonable if litigation in the chosen forum was “so gravely difficult and inconvenient that [the resisting party] will for all practical purposes be deprived of his day in court.”172 In imposing this standard, the Court noted that the clause at issue was the product of “arm’s-length negotiation by experienced and sophisticated businessmen.”173The Bremen Court thus left open the question of whether this same standard would apply when evaluating the enforceability of a forum selection clause in other types of contracts.

Nineteen years later, the Supreme Court addressed the question of when a forum selection clause was “unreasonable” in consumer contracts of adhesion.174 In Carnival Cruise Lines v. Shute, a woman living in Washington slipped and fell while on board a cruise ship. She sued the cruise company in federal court in Washington.175 The company sought to enforce a forum selection clause in the passenger ticket requiring all suits against it to be brought in Florida.176 The Court held that the clause was reasonable—and enforceable—notwithstanding the fact that it was written into a contract of adhesion drafted by a multinational company and required the plaintiff to travel several thousand miles to bring the lawsuit.177 Although the Carnival Cruise court hinted that a clause might be unreasonable if its existence was not “reasonably communicated” to the plaintiff, the effect of the decision was to dramatically curtail the range of cases where a clause might be deemed invalid on the basis of unreasonableness.178

A review of federal court cases decided after Atlantic Marine turned up a mere handful of cases where the federal courts declined to enforce a clause on the grounds that it was unreasonable. These cases suggest that a clause may be deemed unreasonable when: (1) the chosen court lacks subject matter jurisdiction to hear a case; (2) enforcement will result in duplicative litigation; (3) the clause was not reasonably communicated; (4) enforcement would deprive the resisting party of her day in court; or (5) the clause is fundamentally unfair.

Chosen Court Lacks Subject Matter Jurisdiction

If the court named in the forum selection clause lacks subject matter jurisdiction to hear the dispute, then the federal courts will decline to enforce the clause. In Hare v. YJ Sales, Inc., for example, a federal district court refused to enforce a forum selection clause requiring a copyright claim to be brought in the state courts of Rhode Island because state courts do not have subject matter jurisdiction to hear copyright claims.179 In BH Servs. v. FCE Benefit Adm’rs Inc., a federal district court refused to enforce a forum selection clause requiring an ERISA suit to be brought in San Mateo County, California, because there was no federal courthouse in that county and state courts lack subject matter jurisdiction to hear ERISA claims.180 In Alamo Masonry & Constr. Contractors, LLC v. Air Ideal, Inc., a federal district court hearing a claim arising under the Miller Act refused to enforce a forum selection clause calling for disputes to be resolved in Seminole County, Florida, because that county lacked a federal courthouse and state courts lack subject matter jurisdiction to hear claims arising under the Miller Act.181

Enforcement Will Result in Duplicative Litigation

The federal courts have also sometimes held that it is unreasonable to enforce a forum selection clause when to do so will lead to inefficient and duplicative litigation. In Carney v. Beracha, for example, a federal district court in Connecticut held that “enforcement of the forum selection clauses to bar this action from this court would be unreasonable, because it would require piecemeal litigation in multiple fora and, in some cases, might require multiple courts to adjudicate claims covering only portions of each transaction.”182 In Laferte v. myFootpath, LLC, a federal court in California noted that “courts find forum selection clauses unreasonable where there is a possibility of prejudice through conflicting judgments by duplicitous litigation in multiple courts, or where refiling subordinate claims in a separate court would result in judicial inefficiency.”183 And in Idingo LLC v. Cohen, a federal court in New Jersey declined to enforce a forum selection clause because it would lead to “separate cases involving extremely similar facts and claims . . . in two court systems” and “fragmented and duplicative litigation.”184

Clause Was Not Reasonably Communicated

In Carnival Cruise, the Supreme Court offered no opinion on the question of whether a forum selection clause might be unenforceable if it was never “reasonably communicated” to the resisting party. In the years since that case was decided, a number of lower courts have invalidated clauses on this basis. In Azzia v. Royal Caribbean Cruises and Touloumes v. Kerzner Int’l Bah., Ltd., for example, a federal district court in Florida refused to enforce a forum selection clause in a contract between a passenger and a cruise company because the company failed to show that an email containing a contract was ever sent to the plaintiff.185

In Young v. Holland Am. Line, N.V., the court concluded that “a term disclosed only after a purchase is made and at a time when cancellation would cost up to 75% of the ticket price” was unenforceable.186 And in Hussein v. Coinabul, LLC, a federal district court in Illinois declined to enforce a forum selection clause because it was “hidden behind a hyperlink that is tucked away at the bottom of its website.”187 In each of these cases, the court concluded that the existence of the forum selection clause was not reasonably communicated to the plaintiff and was therefore unenforceable.

Enforcement Will Deprive Resisting Party of Day in Court

A handful of federal courts have declined to enforce a forum selection clause on the grounds that it would deprive the resisting party of his day in court. In Grice v. VIM Holdings Grp., LLC, a federal district court in Massachusetts held that enforcing a forum selection clause selecting the courts of Illinois against a single mother residing in Massachusetts earning fifteen dollars per hour would deprive her of her day in court and was therefore unreasonable.188 In Lieberman v. Carnival Cruise Lines, a federal district court in New Jersey held that enforcing a Florida clause against a thirty-nine-year-old mother with four children in New Jersey undergoing chemotherapy for stage four cancer would deprive her of her day in court and was therefore unreasonable.189 And in Harmon v. DynCorp Int’l, Inc., a federal district court in Virginia held that enforcing a United Arab Emirates (“UAE”) clause against a Virginia-based employee would deprive him of his day in court because of the distance and substantial difference between the laws of the United States and the UAE and because the employment contract at issue was not written in Arabic and was hence invalid under the law of the UAE.190

Clause Is Fundamentally Unfair

In a smattering of cases, the federal courts have held that enforcing a clause was so profoundly unfair that it crossed the line into unreasonableness on this basis alone. In one case, a federal district court in Massachusetts refused to enforce an Iowa clause due to the circumstances under which the agreement was signed.191 The court noted that the plaintiff spoke only Spanish and could not read the English-language contract presented to him on a take-it-or-leave-it basis.192 The court pointed out that if the plaintiff had refused to sign, he would have been “stranded in Iowa without bus fare home and with a debt of $2,000.”193 Such a clause was, in the court’s view, unfair and hence unenforceable.194 In Davila v. Adesa Utah, LLC, a federal district court in Utah refused to enforce an Indiana clause due to its one-sided nature.195 Although the plaintiff was required to bring suit against the defendant in Indiana, the clause allowed the defendant to bring suit against the plaintiff wherever it wished.196 In light of this disparity, and amid other concerns about negotiating power and small print, the court concluded that the clause was unfair and therefore unenforceable.197

An Empirical Take on Enforceability

Legal scholars have long distinguished between the “law on the books” and the “law in action.”198 The law on the books is “the content of statutes, regulations, and judicial decisions,” while the law in action “refers to regularities describing how legal authorities enforce the ‘law on the books.’”199 The previous two Sections surveyed the law on the books as it relates to the enforcement of forum selection clauses in federal court. This Section describes how federal courts actually apply that law in the cases that come before them.

To achieve this end, it surveys every federal case relating to the enforceability of forum selection clauses handed down between January 1, 2014, and December 31, 2020. The screening criteria used to identify these cases are set forth in the Appendix. The search produced a dataset of 658 federal cases when the court considered the argument that a clause was unreasonable or contrary to public policy. I then reviewed each of these cases to determine whether the federal court ultimately chose to enforce the clause.200

I found that the federal courts enforced forum selection clauses eighty-eight percent of the time in cases where the issue of enforceability was raised. I also found that there were noteworthy differences in enforcement practices across states. The federal courts in Florida, for example, enforce forum selection clauses ninety-six percent of the time. The federal courts in California, by comparison, enforce forum selection clauses only eighty percent of the time. The results for every state with at least twenty federal court decisions during the applicable time period are set forth in Table 1.

 

I also calculated the rate of enforcement for each federal circuit by looking to all of the federal court decisions within the circuit that addressed the issue of clause enforceability during the relevant time period. The Eleventh Circuit had the highest enforcement rate at ninety-five percent. The Ninth Circuit had the lowest enforcement rate at eighty-one percent. These results are heavily influenced by the presence of Florida in the Eleventh Circuit and the presence of California in the Ninth Circuit. (I did not include the Federal Circuit or the D.C. Circuit in the survey due to their specialized subject areas.) The results for each of the federal courts of appeal are set forth in Table 2.

Overall, these data suggest that—regardless of circuit—federal courts enforce forum selection clauses in the overwhelming majority of cases.201

After calculating the overall enforcement rate by state and by circuit, I turned my attention to those cases where a federal court refused to enforce a forum selection clause. I reviewed each of the non-enforcement cases and coded the reason why the court had declined to give effect to the clause. The data show that the federal courts invoke public policy as basis for non-enforcement in eight percent of cases. They hold that clauses are unreasonable in three percent of cases. They hold that clauses are unenforceable because they were procured by fraud in just one percent of cases, as shown in Table 3.

These findings highlight the reluctance of the federal courts to conclude that a clause is unreasonable. It is far more common for these courts to invalidate a clause on the basis of public policy.

I next reviewed the public policy cases to determine why the court had declined to enforce the clause. I found that the courts invoked state public policy as a basis for non-enforcement in five percent of the cases. They invoked federal public policy as a basis for non-enforcement in three percent of the cases, as reported in Table 4.

The fact that the federal courts are more likely to invoke state public policy rather than federal public policy to invalidate a clause is somewhat surprising because, as discussed above, some federal courts routinely ignore state invalidating statutes. The sheer number of such statutes, however, and the fact that many of these statutes apply to contracts that are regularly litigated in federal court, helps to explain the disparity.

I next reviewed each of the cases where a court concluded that a clause was unreasonable. I found that the most common basis for deeming a clause unreasonable was that it was not reasonably communicated to the plaintiff. The other bases for finding a clause to be unreasonable all appeared in roughly equal numbers, as shown in Table 5.

The data suggest that the courts are not appreciably more likely to declare a clause unreasonable on one basis as opposed to another.

I then reviewed the cases to determine the identity of the defendant that invoked the forum selection clause as basis for transfer or dismissal. I found that ninety-eight percent of the cases involved an entity such as a corporation or a limited liability company. In only two percent of the cases was the sole defendant a natural person, as shown in Table 6.

This finding suggests that the routine enforcement of forum selection clauses by the federal courts overwhelmingly redounds to the benefit of business entities. It is very rare for a natural person, standing alone, to ask a court to enforce one of these provisions.202

Finally, I reviewed the cases to determine the identity of the plaintiff against whom the forum selection clause was invoked. In forty-eight percent of the cases, the plaintiff (or group of plaintiffs) consisted solely of natural persons. In fifty-two percent of the cases, at least one plaintiff was a business entity, as shown on Table 7.

This finding suggests that natural persons are more likely to have a clause enforced against them as plaintiffs than they are to invoke the clause as defendants.

Overall, these data provide important context for the doctrinal rules set forth above. While the federal courts occasionally refuse to enforce forum selection clauses on the grounds that they are contrary to public policy, such decisions are rare. It is even more uncommon for a federal court to refuse to enforce a clause on grounds that it is unreasonable. In the overwhelming majority of cases, the cases indicate that the federal courts will enforce a forum selection clause over the objections of the resisting party.

The Supercharged Forum Selection Clause

The foregoing discussion of validity, interpretation, and enforcement inform the core inquiry as to whether a forum selection clause is “contractually valid” for purposes of Atlantic Marine. With this account in mind, it is now useful to take a step back to consider precisely how much has changed over the past fifty years with respect to forum selection clauses. Each individual rule discussed above, considered in isolation, represents a small shift in the law. Viewed as a collective, however, the accumulated weight of these rules has led to a regime where forum selection clauses have become supercharged. The clause is now a battering ram capable of smashing its way to the courts of the chosen state in virtually every case where it is invoked. This is good for the corporations that rely on these clauses to channel litigation to their home jurisdictions. It is less good for consumers, employees, and other individuals against whom these clauses are routinely enforced.

The Rise of the Forum Selection Clause

In the nineteenth and early twentieth century, it will be recalled, forum selection clauses were per se unenforceable in most cases.203 In 1972, the U.S. Supreme Court jettisoned this rule.204 Henceforth, the Court held, such provisions should be viewed as presumptively enforceable when written into international commercial agreements concluded by sophisticated parties so long as the agreement was reasonable and consistent with public policy of the forum.205 In adopting this framework, the Court effectively rehabilitated the forum selection clause after years in the proverbial wilderness. This decision set the stage for a dizzying array of doctrinal innovations in the years to come, each of which served to amplify the power of the forum selection clause.

In 1974, the Supreme Court chose to modify the rules pertaining to fraud in the context of arbitration clauses.206 The Court held that it was not enough to prove that an arbitration clause was invalid because it was part of an agreement that had been procured by fraud.207 Instead, the resisting party had to show that the clause itself was procured by fraud.208 Although this decision was rendered in an arbitration case, the rule quickly migrated to cases involving forum selection clauses.209 Since it is exceedingly difficult in most cases for the resisting party in most cases to show that the forum selection clause was procured by fraud, the end result of this migration was to defang an important defense—fraud—that would ordinarily provide a contractual basis for proving that the clause should not be given effect.

In 1988, the Supreme Court decided Stewart and held—albeit obliquely—that federal law governed the question of whether a forum selection clause was enforceable when a federal court was asked to transfer a case under Section 1404(a).210 This decision served to undercut the ability of state courts and state legislatures to check the growing power of the forum selection clause.211 If the federal enforcement rule was more pro-enforcement than the state enforcement rule, an out-of-state defendant could evade state law by removing the case to federal court and asking that court to apply federal law.212

In 1991, the Supreme Court recast the reasonability exception established in The Bremen.213 The Court held that forum selection clauses could be reasonable even when written into consumer contracts of adhesion.214 The result was a significant shift in the law. Up to this point, one could plausibly argue that the legal test for enforceability was flexible enough to account for whether the resisting party was a business or a natural person. After Carnival Cruise, this was no longer the case.215 These clauses were now presumptively enforceable even when the resisting party was a natural person who lacked the bargaining power or the sophistication to negotiate the terms of the contract.

In the wake of Carnival Cruise, the federal courts began to create new contract rules that applied exclusively to forum selection clauses, all of which favored enforceability. The courts held that forum selection clauses survive the termination or cancellation of the contract.216 The courts held that non-signatories to a contract may nevertheless take advantage of a forum selection clause if they are so “closely related” to a contract signatory that it was “foreseeable” that they would be bound.217 Whereas before the forum selection clause had been special in that it was subject to review for unreasonableness and public policy, the clause was now special in that it: (1) was subject to unique fraud rules; (2) survived the termination of the contract; and (3) was routinely applied to benefit non-signatories who were not third-party beneficiaries to the agreement.

These innovations were followed by a number of others. Over the past decade, a number of courts have upheld “asymmetric” or “non-mutual” forum selection clauses that require one contracting party to sue in the chosen forum but allow the other party to sue whenever they want.218 Other courts have upheld so called “floating” clauses where the identity of the chosen forum could be changed after the contract was signed.219 Still others enforced clauses even when there was no possibility of recovery in the chosen jurisdiction because the statute of limitations had run.220 Finally, some courts now allow one party to sue the other for damages when it brings a lawsuit in a jurisdiction other than the one named in the forum selection clause.221

The willingness of the courts to adopt each of these doctrinal innovations has generally redounded to the benefit of the party in a position to dictate the terms of the agreement.222 Each and every one of the doctrinal innovations from the past fifty years—the rejection of the rule of per se invalidity, the narrowing of the fraud exception, the federalization of the law relating to enforceability, the narrowing of the reasonableness exception, the decision not to distinguish between business and consumer contracts, allowing clauses to persist beyond the end of the contract, expanding the clause to cover more non-signatories, the embrace of non-mutuality, the endorsement of floating clauses, the decision to enforce clauses even when the statute of limitations has run, and allowing suits for damage—has served to advance the interests of the party with more negotiating power in the drafting process.223 In practice, this means that the modern enforcement regime for forum selection clauses strongly favors the interests of large corporations at the expense of natural persons.224 With the possible exception of its doctrinal cousin—the arbitration clause—there is no contract provision that is so uniquely favored in modern litigation.225

The only body of contemporary judicial doctrine that reliably produces victories for parties resisting a forum selection clause are rules of interpretation.226 If a forum selection clause is non-exclusive, then it cannot compel a court to dismiss or transfer a case to the chosen forum. If a forum selection clause does not cover the claim asserted, then the claim is not subject to the clause. The availability of these interpretive arguments is, however, ultimately dependent on the inattention or carelessness of the contract drafter. If a contract drafter is well advised, it can draft a forum selection clause that is exclusive and broad enough to defeat any and all interpretive arguments put forward by the resisting party. At this point, the full array of doctrinal innovations discussed above may be brought to bear to persuade a court to enforce the clause.

At present, state statutes directing courts to disregard forum selection clauses provide the most robust check on the enforceability of these provisions.227 It will be recalled that state legislatures have enacted statutes directing courts not to enforce outbound forum selection clauses across a range of contract types.228 These laws generally allow individuals who have entered into contracts with large corporations to sue those corporations in that individual’s home jurisdiction. These statutes are, however, only sometimes enforceable in federal court. In jurisdictions where the federal courts discount state public policy as a basis for non-enforcement, these laws will not have any effect.

Recalibrating the Forum Selection Clauses

If one accepts that the pendulum has swung too far in the direction of enforcing forum selection clauses—as it clearly has—then one might also wonder how best to establish a more equitable equilibrium. A federal statute limiting the enforceability of forum selection clauses would accomplish this goal.229 The prospects for enacting such a statute in the current political environment are, however, not encouraging. A decision by the U.S. Supreme Court revisiting its decision in Carnival Cruise would likewise go a long way toward rebalancing the scales. The prospect that the current Court will issue such a decision is, however, similarly discouraging. Viewed through a purely pragmatic lens, therefore, the most viable means of recalibrating the enforcement regime is for the lower federal courts to adopt incremental changes that are permissible under existing precedent.

First, the federal courts should refuse to enforce forum selection clauses when they are contrary to the public policy of the state in which they sit as expressed in a state statute. Some federal courts of appeal have already taken this position.230 The Ninth Circuit, for example, has consistently declined to enforce forum selection clauses when to do so would be contrary to a state statute.231 Other federal courts, however, routinely enforce these provisions even when such action is flatly prohibited by a state statute.232 As discussed above, nothing in Stewart compels the conclusion that federal courts must ignore state statutes voiding forum selection clauses in cases where one party moves to dismiss on the basis of forum non conveniens. Ignoring such statutes, moreover, is inconsistent with the lessons of The Bremen, which specifically provide that a clause is unenforceable if it is contrary to the public policy of the forum.

Second, the federal courts should take a broader view of when a clause is unenforceable because it is unreasonable. As things stand, it is virtually impossible for a party resisting a clause to persuade a federal court that a clause is unreasonable, no matter how aged or disabled or impoverished the resisting party is.233 This unflinching commitment to enforcing clauses against any and all challenges to its reasonableness goes well beyond what is required by Carnival Cruise.

Third, the federal courts should construe special venue provisions in federal statutes as preempting forum selection clauses. As things currently stand, the courts have reached different decisions on this issue depending on the federal statute. This split is particularly curious in light of the fact that the Supreme Court clearly held in Boyd that special venue provisions in the FELA trump forum selection clauses.234 There is no reason why the lower federal courts should reach a different conclusion in claims arising under ERISA or the ADA given the textual parallels between the special venue provisions in these statutes and the one in the FELA.

Fourth, the federal courts should only apply the closely-related-and-foreseeable test to bind non-signatories when they have given their consent.235 The use of the test to bring consenting non-signatories within the scope of a clause is unobjectionable. The use of the test to bind a non-signatory to an agreement without its consent is far more troubling, particularly when the clause is invoked as a basis for asserting personal jurisdiction over a defendant. The closely-related-and-foreseeable test should not be used to determine the rights and obligations of non-consenting non-signatories.

Fifth, the federal courts should make more extensive use of the doctrine of contra proferentem when construing ambiguous forum selection clauses. When a clause is ambiguous, it should be construed against the drafting party. There is nothing remotely controversial about this proposal. Indeed, a smattering of courts have already deployed it to interpret forum selection clauses.236 In light of the bargaining disparities that pervade the drafting of these agreements, a turn to contra proferentem is both appropriate and normatively desirable.

Sixth, and finally, the courts should decline to enforce “non-mutual” forum selection clauses that require one contracting party to sue in the chosen forum but allow the other party to sue whenever they want. What is sauce for the goose should be sauce for the gander. The courts should also decline to uphold “floating” clauses where the identity of the chosen forum could be changed after the contract was signed. If the purpose of these clauses is to promote certainty in dispute resolution, it is difficult to see how this end is furthered by leaving the identity of the chosen jurisdiction unknown until months or years after the contract is signed.

None of the foregoing proposals is revolutionary. A true revolution would require a federal statute or a decision by the U.S. Supreme Court. These proposals do, however, represent a meaningful improvement to the status quo. If adopted, they would allow courts to undertake a more nuanced and balanced analysis of when a forum selection clause is “contractually valid” as that term is used in Atlantic Marine.

Conclusion

In Atlantic Marine, the U.S. Supreme Court provided much-needed guidance to the lower federal courts as to the correct procedural framework for enforcing a forum selection clause.237 The Court commented in a footnote that this framework was applicable whether the clause in question was “contractually valid.”238 It did not, however, offer any guidance as to the meaning of this language.

This Article has sought to remedy this deficit by distilling the relevant doctrinal rules into a concise and readable account of how the lower courts should determine when a forum selection clause is, in fact, contractually valid. It also surveyed hundreds of recent federal cases to determine how these doctrinal rules are applied in practice. These data suggest that federal courts enforce forum selection clauses in the overwhelming majority of cases. A dizzying array of doctrinal innovations in this area have led to a world where these provisions are almost always given effect. Since most of these clauses are drafted and used by large corporations, the effect of this shift is to advantage the interests of these entities.

In an attempt to remedy this imbalance, the Article identified a number of pragmatic reforms that could help to rebalance the scales. First, the federal courts should deem forum selection clauses unenforceable when they are contrary to the statutory public policy of the state in which they sit. Second, they should adopt a broader view of what constitutes an unreasonable clause. Third, they should construe special venue provisions in federal statutes to preempt forum selection clauses. Fourth, they should refrain from applying the closely-related-and-foreseeable test to non-corporate persons, particularly in personal jurisdiction cases. Fifth, they should make more liberal use of the doctrine of contra proferentem when construing forum selection clauses. Finally, they courts should not enforce “non-mutual” forum selection clauses or “floating” forum selection clauses. None of these proposals is a panacea. Individually, each proposal will only move the needle so much. Collectively, however, they represent a meaningful improvement on the status quo and a first step in producing a fairer set of rules to determine whether a clause is contractually valid.    

Appendix

I collected the case data from the federal courts using the following method. First, I conducted a search in LexisAdvance under “Federal District Courts” in each state for the following terms: “choice of court clause” or “forum selection clause” or “choice of forum clause” or “consent to jurisdiction clause” or “venue selection clause.” I narrowed the timeline to the range between January 1, 2014, and December 31, 2020. I then reviewed the resulting hits for cases where: (1) the forum selection clause selected a court located in another U.S. state or a foreign country; (2) the forum selection clause was mandatory; (3) the forum selection clause was broad enough to cover the dispute; (4) the court was asked to transfer the case pursuant to 28 U.S.C. § 1404 or to dismiss the case pursuant to the federal doctrine of forum non conveniens; and (5) the court considered the possibility that the motion should not be granted because the clause was unenforceable. I then repeated the process for cases decided by each federal circuit court of appeal. When my review was complete, I had a dataset of 658 federal cases. That is the dataset analyzed in the Article.

In conducting my review, I excluded a number of cases even though they presented interesting issues relating to forum selection clauses. First, I excluded cases where one party was seeking to remand the case to a state court in the same state. Second, I excluded cases where one party was seeking to transfer the case to a different federal district in the same state where the forum court was located. Third, I excluded cases where the only issue before the court related to clause interpretation and the court did not consider the enforceability of the clause. Fourth, I excluded cases where the primary issue before the court related to an arbitration clause. Fifth, I excluded cases where the forum selection clause was not mandatory. Sixth, I excluded cases where the resisting party argued that the contract was invalid under traditional contract doctrine (e.g., lack of mutual assent). I included, however, cases where one party argued the clause was unenforceable due to fraud. Seventh, I excluded cases where the only issue before the court was whether a third party was bound by the forum selection clause. Eighth, I excluded cases where the case had already been transferred to the forum from somewhere else. Ninth, I excluded cases where the only issue before the court was whether one party had waived its right to invoke the forum selection clause. Tenth, I excluded cases where the only issue before the court was whether the clause conferred subject matter jurisdiction upon the forum court. Finally, I excluded cases where the only issue before the court was whether the party seeking to enforce the clause had sought enforcement via the correct procedural mechanism.

I recognize that there are problems with relying on cases resulting in a published or unpublished decision to empirically assess judicial behavior. Such decisions are not representative of all cases.239 A growing number of scholars have urged empiricists to look to court dockets—rather than judicial opinions—to obtain a more accurate measure of how judges behave.240 These concerns notwithstanding, there are two primary reasons why I adopted the methodological approach set forth above.241 First, while published and unpublished cases may not be “representative” in a statistical sense, they are “representative” in that they are for most scholars, judges, and lawyers the “full population . . . of the cases shaping perceptions of the legal system. Published opinions are all most of us ever work from.”242 Second, while a docket search can tell us the ultimate disposition of a particular case—was it dismissed, transferred, or retained—it can tell us nothing about the reasoning employed by the court to reach that decision. To the extent that I am seeking to measure the outcomes generated by a particular doctrinal test, the only way to meaningfully do so is by a review of published and unpublished decisions in which the courts applied this test. 

  1. [1]. Atl. Marine Constr. Co. v. U.S. Dist. Ct. for W. Dist. of Tex., 571 U.S. 49, 62–68 (2013).

  2. [2]. Id. at 55–68. This Article uses the term “forum selection clause” to refer to a contract provision that selects a court to resolve disputes between the parties. The term does not include arbitration clauses.

  3. [3]. Id.

  4. [4]. 28 U.S.C. § 1404(a) (2018).

  5. [5]. Atl. Marine, 571 U.S. at 60.

  6. [6]. See generally e.g., Gemini Techs., Inc. v. Smith & Wesson Corp., 931 F.3d 911 (9th Cir. 2019) (stating that the analytical framework set forth in Atlantic Marine only applies when a forum selection clause is “contractually valid”); In re Mathias, 867 F.3d 727 (7th Cir. 2017) (discussing the Atlantic Marine decision); Azima v. RAK Inv. Auth., 926 F.3d 870 (D.C. Cir. 2019) (grounding its decision in Atlantic Marine).

  7. [7]. Atl. Marine, 571 U.S. at 62 n.5. In a perfect world, the Court would have stated that its analysis presupposes a “contractually valid [and enforceable] forum selection clause” to more clearly indicate that validity and enforceability are separate inquiries. To date, however, all of the lower courts have construed the “contractually valid” language to encompass issues of enforceability.

  8. [8]. Linda S. Mullenix, Gaming the System: Protecting Consumers from Unconscionable Contractual Forum-Selection and Arbitration Clauses, 66 Hastings L.J. 719, 721 (2015) (“The Court’s unanimous decision, centering on § 1404(a) transfers, begs the primary question of which court should determine the validity of a forum-selection clause, subject to what law, and when.”); Stephen E. Sachs, Five Questions After Atlantic Marine, 66 Hastings L.J. 761, 766 (2015) (“Atlantic Marine places enormous weight on whether a forum-selection clause is valid and enforceable. If it is, enforcement is virtually automatic ... . Yet the opinion says nothing about which clauses are valid in the first place.”).

  9. [9]. See John F. Coyle, Interpreting Forum Selection Clauses, 104 Iowa L. Rev. 1791, 1835–50 (2019).

  10. [10]. Id.

  11. [11]. See Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 15 (1972) (holding forum-selection clauses are not enforceable when they are unreasonable or contrary to public policy); Carnival Cruise Lines v. Shute, 499 U.S. 585, 591 (1991) (same).

  12. [12]. Most scholars writing about Atlantic Marine have focused on the procedural questions that were the focus of the Court’s opinion. See generally, e.g., Robin Effron, Atlantic Marine and the Future of Forum Non Conveniens, 66 Hastings L.J. 693 (2015) (examining “why merging or conflating § 1404(a) and the forum non conveniens doctrine is problematic, both as a general matter and as applied to the specific context of forum-selection clauses”); Matthew J. Sorensen, Note, Enforcement of Forum Selection Clauses in Federal Court After Atlantic Marine, 82 Fordham L. Rev. 2521 (2014) (discussing “the substantive law governing the validity and enforceability of forum-selection clauses”); Bradley Scott Shannon, Enforcing Forum-Selection Clauses, 66 Hastings L.J. 777 (2015) (discussing problems with the enforcement of contractual forum-selection clauses). For one of the few articles that focus on the contractual validity of a forum selection clause after Atlantic Marine, see John M. Doroghazi & David J. Norman, What’s Left to Litigate About Forum Selection Clauses? Atlantic Marine Turns Four, 36 Franchise L.J. 581 (2017) (exploring the remaining problems after Atlantic Marine).

  13. [13]. See infra Part I.

  14. [14]. See infra Section I.B.

  15. [15]. See infra Section I.B.1.

  16. [16]. See infra Section I.B.2.

  17. [17]. See infra Section I.B.3.

  18. [18]. See infra Part II.

  19. [19]. See infra Section II.B.

  20. [20]. See infra Section II.C.

  21. [21]. See infra Section III.B.

  22. [22]. See infra Section III.C.

  23. [23]. Atl. Marine Constr. Co. v. U.S. Dist. Ct. for W. Dist. of Tex., 571 U.S. 49, 60–68 (2013).

  24. [24]. Azima v. RAK Inv. Auth., 926 F.3d 870, 874 (D.C. Cir. 2019) (“As long as the forum-selection clause is applicable, mandatory, valid, and enforceable, the court must almost always grant the motion to dismiss.”). There is currently some disagreement among the federal courts as to whether it is appropriate to inquire as to the availability and adequacy of a foreign forum when one party invokes a forum selection clause and seeks to dismiss on the basis forum non conveniens. See In re Fortinet, Inc., 803 F. App’x 409, 410–11 (Fed. Cir. 2020) (mem.) (collecting cases). This issue of whether a foreign forum is available and adequate is not discussed in this Article.

  25. [25]. Atl. Marine, 571 U.S. at 60–68. This statement does not mean that extraordinary circumstances must be shown to prove that a clause is invalid or unenforceable. It simply makes clear that if a clause is valid, applicable, mandatory, and enforceable, it should then be given effect by transferring or dismissing the case unless there are extraordinary circumstances that dictate a different result.

  26. [26]. Id. As a matter of terminology, the Article refers to the two-part test set forth in Atlantic Marine as the “analytical framework” for evaluating whether to grant a motion to transfer or dismiss when a contract contains a forum selection clause. The first step in this framework requires the court to: (1) determine whether the clause is valid as a matter of contract law;
    (2) interpret the clause to see if it is exclusive and applies to the claims at issue; and (3) determine whether the clause is enforceable. If a clause is deemed “contractually valid” at step one of the analytical framework, the second step requires the court to apply the relevant public interest factors under Section 1404 (or forum non conveniens) to determine whether transfer or dismissal is warranted. Id.

  27. [27]. See Restatement (Second) of Conts. §§ 17–81 (Am. L. Inst. 1981).

  28. [28]. Id. §§ 110–77.

  29. [29]. See e.g., Barnett v. DynCorp Int’l, LLC, 831 F.3d 296, 301–02 (5th Cir. 2016).

  30. [30]. Id. at 304 (“If, instead, the issue of a forum-selection clause’s ‘validity’ is separate from its ‘enforceability’ and not determined by federal law in diversity cases, it seems that the law applicable to that determination would be the same law applicable to forum-selection clause interpretation—that is, the law selected by the forum state’s choice-of-law rules.” (citing Weber v. PACT XPP Techs., AG, 811 F.3d 758, 770–71 (5th Cir. 2016))); Kelley v. MailFinance Inc., 436 F. Supp. 3d 1136, 1141 (N.D. Ill. 2020) (“[D]etermining whether a forum-selection clause is part of a contract is a question of state law.” (citing Bourke v. Dun & Bradstreet Corp., 159 F.3d 1032, 1036 (7th Cir. 1998))); Sorensen, supra note 12, at 2553 (“Courts sitting in diversity should apply substantive state law in determining the answer to the validity question.”); Jason Webb Yackee, Choice of Law Considerations in the Validity & Enforcement of International Forum Selection Agreements: Whose Law Applies?, 9 UCLA J. Int’l L. & Foreign Affs. 43, 46, 84–88 (2004) (arguing that the parties’ chosen law should govern the issue of validity).

  31. [31]. See 14D Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure Jurisdiction and Related Matters § 3803.1 (4th ed. 2022).

  32. [32]. Symeon C. Symeonides, What Law Governs Forum Selection Clauses, 78 La. L. Rev. 1119, 1145–60 (2018).

  33. [33]. See Restatement (Second) of Conflict of L. §§ 198–202 (Am. L. Inst. 1971) (directing courts to apply the rules in Section 188 to determine the governing law for contractual capacity, the requirements of a writing, validity, misrepresentation, duress, undue influence, mistake, and illegality).

  34. [34]. See Avtech Cap., LLC v. C & G Engines Corp., No. 19-cv-00541, 2020 WL 6701880, at *4 (D. Utah Nov. 13, 2020); St. Aubin v. Island Hotel Co. Ltd., No. 16-cv-22023, 2017 WL 998298, at *2–3 (S.D. Fla. Mar. 15, 2017).

  35. [35]. See Wainess v. Smilemakers, Inc., No. 18-12177, 2018 WL 6809654, at *3–4 (E.D. Mich. Dec. 27, 2018); Am. Water Heater Co. v. Taylor-Winfield Techs., Inc., No. 16-cv-125, 2017 WL 4293228, at *6 (E.D. Tenn. Sept. 27, 2017); Duro Textiles, LLC v. Sunbelt Corp., 12 F. Supp. 3d 221, 224 (D. Mass. 2014).

  36. [36]. See Restatement (Second) of Conts. § 163 (Am. L. Inst. 1981).

  37. [37]. John F. Coyle & Katherine C. Richardson, Enforcing Outbound Forum Selection Clauses in State Court, 96 Ind. L.J. 1089, 1144 (2021). This rule derives from the notion that a forum selection clause is a contract between the parties that is separate and distinct from the underlying agreement. This notion is sometimes referred to as the “separability doctrine.” See Intercall Telecomms., Inc. v. Instant Impact, Inc., 376 F. Supp. 2d 155, 160 (D.P.R. 2005). In practice, relatively few courts expressly refer to the separability doctrine when deciding cases. They simply apply the special rule relating to fraud.

  38. [38]. Scherk v. Alberto-Culver Co., 417 U.S. 506, 512–21 (1974); see also Prima Paint Corp. v. Flood & Conklin Mfg. Co., 388 U.S. 395, 401–04 (1967) (“We hold, therefore, that in passing upon a § 3 application for a stay while the parties arbitrate, a federal court may consider only issues relating to the making and performance of the agreement to arbitrate.”).

  39. [39]. Scherk, 417 U.S. at 519­–21.

  40. [40]. Id. at 512–21.

  41. [41]. Id.

  42. [42]. See Moses v. Bus. Card Express, Inc., 929 F.2d 1131, 1138 (6th Cir. 1991); PLAYA/USA Corp. v. PLAYASOL, S.A., No. 80 C 1020, 1981 U.S. Dist. LEXIS 14178, at *3 (N.D. Ill. Aug. 20, 1981).

  43. [43]. See, e.g., Niemi v. Lasshofer, 770 F.3d 1331, 1352 (10th Cir. 2014).

  44. [44]. See, e.g., Brown v. Artec Glob. Media, Inc., No. 16cv2651, 2017 WL 11596885, at *4–7 (S.D. Cal. July 5, 2017); Bulldog Well Testing, LLC v. Eldorado Energy Rentals, LLC, No. civ-16-004, 2016 WL 9558950, at *1–2 (E.D. Okla. Mar. 7, 2016); Armstrong v. Curves Int’l, Inc., No. 15cv1006, 2015 WL 6085553, at *14 (E.D. Mo. Oct. 15, 2015).

  45. [45]. Ramsdell v. Bowles, 64 F.3d 5, 10 (1st Cir. 1995) (quoting F.O. Bailey Co. v. Ledgewood, Inc., 603 A.2d 466, 468 (Me. 1992)). Section 302 of the Restatement (Second) of Contracts states that “a beneficiary of a promise is an intended beneficiary if recognition of a right to performance in the beneficiary is appropriate to effectuate the intention of the parties and
    . . . the circumstances indicate that the promisee intends to give the beneficiary the benefit of the promised performance.” Restatement (Second) of Conts. § 302 (Am. L. Inst. 1981). This rule requires a showing that the contracting parties intended to confer a benefit upon the non-signatory. The mere fact that a person derives an actual benefit from an agreement to which she is not a party is not usually enough to confer third-party beneficiary status.

  46. [46]. See Bruckner Truck Sales, Inc. v. Hoist Liftruck Mfg., LLC, 501 F. Supp. 3d 409, 426 (N.D. Tex. 2020).

  47. [47]. See generally Coyle & Richardson, supra note 37 (presenting statistics and analysis on the enforcement of forum selection clauses in state court); Shannon, supra note 12 (discussing the problems with enforcing forum selection clauses).

  48. [48]. John F. Coyle & Robin J. Effron, Forum Selection Clauses, Non-Signatories, and Personal Jurisdiction, 97 Notre Dame L. Rev. 187, 200–12 (2021).

  49. [49]. Id.

  50. [50]. Id. at 198–205.

  51. [51]. Id.

  52. [52]. Id.

  53. [53]. This rule may be modified by the parties. If the contract contains a no-third-party-beneficiaries clause, for example, affiliates and subsidiaries will not be covered by the forum selection clause even if they otherwise satisfy the requirements of the closely-related-and-foreseeable test. See, e.g., Cooper v. Meridian Yachts, Ltd., 575 F.3d 1151, 1169–70 (11th Cir. 2009); Casville Invs., Ltd. v. Kates, No. 12 Civ. 6968, 2013 WL 3465816, at *5–6 (S.D.N.Y. July 8, 2013); Bensinger v. Denbury Res. Inc., No. 10-cv-1917, 2011 WL 3648277, at *5–6 (E.D.N.Y. Aug. 17, 2011); APA Excelsior III L.P. v. Premiere Techs., Inc., 49 F. Supp. 2d 664, 671–73 (S.D.N.Y. 1999). In the absence of such a clause, however, the test will generally serve to expand the range of actors entitled to invoke the forum selection clause beyond the usual cast of third-party beneficiaries.

  54. [54]. Timothy Murray, 13 Corbin on Contracts § 67.2 (2022).

  55. [55]. Id. (“[N]either termination nor cancellation affect those terms that relate to the settlement of disputes or choice of law or forum selection clauses.”). This rule derives from the notion that a forum selection clause is a contract between the parties that is separate and distinct from the underlying agreement. See Karon v. Elliott Aviation, 937 N.W.2d 334, 353 (Iowa 2020) (Appel, J., dissenting) (criticizing doctrine of separability as applied to forum selection clauses).

  56. [56]. U.S. Smoke & Fire Curtain, LLC v. Bradley Lomas Electrolok, Ltd., 612 F. App’x 671, 672-73 (6th Cir. 2015) (per curiam) (noting the “greater weight of authority” holds that “dispute-resolution provisions, such as forum-selection clauses, are enforceable beyond the expiration of the contract”); Sensify (US) Inc. v. Intelligent Telematics N. Am., Inc., No. 16-24069-Civ, 2017 WL 1541424, at *2–3 (S.D. Fla. Apr. 28, 2017) (“[C]ourts consistently have rejected the notion that termination of an agreement necessarily extinguishes its forum-selection clause.”); VERSAR, Inc. v. Ball, No. Civ. A. 01-1302, 2001 WL 818354, at *2 (E.D. Pa. July 12, 2001) (“Unless otherwise expressed, a choice of forum clause does not expire upon termination of the contract from which it derives.”); Granite Re, Inc. v. Hutton, Inc., No. 19-cv-3074, 2020 WL 4735309, at *3-4 (D. Minn. Aug. 14, 2020) (“Although ‘“[n]ormally, the parties’ obligations expire when a contract terminates, ... provisions related to the manner in which disputes are resolved generally survive the contract’s termination’ and continue to apply to disputes that ‘accrued before the agreement’s termination.’” (quoting Serv. Team of Pros., Inc. v. Folks, No. 18-cv-0048, 2018 WL 2051516, at *2 (W.D. Mo. May 2, 2018))); see also Cottman Ave. PRP Grp. v. AMEC Foster Wheeler Env’t Infrastructure Inc., 439 F. Supp. 3d 407, 434–38 (E.D. Pa. 2020) (concluding that forum selection clauses survive the termination of the contract, even if they are not listed among the terms in the survival clause).

  57. [57]. The Part does not discuss whether an ambiguous forum selection clause should be interpreted to select the state and federal courts or whether it should be read to select the state courts to the exclusion of the federal courts. This interpretive issue is analyzed at length elsewhere. See Coyle, supra note 9, at 1826-30 (2019).

  58. [58]. See e.g., Collins v. Mary Kay, Inc., 874 F.3d 176, 182-83 (3d Cir. 2017) (collecting cases).

  59. [59]. See Manetti-Farrow, Inc. v. Gucci Am., Inc., 858 F.2d 509, 513 (9th Cir. 1988) (“[B]ecause enforcement of a forum clause necessarily entails interpretation of the clause before it can be enforced, federal law also applies to interpretation of forum selection clauses.”); see also Fouad v. Qatar, 846 F. App’x 466, 469 (9th Cir. 2021) (“Federal law applies to interpreting a forum selection clause.”); Doe 1 v. AOL LLC, 552 F.3d 1077, 1081 (9th Cir. 2009) (“We apply federal law to the interpretation of the forum selection clause.”).

  60. [60]. See Collins, 874 F.3d at 182-83.

  61. [61]. See supra Section I.A; see also Martinez v. Bloomberg LP, 740 F.3d 211, 217-18 (2d Cir. 2014) (observing that when a contract contains a choice-of-law clause, the interpretation of the forum-selection clause is governed by the law chosen by the parties in the choice-of-law clause); Yavuz v. 61 MM, Ltd., 465 F.3d 418, 427–28 (10th Cir. 2006) (“[W]hen a court interprets a contract, as a general matter it applies the law that the parties selected in their contract . . . . A forum-selection clause is part of the contract. We see no particular reason ... why a forum-selection clause, among the multitude of provisions in a contract, should be singled out as a provision not to be interpreted in accordance with the law chosen by the contracting parties.”); Barnett v. DynCorp Int’l, LLC, 831 F.3d 296, 304–08 (5th Cir. 2016) (“If, instead, the issue of a forum-selection clause’s ‘validity’ is separate from its ‘enforceability’ and not determined by federal law in diversity cases, it seems that the law applicable to that determination would be the same law applicable to forum-selection clause interpretation—that is, the law selected by the forum state’s choice-of-law rules.”).

  62. [62]. See Coyle, supra note 9, at 1799-803 (2019).

  63. [63]. Id. at 1800-01.

  64. [64]. Id. at 1802-03.

  65. [65]. Id. at 1802.

  66. [66]. Id.

  67. [67]. Id. at 1800.

  68. [68]. Id.

  69. [69]. Id.

  70. [70]. Id. at 1802.

  71. [71]. Id.

  72. [72]. Id. at 1803–20.

  73. [73]. Id. at 1804.

  74. [74]. Id. at 1805­–08.

  75. [75]. Id. at 1807–10.

  76. [76]. Id. at 1808.

  77. [77]. Id. at 1794–810.

  78. [78]. Id.

  79. [79]. Id. at 1809–12 (quoting Hansa Consult of N. Am., LLC v. Hansaconsult Ingenieurgesellschaft mbH, 35 A.3d 587, 595 (N.H. 2011)).

  80. [80]. Id. at 1810–11.

  81. [81]. Id. at 1812.

  82. [82]. Id. at 1812–18.

  83. [83]. Id.

  84. [84]. Id.

  85. [85]. Id.

  86. [86]. Id. at 1818–19.

  87. [87]. Id. at 1818–23 (citing Terra Int’l, Inc. v. Miss. Chem. Corp., 119F.3d 688, 693 (8th Cir. 1997)).

  88. [88]. Id.

  89. [89]. See David Marcus, The Perils of Contract Procedure: A Revised History of Forum Selection Clauses in the Federal Courts, 82 Tul. L. Rev. 973, 975–80 (2008) (noting that there was some limited enforcement of forum selection clauses in the admiralty context prior to 1972).

  90. [90]. See id.

  91. [91]. Nute v. Hamilton Mut. Ins. Co., 72 Mass. 174, 184 (1856).

  92. [92]. See Marcus, supra note 89, at 975–80.

  93. [93]. Ins. Co. v. Morse, 87 U.S. 445, 451 (1874).

  94. [94]. In a prior paper, I reviewed twenty-five contract form books published between 1860 and 2019 to see how many contained choice-of-law clauses. See John F. Coyle, A Short History of the Choice-of-Law Clause, 91 U. Colo. L. Rev. 1147, 1171 (2020). In the course of this review, I also found that virtually none of these form books published prior to 1950 contained forum selection clauses.

  95. [95]. Model Choice of Forum Act § 3 (Unif. L. Comm’n 1968). See also generally Willis L. M. Reese, The Model Choice of Forum Act, 17 Am. J. Compar. L. 292 (1969) (describing the statute).

  96. [96]. Model Choice of Forum Act §§ 2–3.

  97. [97]. Restatement (Second) of Conflict of L. § 80 cmt. a (Am. L. Inst. 1971).

  98. [98]. See generally Bremen v. Zapata Off-Shore Co., 407 U.S. 1 (1972) (holding forum-selection clauses are presumptively enforceable unless contrary to public policy or unreasonable).

  99. [99]. Id. at 9–12, 18–20.

  100. [100]. Id.

  101. [101]. Id. at 17–19.

  102. [102]. Id. at 16–20.

  103. [103]. Id. at 15–17.

  104. [104]. Id. at 18–20.

  105. [105]. See generally Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22 (1988) (holding that federal law applies in determining if a forum selection clause is enforceable).

  106. [106]. Id. at 24–26.

  107. [107]. Id.

  108. [108]. Id.

  109. [109]. Id.

  110. [110]. Id.

  111. [111]. Stewart Org., Inc. v. Ricoh Corp., 810 F.2d 1066, 1067 (11th Cir. 1987) (“The threshold question is whether federal or state law governs the enforceability of a forum selection clause.”).

  112. [112]. Id. at 1070–71.

  113. [113]. Stewart, 487 U.S. at 28–31.

  114. [114]. Id.

  115. [115]. Id. at 31–33.

  116. [116]. See Wright & Miller supra note 31, at § 3803.1 (“A question of obvious importance in diversity of citizenship cases is what law—state or federal—shall govern on the question of whether a forum selection clause is enforceable. It seems rather clear that federal law should govern.”) (footnotes omitted). A number of scholars have argued that, in fact, the threshold determination of whether a clause is contractually valid should be governed by state law. See, e.g., Kermit Roosevelt III & Bethan R. Jones, Adrift on Erie: Characterizing Forum-Selection Clauses, 52 Akron L. Rev. 297, 317 (2018) (“Whether the clause confers rights under state law is a question of state substantive law, to be decided under the state law that governs the contract. The effect of those rights in federal court is a question of federal procedural law. As with a § 1404(a) motion, a federal court might decide that a valid forum-selection clause does not justify dismissal, and it might decide that an invalid clause does.”); Adam N. Steinman, Atlantic Marine Through the Lens of Erie, 66 Hastings L.J. 795, 797 (2015) (“Atlantic Marine does not mandate unflinching enforcement of forum-selection clauses without any mechanism for parties to raise legitimate concerns about the use and operation of such clauses in certain contexts. Rather, Atlantic Marine should be read to defer to state law on such matters.”). In his dissent in Stewart, Justice Scalia also argued that the enforceability issue should be governed by state law. Stewart, 487 U.S. at 33–41 (Scalia, J., dissenting).

  117. [117]. Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 15 (1972).

  118. [118]. As a rule, the federal courts focus exclusively on statutes enacted by the forum in considering whether to apply the public policy exception. See Willis re Inc. v. Herriott, 550 F. Supp. 3d 68, 97 (S.D.N.Y. 2021) (“[F]ederal courts sitting in diversity in New York may refuse to enforce a contractual forum-selection provision if enforcing it would contravene New York’s public policy; the public policy of California or any other state is irrelevant.”). They do not—and should not—consider the public policy of any other state.

  119. [119]. 45 U.S.C. § 56.

  120. [120]. 49 U.S.C. § 14706(d)(2).

  121. [121]. 40 U.S.C. § 3133.

  122. [122]. See Boyd v. Grand Trunk W. R.R. Co., 338 U.S. 263, 265 (1949) (“[C]ontracts limiting the choice of venue are void as conflicting with the Liability Act.”).

  123. [123]. See Scotlynn USA Div., Inc. v. Singh, No. 215-cv-381-FTM-29, 2016 WL 8679295, at *2 (M.D. Fla. Sept. 9, 2016) (“Courts that have addressed venue in the context of household goods have determined that ‘the Carmack Amendment essentially prohibits enforcement of forum-selection clauses and provides that suit may be brought against a carrier in a forum convenient to the shipper.’” (quoting Stewart v. Am. Van Lines, No. 12CV394, 2014 WL 243509, at *4 (E.D. Tex. Jan. 21, 2014))); Dabecca Nat. Foods, Inc. v. RD Trucking, LLC, No. 14 C 6100, 2015 WL 2444505, at *6 (N.D. Ill. May 20, 2015) (collecting cases).

  124. [124]. See In re Mathias, 867 F.3d 727, 730–34 (7th Cir. 2017); Smith v. Aegon Cos. Pension Plan, 769 F.3d 922, 931–34 (6th Cir. 2014), cert. denied, 577 U.S. 1061 (2016). But see Dumont v. PepsiCo, Inc., 192 F. Supp. 3d 209, 223 (D. Me. 2016) (“Because ERISA has a strong public policy in favor of ready access to federal courts, enforcement of the forum selection clause against Dumont would be unreasonable.”). See generally Christine P. Bartholomew & James A. Wooten, The Venue Shuffle: Forum Selection Clauses and ERISA, 66 UCLA L. Rev. 862 (2019) (arguing that forum selection clauses should not be enforced in ERISA cases).

  125. [125]. See United States ex rel. River Front Recycling & Aggregate, LLC v. Kallidus Techs., Inc., No. CV 18-9141, 2019 WL 2082948, at *6–7 (D.N.J. May 13, 2019) (discussing the Miller Act); In re Marquette Transp. Co. Gulf-Inland LLC, No. 18-cv-00074, 2018 WL 4443141, at *4–5 (S.D. Tex. Sept. 4, 2018), report and recommendation adopted, No. 18-cv-00074, 2018 WL 4408937 (S.D. Tex. Sept. 17, 2018) (discussing the Jones Act).

  126. [126]. Compare Luderus v. United States Helicopters, Inc., No. 12-cv-5094, 2013 WL 677814, at *5 (N.D. Ill. Feb. 25, 2013) (clause is unenforceable); Smith v. Kyphon, Inc., 578 F. Supp. 2d 954, 961 (M.D. Tenn. 2008) (same); Thomas v. Rehab. Servs. of Columbus, Inc., 45 F. Supp. 2d 1375, 1381 (M.D. Ga. 1999) (same), with DeBello v. VolumeCocomo Apparel, Inc., 720 F. App’x 37, 41 (2d Cir. 2017) (clause is enforceable on facts presented); Kessler v. Direct Consulting Assocs. LLC, No. 17-11943, 2018 WL 7890862, at *9 (E.D. Mich. July 6, 2018) (same).

  127. [127]. In re Marquette Transp., 2018 WL 4443141, at *4 (collecting cases); see also In re OSG Ship Mgmt., 514 S.W.3d 331, 340–43 (Tex. App. 2016) (discussing the relevance of Boyd to Jones Act cases).

  128. [128]. See e.g., Martinez v. Bloomberg LP, 740 F.3d 211, 228–29 (2d Cir. 2014).

  129. [129]. 15 U.S.C. § 77n (“Any condition, stipulation, or provision binding any person acquiring any security to waive compliance with any provision of this subchapter or of the rules and regulations of the Commission shall be void.”); id. § 78cc(a) (“Any condition, stipulation, or provision binding any person to waive compliance with any provision of this chapter or of any rule or regulation thereunder, or of any rule of a self-regulatory organization, shall be void.”).

  130. [130]. See John F. Coyle, Cruise Contracts, Public Policy, and Foreign Forum Selection Clauses, 75 U. Miami L. Rev. 1087, 1098-99 (2021); cf. Coyle & Richardson, supra note 37, at 1198-200 (discussing empirical analysis of forum selection clauses). The Seventh Circuit recently invoked the anti-waiver provisions in the federal securities acts in refusing to enforce a forum selection bylaw requiring federal claims to be brought in the Delaware Court of Chancery. See Seafarers Pension Plan ex rel. Boeing Co. v. Bradway, 23 F.4th 714, 724–28 (7th Cir. 2022). In a subsequent case, the Ninth Circuit held that these same anti-waiver provisions did not foreclose the enforcement of a different forum selection bylaw selecting the same court. See Lee v. Fisher, 34 F.4th 777, 781–82 (9th Cir. 2022).

  131. [131]. See Darrell Hall, Note, No Way Out: An Argument Against Permitting Parties to Opt Out of U.S. Securities Laws in International Transactions, 97 Colum. L. Rev. 57, 81–84 (1997) (describing how different courts have approached this issue).

  132. [132]. See Lipcon v. Underwriters at Lloyd’s, 148 F.3d 1285, 1297–98 (11th Cir. 1998); Haynsworth v. Corp., 121 F.3d 956, 960–62 (5th Cir. 1997); Allen v. Lloyd’s of London, 94 F.3d 923, 927–29 (4th Cir. 1996); Shell v. R.W. Sturge, Ltd., 55 F.3d 1227, 1229–30 (6th Cir. 1995); Bonny v. Soc’y of Lloyd’s, 3 F.3d 156, 158–60 (7th Cir. 1993); Roby v. Corp. of Lloyd’s, 996 F.2d 1353, 1363–65 (2d Cir. 1993); Riley v. Kingsley Underwriting Agencies, Ltd., 969 F.2d 953, 958 (10th Cir. 1992).

  133. [133]. See supra notes 115-31 and accompanying text.

  134. [134]. Id. The federal courts have also invoked the logic of anti-waiver in weighing whether to enforce foreign forum selection clauses in cases arising under federal civil rights laws and the Carriage of Goods by Sea Act (COGSA). See Coyle, supra note 130, at 1100–04.

  135. [135]. See Coyle, supra note 130, at 1100–04.

  136. [136]. See id.

  137. [137]. Jonathan C. Lipson, Fighting Fiction with Fiction—The New Federalism in (a Tobacco Company) Bankruptcy, 78 Wash. U. L.Q. 1271, 1279–80 (2000) (“The Bankruptcy Code seeks to channel all claims against a debtor through a single foruma federal bankruptcy courtand to provide a single, effective mechanism through which such claims are treated.”).

  138. [138]. See id.

  139. [139]. See e.g., Walker v. Got’cha Towing & Recovery, LLC (In re Walker), 551 B.R. 679, 690–91 (Bankr. M.D. Ga. 2016) (“Because the bankruptcy system implicates interests far broader than the private rights of the two parties to the contract in question, it is not unusual for prepetition contractual obligations, particularly those dictating forum or waiving the protections of the automatic stay, to be modified or even ignored in a bankruptcy case.”) (footnotes omitted); In re John Q. Hammons Fall 2006, LLC, No. 16-21142, 2017 WL 4620872, at *8 (Bankr. D. Kan. Oct. 13, 2017) (“The Court is persuaded that in this case the strong public policy of centralized resolution of claims supports keeping the matter in bankruptcy court and not enforcing the forum selection clause in the rejected ROFR.”); Harpole Constr., Inc. v. Medallion Midstream, LLC (In re Harpole Constr., Inc.), No. 15-12630, 2016 WL 7373780, at *6 (Bankr. D.N.M. Nov. 23, 2016) (“In accordance with the case law cited above, and after considering the circumstances of this case, the Court concludes that the interest in centralizing all of Harpole’s disputes [in the bankruptcy court] outweighs the policy of enforcing the forum selection clause.”); see also Kismet Acquisition, LLC v. Icenhower (In re Icenhower), 757 F.3d 1044, 1051 (9th Cir. 2014) (concluding “[t]he bankruptcy court properly declined to enforce the forum selection clauses” because one of the Code’s “primary objectives is ‘centralization of disputes concerning a debtor’s legal obligations’” (quoting In re Eber,687 F.3d 1123, 1131 (9th Cir. 2012))); Haigler v. Dozier (In re Dozier Fin., Inc.), 587 B.R. 637, 650–51 (Bankr. D.S.C. 2018) (declining to enforce an FSC in a bankruptcy case); Alsohaibi v. Arcapita Bank B.S.C.(c) (In re Arcapita Bank B.S.C.(c)), 508 B.R. 814, 820 (S.D.N.Y. 2014) (finding “[a] debtor-in-possession ... is not bound by a forum selection clause in an agreement provided the litigation at issue amounts to a core proceeding and is not inextricably intertwined with non-core matters.” (quoting Statutory Comm. of Unsecured Creditors v. Motorola, Inc. (In re Iridium Operating LLC),285 B.R. 822, 837 (S.D.N.Y. 2002))); Argosy Cap. Grp. III, L.P. v. Triangle Cap. Corp., No. 17 Civ. 9845, 2019 WL 140730, at *7–8 (S.D.N.Y. Jan. 9, 2019) (finding the forum selection clause was not enforceable).

  140. [140]. Coyle & Richardson, supra note 37, at 1234–40.

  141. [141]. Id.

  142. [142]. Id.

  143. [143]. Idaho Code § 29-110(1) (2022).

  144. [144]. This happens a lot. See Davis v. Oasis Legal Fin. Operating Co., LLC, 936 F.3d 1174, 1179–81 (11th Cir. 2019); Rob & Bud’s Pizza, LLC v. Papa Murphy’s Int’l, Inc., No. 15-cv-5090, 2015 WL 3901611, at *2 (W.D. Ark. June 24, 2015); Advanced Int’l Mktg., LLC v. LXR Biotech, LLC, No. 17-cv-05086, 2017 WL 4780628, at *3 (W.D. Ark. Oct. 23, 2017); Weber v. Saladworks, LLC, No. SA CV 13-01049, 2014 WL 12581768, at *6 (C.D. Cal. Jan. 27, 2014); Frango Grille USA, Inc. v. Pepe’s Franchising Ltd., No. CV 14-2086, 2014 WL 7892164, at *3 (C.D. Cal. July 21, 2014); Devore v. H&R Block Tax Servs., LLC, No. CV 16-946, 2016 WL 11520709, at *2 (C.D. Cal. Mar. 10, 2016); Friedman v. Glob. Payments Inc., No. CV 18-3038, 2019 WL 1718690, at *3 (C.D. Cal. Feb. 5, 2019); Maced. Distrib., Inc. v. S-L Distrib. Co., LLC, No. SACV 17-1692, 2018 WL 6190592, at *3 (C.D. Cal. Aug. 7, 2018); Alabsi v. Savoya, LLC, No. 18-cv-06510, 2019 WL 1332191, at *7 (N.D. Cal. Mar. 25, 2019); Lyon v. Neustar, Inc., No. 19-cv-00371, 2019 WL 1978802, at *8 (E.D. Cal. May 3, 2019); Karl v. Zimmer Biomet Holdings, Inc., No. C 18-04176, 2019 WL 2775567, at *5 (N.D. Cal. July 2, 2019); Yeomans v. World Fin. Grp. Ins. Agency, Inc., No. 19-cv-00792, 2019 WL 5789273, at *7 (N.D. Cal. Nov. 6, 2019); Miller-Garcia v. Avani Media, LLC, No. 19-cv-04130, 2020 WL 95635, at *4 (N.D. Cal. Jan. 8, 2020); Bell v. L.P. Brown Co., No. CV 14-02772, 2015 WL 429973, at *6 (W.D. La. Feb. 2, 2015); Waguespack v. Medtronic, Inc., 185 F. Supp. 3d 916, 925 (M.D. La. 2016); Town of Jonesboro v. Pittsburg Tank & Tower Maint. Co., No. 17-1589, 2018 WL 3199476, at *3 (W.D. La. Feb. 12, 2018); Swank Enters., Inc. v. NGM Ins. Co., No. CV 19-200, 2020 WL 1139607, at *5 (D. Mont. Mar. 9, 2020); Crest Furniture, Inc. v. Ashley Homestores, Ltd., No. 20-cv-01383, 2020 WL 6375808, at *7 (D.N.J. Oct. 30, 2020); Fam. Wireless #1, LLC v. Auto. Techs., Inc., No. 15-11215, 2015 WL 5142350, at *5 (E.D. Mich. Sept. 1, 2015); Live Cryo, LLC v. CryoUSA Imp. & Sales, LLC, No. 17-cv-11888, 2017 WL 4098853, at *5 (E.D. Mich. Sept. 15, 2017); J. Lilly, LLC v. Clearspan Fabric Structures, Int’l, Inc., No. 18-cv-01104, 2018 WL 4773545, at *3 (D. Or. Oct. 2, 2018); Jorgenson Forge Corp. v. Illinois Union Ins. Co., No. 13-cv-01458, 2014 WL 12103362, at *3 (W.D. Wash. June 17, 2014).

  145. [145]. Gemini Techs., Inc. v. Smith & Wesson Corp., 931 F.3d 911, 915–17 (9th Cir. 2019); see also Davis, 936 F.3d at 1177 (applying state statute to invalidate outbound forum selection clause); DePuy Synthes Sales, Inc. v. Howmedica Osteonics Corp., 28 F.4th 956, 967 (9th Cir. 2022) (same).

  146. [146]. Gemini, 931 F.3d at 913–15.

  147. [147]. Id.

  148. [148]. Id.

  149. [149]. Id.

  150. [150]. Id.

  151. [151]. Id. at 915–17.

  152. [152]. Id. at 916 (citations omitted).

  153. [153]. Id. at 916–17.

  154. [154]. Id. Some federal courts have declined to follow this approach. See e.g., Bowen Eng’g Corp. v. Pac. Indem. Co., 83 F. Supp. 3d 1185, 1191–93 (D. Kan. 2015) (collecting cases). These decisions generally fail to account, however, for important differences between Stewart and Atlantic Marine. See Pierman v. Stryker Corp., No. 19-cv-00679, 2020 WL 406679, at *5 (S.D. Cal. Jan. 24, 2020); Steinman, supra note 116, at 796 (“Properly understood, Atlantic Marine opens the door for state law to play a more significant role than many anticipated in the wake of Stewart.”). The Supreme Court held in Stewart, for example, that state common law was not dispositive when the defendant sought to transfer a case pursuant to Section 1404(a). Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 28–32 (1988). That case said nothing, however, about the role that state law should play when the defendant moves to dismiss on the basis of forum non conveniens, as was the case in Gemini. Harding Materials, Inc. v. Reliable Asphalt Prods., Inc., No. 16-cv-02681, 2017 WL 495787, at *4 (S.D. Ind. Feb. 6, 2017). Other federal courts have distinguished Stewart in other ways. See Kirkland v. Deluxe Small Bus. Sales, Inc., No. CV 16-73, 2016 WL 9402787, at *6 (M.D. La. July 27, 2016) (“However, neither Stewart nor Atlantic Marine considered the threshold issue of whether the Forum Selection Clause is valid. That analysis, as set forth above, is dictated in this circuit by Haynesworth and requires consideration of whether enforcement would contravene a strong public policy of the forum state.” (citing Haynsworth v. Corp., 121 F.3d 956, 962 (5th Cir. 1997)); Black Hills Truck & Trailer, Inc. v. Mac Trailer Mfg., Inc., 2014 WL 5782452, at *15 (D.S.D. Nov. 6, 2014) (“Stewart makes clear that federal law, not state law, applies to a motion to transfer under 1404(a). But in situations where the court should apply Bremen, Stewart does not alter that approach. Therefore, Stewart does not undermine the applicability of Bremen ... to the determination of the enforceability of a forum-selection clause.”); Nat’l Frozen Foods Corp. v. Berkley Assurance Co., No. C17-339 RSM, 2017 WL 3781706, at *9 (W.D. Wash. Aug. 31, 2017) (“Stewart is likewise unhelpful because it dealt with ‘Alabama’s putative policy regarding forum-selection clauses,’ not a state law making a forum selection clause void.”) (citation omitted)).

  155. [155]. Presidential Hosp., LLC v. Wyndham Hotel Grp., LLC, 333 F. Supp. 3d 1179, 1222 (D.N.M. 2018) (“[T]he Court concludes that, under [Stewart], the Court cannot properly consider state statutes voiding forum selection clauses when a party moves for a 28 U.S.C. § 1404(a) transfer.”).

  156. [156]. This weighing of factors also sometimes occurs as part of the broader 1404 analysis rather than the enforcement analysis. See Redmond v. Sirius Int’l Ins. Corp., No. 12-cv-587, 2014 WL 197909, at *4 (E.D. Wis. Jan. 15, 2014) (“The court also recognizes that, although it is unenforceable under Wisconsin law, the fact that the parties agreed to a forum selection may be given some weight in the analysis under § 1404(a). However, the fact of the parties’ agreement is counterbalanced by Wisconsin’s strong public policy against forum selection clauses in insurance contracts; thus, the interests of justice lead to the conclusion that this fact merits negligible weight.” (citations omitted)); Ha Thi Le v. Lease Fin. Grp., LLC, No. CV 16-14867, 2017 WL 2915488, at *5 (E.D. La. May 8, 2017) (“While the Court recognizes that it still has the power to transfer to New York on the basis of the other factors identified in 28 U.S.C. § 1404(a), the Court declines to do so.” (citation omitted)).

  157. [157]. Gita Sports Ltd. v. SG Sensortechnik GmbH & Co. KG, 560 F. Supp. 2d 432, 434–35 (W.D.N.C. 2008).

  158. [158]. Id. at 434–42.

  159. [159]. Id. at 436–37, 440–41.

  160. [160]. Id. at 437 (quoting Allen v. Lloyd’s of London,94 F.3d 923, 928 (4th Cir. 1996)).

  161. [161]. Id. at 436–42.

  162. [162]. Id. at 438–43.

  163. [163]. Id.; see also Brand Energy Servs., LLC v. Enerfab Power & Indus., Inc., No. 15-cv-01530, 2016 WL 10650607, at *4 (M.D. Tenn. Oct. 26, 2016) (“[T]his Court will consider Tenn. Code Ann. § 66-11-208(a) as one factor that weighs against enforcement of the Forum Selection Clause. To determine whether to invalidate the Clause in its entirety, the Court must also consider ‘(1) [w]hether the clause was obtained by fraud, duress, or other unconscionable means;
    (2) whether the designated forum would ineffectively or unfairly handle the suit; and (3) whether the designated forum would be so seriously inconvenient such that requiring plaintiff to bring suit there would be unjust.” (quoting Wong v. PartyGaming Ltd., 589 F.3d 821, 826 (6th Cir. 2009)) (alteration in original) (emphasis omitted)).

  164. [164]. Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 15–17 (1972).

  165. [165]. Id. at 18–20

  166. [166]. See generally Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938) (holding federal courts sitting in diversity must generally apply state law). This problem is felt even more acutely in the smattering of cases where the federal courts have held that state law is “irrelevant” to the enforceability inquiry. See, e.g., WCC Cable, Inc. v. G4S Tech. LLC, No. 17-cv-00052, 2017 WL 6503142, at *7 (W.D. Va. Dec. 15, 2017).

  167. [167]. Erie, 304 U.S. at 78–80.

  168. [168]. Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 39 (1988) (Scalia, J., dissenting) (observing that allowing federal courts to disregard state law in determining whether a forum selection clause is enforcing “clearly encourages forum shopping”).

  169. [169]. Steinman, supra note 116, at 804–10 (discussing Erie problems that arise when federal law governs enforceability issue).

  170. [170]. In re Hulu, LLC, No. 2021-142, 2021 WL 3278194, at *2 (Fed. Cir. Aug. 2, 2021) (listing seven factors); Rsch. Automation, Inc. v. Schrader-Bridgeport Int’l, Inc., 626 F.3d 973, 978–79 (7th Cir. 2010) (listing nine factors); Strategic Power Sys. v. Sciemus, Ltd., No. 16-cv-859, 2017 WL 3402082, at *3 (W.D.N.C. Aug. 8, 2017) (listing eleven factors); see also Maggie Gardner, Retiring Forum Non Conveniens, 92 N.Y.U. L. Rev. 390, 414 (2017) (“‘Forum non conveniens’ never meant ‘inconvenient forum’; it translates more correctly to ‘inappropriate’ or ‘unsuitable’ forum.”).

  171. [171]. Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 10 (1972).

  172. [172]. Id. at 18.

  173. [173]. Id. at 12.

  174. [174]. Carnival Cruise Lines, Inc. v. Shute, 499 U.S. 585, 590 (1991).

  175. [175]. Id. at 587–90.

  176. [176]. Id.

  177. [177]. Id. at 590–97.

  178. [178]. Id.

  179. [179]. Hare v. YJ Sales, Inc., No. SACV 17-00373, 2017 WL 7163926, at *4–5 (C.D. Cal. June 15, 2017).

  180. [180]. BH Servs. Inc. v. FCE Benefit Adm’rs Inc., No. 16-cv-05045, 2017 WL 3635186, at *4–7 (D.S.D. Aug. 23, 2017).

  181. [181]. Alamo Masonry & Constr. Contractors, LLC v. Air Ideal, Inc., No. 13-cv-00448, 2014 WL 1391024, at *2–3 (S.D. Tex. Apr. 8, 2014).

  182. [182]. Carney v. Beracha, 996 F. Supp. 2d 56, 71 (D. Conn. 2014). But see SSAB Ala., Inc. v. Kem-Bonds, Inc., No. CV 17-0175, 2017 WL 6345809, at *6 (S.D. Ala. Dec. 12, 2017) (“Atlantic Marine calls for enforcement of the forum-selection clause, notwithstanding objections grounded in fears of duplicative litigation or judicial economy.”).

  183. [183]. Laferte v. myFootpath, LLC, No. CV 12-10118, 2014 WL 12591801, at *3 (C.D. Cal. July 18, 2014). But see Valspar Corp. v. E.I. DuPont de Nemours & Co., 15 F. Supp. 3d 928, 934–35 (D. Minn. 2014) (“It is always more expeditious to try related claims in one forum rather than several, but allowing efficiency and economy to rule the day would effectively swallow Atlantic Marine’s holding in every case with multiple defendants.”).

  184. [184]. Indingo LLC v. Cohen, No. 16-6525, 2017 WL 59204, at *5 (D.N.J. Jan. 5, 2017).

  185. [185]. Azzia v. Royal Caribbean Cruises, Ltd., No. 15-cv-24776, 2016 WL 11395237, at *3 (S.D. Fla. Aug. 31, 2016); see also Fluence Energy, LLC v. M/V BBC Fin., No. 21-cv-01239, 2022 WL 378197, at *7 (S.D. Cal. Feb. 8, 2022) (“To the extent Fluence was unaware of the Bills of Lading, and the forum-selection clause contained therein, it would be unreasonable to enforce the clause and disregard Fluence’s choice of forum.”); Touloumes v. Kerzner Int’l Bah. Ltd., No. 13-24053-civ, 2014 WL 10102248, at *1–3 (S.D. Fla. Sep. 26, 2014) (“The Court finds ... the private and public interest factors do not weigh in favor of dismissal.”). But see Rivas v. Greyhound Lines, Inc., No. EP-14-cv-0166, 2017 WL 8890775, at *7 (W.D. Tex. Oct. 4, 2017),report and recommendation adopted in part, rejected in part,No. 14-cv-166, 2018 WL 1896413 (W.D. Tex. Mar. 1, 2018) (enforcing clause notwithstanding magistrate judge’s finding that the top half of the ticket is “absolutely illegible because the font is too small. Indeed, I cannot even read it using a magnifying glass. The bottom half of Alejandro’s ticket is even more illegible, and the defendants did not even attempt a translation of it. I have no evidence before me that the plaintiffs received a legible version of the bottom half of the ticket that would have put them on notice”) (footnotes omitted).

  186. [186]. Young v. Holland Am. Line, N.V., No. 16-cv-04820, 2016 WL 7451563, at *3 (N.D. Cal. Dec. 28, 2016). But see Santos v. Costa Cruise Lines, Inc., 91 F. Supp. 3d 372, 380 (E.D.N.Y. 2015) (upholding forum selection clause notwithstanding fact that plaintiffs would have had to forfeit fifty percent of their ticket cost to reject the terms).

  187. [187]. Hussein v. Coinabul, LLC, No. 14 C 5735, 2014 WL 7261240, at *3 (N.D. Ill. Dec. 19, 2014); see also Live Face on Web, LLC v. Complete Fam. Dentistry, P.C., No. CV 16-7, 2016 WL 8813993, at *5 (W.D. Pa. Nov. 18, 2016) (declining to enforce a forum selection when “[t]here [was] no evidence Complete had reasonable notice of or assented to the forum selection clause in the Terms of Use on Solution’s website. The forum selection clause appeared only on Solution’s website. There is no evidence Complete knew or should have known about the forum selection clause or to look at Solution’s website”). But see Omnibus Trading, Inc. v. Gold Creek Foods, LLC, No. 18-cv-02598, 2019 WL 3429048, at *4–5 (N.D. Tex. July 30, 2019) (concluding that a clause was reasonably communicated when the contract contained a link to the General Terms and Conditions and the forum selection clause was contained therein).

  188. [188]. Grice v. VIM Holdings Grp., LLC, 280 F. Supp. 3d 258, 283 (D. Mass. 2017). But see Get in Shape Franchise, Inc. v. TFL Fishers, LLC, 167 F. Supp. 3d 173, 204–07 n.9 (D. Mass. 2016) (enforcing clause notwithstanding the fact that the plaintiff “had an annual income of $24,500 in 2014 and $36,000 in 2013. She also has approximately $45,000 in unspecified ‘debts,’ ‘no liquid assets other than a minor amount in a checking account,’ and ‘does not own a home.’” (citations omitted)); Horne v. Ace Ltd., No. 12-cv-1142, 2014 WL 12788989, at *2 (D. Nev. Mar. 13, 2014) (enforcing clause requiring litigation to proceed in Argentina notwithstanding fact that “plaintiff describes his monthly household income as $3,800, with his bills exceeding $4,000 monthly” and “plaintiff addresses his physical limitations including a hernia, ‘24/7 pain in [his] feet,’ and difficulty sleeping”).

  189. [189]. Lieberman v. Carnival Cruise Lines, No. Civ. A. 13-4716, 2014 WL 3906066, at *12 (D.N.J. Aug. 7, 2014).

  190. [190]. Harmon v. Dyncorp Int’l, Inc., No. 13-cv-1597, 2015 WL 518594, at *9 (E.D. Va. Feb. 6, 2015), aff’d, 624 F. App’x 104 (4th Cir. 2015).

  191. [191]. Montoya v. CRST Expedited, Inc., 285 F. Supp. 3d 493, 497–501 (D. Mass. 2018).

  192. [192]. Id.

  193. [193]. Id. at 499.

  194. [194]. Id. at 497–501.

  195. [195]. Davila v. Adesa Utah, LLC, No. 20-cv-00055, 2020 WL 4784766, at *2. (D. Utah Aug. 18, 2020).

  196. [196]. Id.

  197. [197]. Id. at *2–3.

  198. [198]. See e.g., Rebecca Stone, Legal Design for the “Good Man,102 Va. L. Rev. 1767, 1796–800 (2016).

  199. [199]. Id. at 1798 (citing Roscoe Pound, Law in Books and Law in Action, 44 AM. L. REV. 12 (1910)).

  200. [200]. The dataset contains cases where the court addressed the issue of whether a clause was enforceable as part of its broader inquiry into whether a clause was “contractually valid” for purposes of Atlantic Marine plus cases where one party argued the clause was invalid due to fraud.

  201. [201]. This finding provides empirical support for intuitions long voiced by scholars in this area. See Mullenix, supra note 8, at 750 (“If one sifts through the thousands of reported federal forum selection clause decisions since Zapata—and there are thousands of such decisions—one cannot help but be struck by the following fact: in virtually every case the party seeking enforcement of the clause wins, and the party seeking to invalidate the clause loses.” (footnotes omitted)).

  202. [202]. In many cases, there were a combination of natural persons and entities named as defendants. I coded a case under the “natural persons” category only when the natural person was the only defendant named in the case.

  203. [203]. See supra notes 80–110 and accompanying text.

  204. [204]. See id.; Bremen v. Zapata Off-Shore Co., 407 U.S. 1, 18–20 (1972).

  205. [205]. See supra note 80–110 and accompanying text; Bremen, 407 U.S. at 18–20.

  206. [206]. See Scherk v. Alberto-Culver Co., 417 U.S. 506, 519–20 (1974).

  207. [207]. Id.

  208. [208]. See supra notes 37–56 and accompanying text.

  209. [209]. Id.

  210. [210]. Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 28–32 (1988); see supra notes 89–116 and accompanying text.

  211. [211]. See supra notes 89–116 and accompanying text.

  212. [212]. See supra Section III.B.2.ii.

  213. [213]. See supra notes 170–87 and accompanying text.

  214. [214]. See id.

  215. [215]. See id.

  216. [216]. See supra Section I.B.3.

  217. [217]. See supra Section I.B.2.

  218. [218]. Mao v. Sanum Invs., Ltd., No. 14-cv-00721, 2014 WL 5292982, at *2 (D. Nev. Oct. 15, 2014) (“Even though, as Mao points out, the clause permits Bridge Capital and Sanum Investments to bring suit in other jurisdictions but prohibits Mao from doing so, this does not invalidate the clause. Unequal contract terms and unequal bargaining power will not invalidate a forum-selection clause.”); Carter’s of New Bedford, Inc. v. Nike, Inc., No. Civ A 13-11513, 2014 WL 1311750, at *4 (D. Mass. Mar. 31, 2014), aff’d, 790 F.3d 289, 294 (1st Cir. 2015) (“Carter’s broadest argument is that the forum selection clause is simply unfair; both in that it requires a small family-owned business from Massachusetts to litigate disputes with Nike across the country in Oregon, and in that Nike is not similarly restricted in its ability to select a forum.”).

  219. [219]. Anderson Holdings, Inc. v. Cyclebar Franchising, LLC, No. 18-cv-131, 2018 WL 6430828, at *3 (S.D. Miss. Sept. 24, 2018).

  220. [220]. The fact that the statute of limitations has run in the chosen court typically does not provide a valid basis for declining to enforce a forum selection clause. See Barnett v. DynCorp Int’l LLC, No. 15-cv-233, 2015 WL 12714715, at *4 (N.D. Tex. July 13, 2015) (observing “that the vast majority of courts have found that the enforcement of foreign forum selection clauses is not unreasonable, even when the contractual forum’s statute of limitations would bar a plaintiff’s action” (emphasis omitted)).

  221. [221]. See Tanya Monestier, Damages for Breach of a Forum Selection Clause, 58 Am. Bus. L.J. 271, 280 (2021).

  222. [222]. See Doroghazi & Norman, supra note 12, at 581 (“It is no secret that home turf is an advantage. Plants grow best in their native soil and climate. Sports teams win more often on their home court or field. This trope remains true in litigation. An attorney litigating in his or her home court knows the judges and can tailor litigation strategy to the assigned judge’s preferences and proclivities.” (footnotes omitted)); John C. Jorgenson, Note, Drafting Effective Delaware Forum-Selection Clauses in the Shadow of Enforcement Uncertainty, 102 Iowa L. Rev. 353, 378 (2016) (encouraging Delaware corporations to write exclusive forum selection clauses into their corporate bylaws so as to “achieve[] the upside of the ability to litigate selectively in a convenient forum”).

  223. [223]. Cara Reichard, Note, Keeping Litigation at Home: The Role of States in Preventing Unjust Choice of Forum, 129 Yale L.J. 866, 869 (2020) (“[Forum selection clauses] can create a significant obstacle for potential litigants—particularly employees, consumers, or other relatively powerless individuals who might be wronged at the hands of a corporate entity.”).

  224. [224]. As discussed above, only two percent of the defendants in the dataset were natural persons without any affiliated entities. See also Mullenix, supra note 8, at 737 (“Because of the enormous strategic advantage conferred by contractual forum-selection clauses on defendants and the fundamental unfairness of the law to consumers governing such provisions, it is thought provoking to view forum-selection clauses, then, as a strategic mechanism to game the system rather than through the lens of sanctified contract principles.”).

  225. [225]. The uniquely favorable treatment of arbitration clauses is largely attributable to the existence of the Federal Arbitration Act. There is no comparable federal statute that governs forum selection clauses.

  226. [226]. See supra notes 65-87.

  227. [227]. Reichard, supra note 223, at 872 (“Large corporate powers today have nearly every advantage over the individuals with whom they contract, not least because they prescribe the terms of those contracts. Anti-choice-of-forum laws, including those already adopted by many states, offer a rare opportunity to redistribute power by ensuring that, in the event of a legal claim, the forum is one that does not disadvantage the relatively powerless individual. In litigation against corporate entities, individuals already face enough challenges.”).

  228. [228]. Coyle & Richardson, supra note 37, at 1234.

  229. [229]. Mullenix, supra note 8, at 757–58 (“If consumers are to be afforded meaningful relief from such clauses, then federal statutory substantive law is needed to determine the validity and enforcement of a forum-selection or choice-of-law clause challenged by a plaintiff ... .” (footnotes omitted)).

  230. [230]. See supra notes 135–64 and accompanying text.

  231. [231]. See, e.g., DePuy Synthes Sales, Inc. v. Howmedica Osteonics Corp., 28 F.4th 956, 957–60 (9th Cir. 2022); Gemini Techs., Inc. v. Smith & Wesson Corp., 931 F.3d 911, 912–14 (9th Cir. 2019); see also Davis v. Oasis Legal Fin. Operating Co., LLC, 936 F.3d 1174, 1177–83 (11th Cir. 2019) (declining to enforce a forum selection clause).

  232. [232]. See supra note 147–78 and accompanying text; see also Albemarle Corp. v. Astrazeneca UK Ltd., 628 F.3d 643, 652 (4th Cir. 2010) (“The Bremen would have little effect if states could effectively override the decision by expressing disagreement with the decision’s rationale” (emphasis added)).

  233. [233]. See, e.g., Skoglund v. PetroSaudi Oil Servs. (Venez.), No. 18-386, 2018 WL 6112946, at *6 (E.D. La. Nov. 20, 2018) (enforcing a forum selection clause in a contract of adhesion requiring a blind and crippled oil rig worker to travel to a foreign country located more than four thousand miles away to bring suit against an oil company that lacked a significant connection to the chosen forum).

  234. [234]. See Boyd v. Grand Trunk W. R.R. Co., 338 U.S. 263, 265 (1949) (“[C]ontracts limiting the choice of venue are void as conflicting with the Liability Act.”); see also 14 Penn Plaza LLC v. Pyett, 556 U.S. 247, 273–74 (2009) (“[A] substantive waiver of federally protected civil rights will not be upheld . . . .”); Alexander v. Gardner-Denver Co., 415 U.S. 36, 51 (1974) (“[T]here can be no prospective waiver of an employee’s rights under Title VII.”).

  235. [235]. See Coyle & Effron, supra note 48, at 198–205.

  236. [236]. See Coyle, supra note 9, at 1796 n.12.

  237. [237]. Atl. Marine Constr. Co. v. U.S. Dist. Ct. for W. Dist. of Tex., 571 U.S. 49, 62–68 (2013).

  238. [238]. Id. at 62 n.5.

  239. [239]. See William H.J. Hubbard, The Effects of Twombly and Iqbal, 14 J. Empirical Legal Stud. 474, 481 (2017) (observing that cases identified through “databases of published judicial opinions” are “notrepresentativeofcasesas a whole, both because published opinions are not a random sample of all judicial decisions, and becausecaseswith judicial decisions are not a random sample of allcases”).

  240. [240]. See, e.g., David A. Hoffman, Alan J. Izenman & Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash. U. L. Rev. 681, 686–89 (2007).

  241. [241]. Empirical legal scholars have recognized that a “systematic review” of cases that is transparent about collection methods and findings may serve to make doctrinal work more rigorous. William Baude, Adam S. Chilton & Anup Malani, Making Doctrinal Work More Rigorous: Lessons from Systematic Reviews, 84 U. Chi. L. Rev. 37, 51 (2017) (“We propose a four-step process for making claims about the state of legal doctrine: (1) clearly stating the legal question that is being answered; (2) defining the sample of cases that will be used; (3) explaining how the cases in the sample will be weighted; (4) conducting the analysis of the sample of cases and stating the conclusion.”). The data-collection methods utilized in this article are consistent with each of this recommended approach.

  242. [242]. Theodore Eisenberg & Sheri Lynn Johnson, The Effects of Intent: Do We Know How Legal Standards Work?,76 Cornell L. Rev. 1151, 1195 (1991).

*

Reef C. Ivey II Distinguished Professor of Law, University of North Carolina at Chapel Hill.



Thanks to Pam Bookman, Bill Dodge, Robin Effron, Andy Hessick, Leigh Osofsky, Katie Richardson, Kim Roosevelt, Aaron Simowitz, Kathleen Thomas, Andrea Wang, and Diego Zambrano, for their comments on an earlier version of this paper.