chevron-down Created with Sketch Beta.

    The Business Lawyer

    Winter 2024-2025 | Volume 80, Issue 1

    Third-Party Boilerplate Providers and Arbitration Boilerplate

    Christopher R Drahozal

    Mar 05, 2025

    28 min read

    Arbitration Bankruptcy & Finance Business Litigation Business Litigation & Dispute Resolution Business Regulation & Regulated Industries Class Actions Commercial Transactions Consumer Finance Dispute Resolution Labor & Employment Law

    Summary

    • This article offers some preliminary thoughts on whether contractual boilerplate from third-party boilerplate providers avoids the problems with contractual boilerplate identified by Stephen J. Choi, Mitu Gulati, and Robert E. Scott. 
    • It focuses on arbitration institutions as third-party boilerplate providers and identifies possible examples of encrustation, contractual black holes, and contractual landmines in arbitration boilerplate.
    • It concludes that third-party boilerplate providers do not avoid the problems identified by Choi, Gulati, and Scott, nor do they necessarily result in prompt correction of problems that do occur.
    Third-Party Boilerplate Providers and Arbitration Boilerplate
    iStock.com/Sjo

    Jump to:

    I. Introduction

    Boilerplate terms in contracts are “terms that are similar or identical in other agreements of a similar type.” Examples include choice-of-law clauses, forum selection clauses, damages limitations, and, of particular interest here, arbitration clauses. The commonly held view is that even in commercial contracts, boilerplate terms “are frequently pasted from prior deals: these are terms that no one reads, let alone bargains over.” As a result, while boilerplate terms are inexpensive to produce, they can lead to “errors that remain in the contract even in the face of adverse legal consequences.”

    Stephen J. Choi, Mitu Gulati, and Robert E. Scott have identified several types of errors that can result from the use of boilerplate contract terms. Repetition in the use of a boilerplate term can result in “rote usage”—when “standardized terms . . . get used by rote so consistently that they lose a shared meaning and become a ritualized legal incantation”—and “encrustation”—when “the intelligibility of language deteriorates significantly as legal jargon is added to standard formulations, leading to linguistic variations of the same clause.” Eventually, a boilerplate term “can be emptied of any recoverable meaning” and become (in their parlance) a “contractual black hole.” Boilerplate terms that have become contractual black holes pose “a heightened risk that courts may be persuaded to adopt an interpretation of the term at issue that is antithetical to the functioning of the market that relies on the standard contract to regulate the rights and duties of the participating parties.”

    Or the boilerplate terms can be “landmines”—“embedded language that lies dormant, sometimes for many years,” before “the harmful mutation is used as leverage in a subsequent dispute.” Contractual landmines can arise in a variety of circumstances. They can be “historic holdovers” that are “empty vessels” until “experienced litigators assert an unexpected meaning for the term.” They can be random errors that result when a drafter “inserts language that might have made sense in a different context . . . but is problematic” in the contract in which it is used. Or they can result when “a standardized, and well understood, term is changed by the intentional addition (or omission) of a new phrase or clause because of a conflict at the deal making stage that must be overcome to get the transaction done.” Regardless of the source, these “blunders, gaps, and booby traps in the documents do not diminish over time with repeated use; if anything, the general pattern is that the contracts develop more landmines over time.”

    The paradigmatic case of contractual boilerplate studied by Choi, Gulati, and Scott is the pari passu clause in sovereign bond contracts. This clause, “a boilerplate formulation common to sovereign debt contracts for nearly 200 years,” continued to be included in such contracts by big firm lawyers long after courts interpreted the clause in a way that major players in the market agreed was incorrect and that imposed serious costs on the parties. The “lawyers had no incentive to revise the standard terms for their individual clients,” despite the fact that revision was in their clients’ collective interests.

    But firms obtain contractual boilerplate not only from their lawyers, but also from a variety of third-party boilerplate providers. Thus, as stated by Lisa Bernstein, “[t]rade associations are, and traditionally have been, important sources of standard-form contracts—in the form of both traditional contracts and trading rules that can be incorporated into contracts by reference.” Examples include the American Institute of Architects, local realtors associations, and the trade associations of commodities firms studied by Bernstein. Form sellers, such as TruStage Compliance Solutions (formerly CUNA Mutual Group) with its financial services contracts for credit unions, sell standard form contracts, typically to smaller entities that do not want to pay lawyers to draft forms customized for them. The Consumer Financial Protection Bureau, in its study of consumer financial services arbitration, reported that “[a]t least 83 of the 141 small to mid-sized banks (58.9%) in the checking account sample used some version of a standard form prepared by a single form provider.” While some third-party boilerplate providers provide entire standard form contracts, others provide only specialized contract terms. The International Chamber of Commerce, for instance, publishes its Incoterms and Uniform Customs and Practice for Documentary Credits, among others, while arbitration institutions provide both model arbitration clauses and standard form arbitration rules.

    This article examines whether contractual boilerplate from third-party boilerplate providers avoids the problems with contractual boilerplate identified by Choi, Gulati, and Scott, with a particular emphasis on arbitration institutions and arbitration boilerplate. The article first offers some general thoughts on contractual boilerplate and third-party boilerplate providers. Then it discusses arbitration institutions as third-party boilerplate providers. Finally, the article identifies possible examples of encrustation, contractual black holes, and contractual landmines in arbitration boilerplate. It concludes that third-party boilerplate providers do not avoid the problems identified by Choi, Gulati, and Scott, nor do they necessarily result in prompt correction of problems that do occur.

    II. Third-Party Boilerplate Providers and Contractual Boilerplate

    Third-party boilerplate providers differ in a variety of ways from the law firms studied by Choi, Gulati, and Scott as sources of contractual boilerplate. This part offers some preliminary thoughts on those differences and their possible implications for contractual boilerplate as analyzed by Choi, Gulati, and Scott. It focuses on three ways in which at least some third-party boilerplate providers differ from law firms that draft sovereign bond contracts: some third-party boilerplate providers (1) operate on a not-for-profit rather than a for-profit basis (“not-for-profit”); (2) create industry standard forms (“standardization”); and (3) specialize in particular types of boilerplate terms (“specialization”).

    A. Not-for-Profit

    Unlike law firms, some third-party boilerplate providers are not-for-profit entities. Kevin Davis has identified a number of differences between not-for-profit entities and for-profit entities as providers of boilerplate: not-for-profit entities may (1) “take into account benefits and costs that are not recognized by for-profit organizations”; (2) be “relatively well placed to stimulate demand for contracts by credibly assuring prospective users of their value”; (3) “produce contracts of a given quality at a relatively low cost because they have superior ability to attract volunteer labor”; and (4) “produce contracts at a relatively low cost because they enjoy preferential tax treatment.”

    The extent to which the incentives facing not-for-profit entities, especially ones competing with for-profit entities in the marketplace, in fact differ or alter their behavior relative to for-profit entities is uncertain. Even if not-for-profit entities do take into account different costs and benefits from for-profit entities, it is not obvious whether that would make them more or less likely to provide problematic boilerplate. That said, the lower costs faced by not-for-profit entities producing contractual boilerplate might make it economical for them to produce greater documentation surrounding the boilerplate they provide, thus reducing the likelihood that a contractual black hole will form. Likewise, not-for-profit entities may have greater incentive to make public such documentation (perhaps to enhance their fundraising and other efforts), also reducing the likelihood that the original meaning will be lost. Conversely, not-for-profits might be slower to respond to the need for changes in their boilerplate (such as to address contractual black holes or landmines), either because of the differing incentives they face or their volunteer labor force. And to the extent the not-for-profit boilerplate provider has the characteristics of a private legislature, the drafting and revision process may face additional limitations and hurdles to overcome.

    B. Standardization

    Trade associations might provide boilerplate that serves as the de facto industry standard. If so, one certainly might still see rote reiteration of boilerplate but perhaps less encrustation (if nothing else because there will be fewer variations in contract language). To the extent the standardized boilerplate is provided by a trade association, the trade association might be likely to revisit and revise it on a regular basis (such that black holes and landmines might be less likely to persist):

    These contracts and rules [provided by trade associations] were regularly revised in response to problems that arose, changes in technology, and other changes in market conditions. The process of adopting these changes was—and in the case of trading rules, continues to be—costly and time-consuming. . . . Changes are researched and debated extensively. In most groups, rule changes must be approved by a majority of group members, making it important for the revisers to clearly articulate the reasons for the proposed change. Although changes in optional standard-form contracts do not typically require membership approval, in practice associations go to great lengths to adopt only those changes that will be widely accepted in the trade.

    So, like boilerplate provided by not-for-profits more generally, standardized boilerplate may be more likely to have its drafting history documented and to be revised regularly, albeit through an arduous and demanding process.

    In addition, trade associations may provide an internal dispute resolution system under which disputes between members are resolved in arbitration with industry-expert arbitrators rather than in court. By keeping disputes between parties to standardized boilerplate out of the courts, trade association arbitration may reduce the likelihood that the meaning of contractual boilerplate will be lost or that landmines will appear because the arbitrators deciding any disputes will be “well versed in the meaning of trade rules and standard contract provisions” (even if the arbitrators apply those rules formalistically rather than flexibly).

    C. Specialization

    Some third-party boilerplate providers provide standard form contracts in their entirety. Others provide individual terms as to which they have particular expertise, which parties and their lawyers can incorporate (either word-for-word or by reference) into their standard form contracts.

    Specialization would seem to reduce the likelihood that boilerplate would lose its meaning or result in contractual landmines. First, specialization might mean that the provider has fewer contractual terms to draft in the first instance and hence to monitor on an ongoing basis. The lower ongoing cost of monitoring and updating specialized boilerplate makes it easier to retain knowledge of the meaning of the boilerplate terms and avoid landmines. It also might enable the specialized boilerplate provider to identify court interpretations that require the meaning of boilerplate terms to be clarified. Second, specialization is likely to result from and lead to greater expertise in a particular type of term. The greater expertise of specialized third-party boilerplate providers likewise should make them better able both to retain information about the meaning of boilerplate terms and to recognize the need for clarification. Accordingly, boilerplate provided by specialized providers seems less likely to result in contractual black holes and landmines and more likely to be revised in response to an aberrant court interpretation.

    III. Arbitration Boilerplate and Third-Party Boilerplate Providers

    This part looks at arbitration institutions as a type of third-party boilerplate provider. It first provides a brief introduction to arbitration and discusses arbitration clauses as an example of contractual boilerplate. It then describes the role of arbitration institutions as third-party providers of arbitration boilerplate.

    A. Arbitration Clauses as Contractual Boilerplate

    Arbitration is a common albeit not ubiquitous form of dispute resolution: the use of arbitration clauses varies by industry and by size of firm within an industry. By agreeing to arbitrate, parties agree to have their disputes resolved by private judges (arbitrators) rather than in the public court system. As often stated by the U.S. Supreme Court, arbitration is “a matter of contract”: parties cannot be required to arbitrate if they have not agreed to do so. Stated otherwise, court litigation is the default means of resolving disputes. Parties can contract out of that default rule by agreeing to arbitration.

    An arbitration clause is a classic example of a boilerplate contract provision. Parties typically agree to arbitration before a dispute arises by including a (pre-dispute) arbitration clause in their contract. In negotiated contracts, the negotiation and drafting of arbitration clauses appears to track the story of commercial boilerplate production more generally, as told by Choi, Gulati, and Scott. Arbitration clauses (like other boilerplate terms such as choice-of-law clauses)

    are often added at the last minute and addressed at the end of contract negotiations with very little thought and consideration given to the consequences and intricacies of the clause, hence earning the name “midnight” clause. The excuse for this late consideration by some is fear that negotiations will break down by damaging the optimistic structuring of a deal by introducing the idea that something might go wrong.

    Because of the tight timeframe, rather than being drafted from scratch, often “boilerplate language from another unrelated contract is lifted and inserted in the deal contract.” As a result, commentators conclude, “poorly drafted dispute resolution clauses frequently trap parties in a less than optimal dispute resolution process.” By comparison, arbitration clauses (and other boilerplate terms) in standard form contracts presumably are not produced by the same last-minute negotiation and drafting process. Nonetheless, those clauses can be less than optimal as well.

    B. Arbitration Institutions as Third-Party Boilerplate Providers

    Although not required to do so, parties commonly agree to have an arbitration institution administer any arbitration that might occur. Thus, 89.5 percent of a sample of arbitration clauses in international supply contracts (77 out of 86) provided for institutional (rather than ad hoc) arbitration, with the American Arbitration Association (AAA) (or its international wing, the International Centre for Dispute Resolution (ICDR)) and the International Chamber of Commerce (ICC) International Court of Arbitration the most commonly chosen institutions. Essentially all U.S. consumer financial service contracts with an arbitration clause likewise provided for an administering institution, almost always either the AAA or JAMS.

    Arbitration institutions provide administrative services to parties during the course of the arbitration proceeding. The arbitration institution may act as a court clerk’s office, accepting and distributing filings made in the arbitration, scheduling hearings, and the like. It can serve as the appointing authority, appointing arbitrators when a party fails to do so and ruling on challenges to arbitrators. It may also handle the fees charged to the parties both for its services and for the arbitrators’ services.

    But arbitration institutions also serve as third-party boilerplate providers for parties who use their services. First, institutions draft and publish model arbitration clauses that parties can include in their contracts. Use of the model clauses is not mandatory, and, as noted below, parties rarely use the model clause without modification. Increasingly, institutions go beyond simply providing a model arbitration clause and sometimes provide an online tool to assist parties in drafting a more customized arbitration clause.

    Second, institutions provide a bundle of standard form arbitration rules that parties can incorporate by reference into their arbitration agreement. Indeed, the rules commonly provide that by agreeing to have an institution administer their arbitration the parties also agree to be subject to the institution’s arbitration rules. Arbitration institutions often enlist both internal and external expertise (e.g., from users of the institution’s arbitration services, arbitration practitioners, and arbitrators) to prepare and revise their arbitration rules, albeit typically with less public involvement than, say, the process for revising federal court rules. Institutional arbitration rules cover the full extent of the arbitration proceeding, from the filing of the claim to the issuance of the award. With some exceptions, the parties are free to—and commonly do—modify various of the institutional rules.

    In a minority of cases, parties to arbitration agreements do not specify an arbitration institution but instead agree to have the arbitration proceed on an ad hoc basis—i.e., with the arbitrators themselves handling the administrative services and with some other third party (or court) serving as appointing authority. In such cases, the parties can either draft arbitration rules themselves or agree to a set of arbitration rules created for use in ad hoc proceedings. An example of the latter is the UNCITRAL Arbitration Rules, which were promulgated in 1976 (and amended in 2010 and 2013) by the United Nations Commission on International Trade Law (UNCITRAL). The UNCITRAL Arbitration Rules have a well-documented drafting history and have influenced the development of institutional arbitration rules.

    Arbitration institutions reflect several (although not all) of the characteristics of third-party boilerplate providers discussed in Part II. Arbitration institutions are both for-profit (e.g., JAMS) and not-for-profit (e.g., AAA) entities. Each institution focuses on a single, specialized type of boilerplate—arbitration clauses and rules. Arbitration boilerplate is not standardized, however. No single arbitration institution has a sufficient market share to impose a standardized arbitration clause, and the costs of modifying model arbitration clauses and institutional arbitration rules in contracts evidently is low enough that parties do so with some frequency.

    IV. Arbitration Boilerplate and the Theory of Contractual Boilerplate

    This part considers whether the problems with contractual boilerplate identified by Choi, Gulati, and Scott occur in arbitration boilerplate, despite (or perhaps because of ) the involvement of third-party boilerplate providers (arbitration institutions). Its analysis is anecdotal, rather than empirical. Nonetheless, it provides evidence that, at a minimum, third-party boilerplate providers do not avoid the problems with contractual boilerplate that Choi, Gulati, and Scott identify.

    This part first examines scope provisions, which define the set of disputes the parties have agreed to arbitrate, as an example of rote usage and encrustation of contract language. Second, it discusses whether institutional arbitration rules providing that arbitrators have authority to resolve challenges to their own jurisdiction, when interpreted as a delegation clause, constitute a contractual black hole. Third, it looks at several types of arbitration boilerplate that are, or might be, landmines.

    A. Scope Provisions and Encrustation

    An arbitration clause must define the set of disputes that the parties are agreeing to submit to arbitration—i.e., its scope. Whether a dispute falls within the scope of an arbitration clause is a commonly litigated issue in disputes over the enforceability of arbitration clauses. Parties wishing to avoid arbitration argue that the dispute is outside the scope of the clause and thus must be decided in court, while parties wishing to arbitrate argue the opposite.

    Although arbitration institutions commonly offer model arbitration clauses for parties to incorporate into their contracts, those model clauses vary in how they define the scope of the obligation to arbitrate, as shown in Table 1. The AAA clause, for example, applies to “[a]ny controversy or claim,” while the ICC clause applies to “[a]ll disputes” and the JAMS clauses apply to “[a]ny dispute, claim or controversy” (for domestic contracts) and “[a]ny dispute, controversy or claim” (for international disputes). The language in the JAMS clause for international disputes is the same as in the UNCITRAL clause. The AAA clause, as well as the JAMS and UNCITRAL clauses, use the connecting phrase “arising out of or relating to,” while the ICC clause uses the connecting phrase “arising out of or in connection with.” The ICC clause then concludes with “the present contract.” The AAA clause adds “or the breach thereof,” the UNCITRAL clause adds “or the breach, termination or invalidity thereof,” and the JAMS clauses add “or the breach, termination, enforcement, interpretation or validity thereof, including the determination of the scope or applicability of this agreement to arbitrate” or “including the formation, interpretation, breach or termination thereof, including whether the claims asserted are arbitrable.” All of the model clauses aim to include a broad range of disputes within their scope. Despite the wording differences, there is no indication that any model clause seeks to cover a different set of disputes than the others.

    Moreover, parties rarely follow the model clauses, even as to such core provisions as scope. For example, in a sample of eighty-six international supply contracts with arbitration clauses, only nine of the clauses (or 10.5 percent) included language matching one of the model clauses quoted in Table 1. Prior studies have similar findings.

    Table 1 Scope Provisions in Model Arbitration Clauses

    American Arbitration Ass’n

    “Any controversy or claim arising out of or relating to this contract, or the breach thereof … .” *

    Int’l Chamber of Commerce

    “All disputes arising out of or in connection with the present contract … .” †

    JAMS (domestic cases)

    “Any dispute, claim or controversy arising out of or relating to this Agreement or the breach, termination, enforcement, interpretation or validity thereof, including the determination of the scope or applicability of this agreement to arbitrate … .” ‡

    JAMS (int’l cases)

    “Any dispute, controversy or claim arising out of or relating to this contract, including the formation, interpretation, breach or termination thereof, including whether the claims asserted are arbitrable … .” §

    UNCITRAL Arbitration Rules

    “Any dispute, controversy or claim arising out of or relating to this contract, or the breach, termination or invalidity thereof … .” ||

    * AAA Commercial Arbitration Rules, supra note 48, at 8; see also Int’l Centre for Disp. Resol., Guide to Drafting International Dispute Resolution Clauses 2 (undated), www.adr.org/sites/default/files/document_repository/ICDR%20Guide%20to%20Drafting%20International%20Dispute%20Resolution%20Clauses%20-%20English.pdf.

    † Int’l Chamber of Commerce, Standard ICC Arbitration Clauses (Mar. 1, 2017), www.iccwbo.org/products-and-services/arbitration-and-adr/arbitration/standard-icc-arbitration-clauses/.

    ‡ JAMS Clause Workbook, supra note 67, at 2.

    § Id.

    || UNCITRAL Arbitration Rules, supra note 61, Annex (“Model arbitration clause for contracts”).

    More striking, however, is the degree of variation in each of the central elements of the scope provisions in the arbitration clauses in the international supply contracts. Those arbitration clauses included twenty different descriptions of the disputes subject to the clause, thirty-five different formulations to describe the contract in which the arbitration clause is used, and twenty-one variations of the language describing the relationship between the two. Overall, the eighty-six clauses contained at least seventy different formulations of scope language, with the most common formulation appearing in only four contracts. If instead the parties had used the model clauses suggested by arbitration institutions, the degree of variation would have been substantially less.

    U.S. courts typically do not try to parse or distinguish among these numerous variations in contractual language. Instead, they usually deal with the wide variety of scope formulations by ignoring the variations in language and classifying arbitration clauses as either “broad” or “narrow.” If the court characterizes the scope of an arbitration clause as “broad,” then it is likely to find it to apply to a range of disputes collateral to the contract. If the court characterizes the scope of an arbitration clause as “narrow,” then it is much less likely to find collateral disputes subject to arbitration.

    A main advantage of this sort of “simplified classification” as a response to rote usage and encrustation is that it reduces the risk of arbitrary contractual interpretations by the courts. Given that, as Gary Born states, “the notion that business men or women compare different formulae, such as ‘arising out of,’ ‘relating to,’ or ‘in connection with,’ is difficult to accept in most settings,” the simplified classification approach reduces the risk of decisions arbitrarily divorced from the meaning of the boilerplate language. But it does so at the expense of any consideration of parties’ individualized intent in an appropriate case. As the Macneil treatise explains, under this approach, “the form of an arbitration agreement may assume an unfortunate talismanic function, going somewhat beyond simply looking at the language to ascertain party intention,” which risks overbroad or underinclusive interpretations of the scope provision.

    B. Delegation Clauses and Contractual Black Holes

    The Federal Arbitration Act (as construed by the courts) sets out default rules governing who decides challenges to the enforceability of arbitration agreements. The general line is between challenges to the enforceability of an arbitration clause itself, over which courts have the final say, and challenges to the enforceability of the main contract that includes the arbitration clause, which are for the arbitrators to decide. One exception is that issues of assent, whether directed to the arbitration clause or to the main contract, remain for the court to decide.

    In Rent-A-Center, West, Inc. v. Jackson, the U.S. Supreme Court held that parties can change the default allocation of authority by delegating challenges to the enforceability of the arbitration clause itself to the arbitrators for final decision. The arbitration clause at issue in Rent-A-Center contained the following provision, which the Court (following the parties) referred to as a “delegation provision”:

    The Arbitrator, and not any federal, state, or local court or agency, shall have exclusive authority to resolve any dispute relating to the interpretation, applicability, enforceability or formation of this Agreement including, but not limited to any claim that all or any part of this Agreement is void or voidable.

    After Rent-A-Center, if the parties’ arbitration agreement contains a delegation provision, challenges to the enforceability not only of the main contract but also the arbitration clause are for the arbitrator to decide. Only challenges to the enforceability of the delegation clause itself (and, of course, issues of assent) are for the court to decide.

    Most institutional arbitration rules address the authority of arbitrators to decide challenges to the enforceability of the arbitration agreement (so-called competence-competence clauses)—albeit without the exclusivity language included in the Rent-A-Center delegation provision. For example, the AAA Commercial Arbitration Rules provide that “[t]he arbitrator shall have the power to rule on his or her own jurisdiction, including any objections with respect to the existence, scope, or validity of the arbitration agreement or to the arbitrability of any claim or counterclaim, without any need to refer such matters first to a court.” Other institutional arbitration rules have similar competence-competence clauses albeit with somewhat different language.

    U.S. courts have, with only a few exceptions, concluded that the competence-competence clauses in institutional arbitration rules (at least as between sophisticated parties) have the same effect as the delegation provision in Rent-A-Center, despite the differing language—i.e., that they delegate to the arbitrator conclusive authority to rule on most challenges to the enforceability of an arbitration agreement. Court have so held as to the competence-competence clauses in the AAA, ICDR, JAMS, ICC, and UNCITRAL arbitration rules, among others, typically without considering the varying language of the rules.

    But there is a strong argument that U.S. courts have misconstrued the competence-competence clauses in institutional arbitration rules, with their original meaning seemingly lost to the courts. As explained in a Reporters’ Note to the Restatement of the U.S. Law of International Commercial and Investor-State Arbitration, the “principal reason” for including such competence-competence clauses in institutional arbitration rules “was to dispel the notion that arbitrators could only decide their own jurisdiction to the extent that the parties expressly authorize them to do so.” On this view, without such boilerplate language in the arbitration rules, the arbitrators might have to stop the case and wait until a court rules on a challenge to the enforceability of the arbitration agreement rather than addressing the issue themselves. But as the Reporters’ Note states: “the evidence generally suggests that [the arbitral rules of the major international arbitration institutions] were not intended to render exclusive the competence of arbitral tribunals to make jurisdictional determinations.”

    The one exception is the AAA’s commercial and international arbitration rules. I will not discuss the drafting history of those rules in detail here. But that history shows two things: first, that the AAA committees that drafted amendments to the rules believed (at least when the AAA was amending its international rules) that the competence-competence clauses in the rules act as a delegation provision; and second, that the language and source of the competence-competence clauses do not support such an interpretation. Instead, the language of the rules and its source (the UNCITRAL Arbitration Rules) support the Restatement’s view that the rules do not give arbitrators the exclusive competence to rule on their own jurisdiction. Moreover, at a minimum, even if the subjective intent of the AAA drafting committees does control as applied to the AAA and ICDR rules, it should not control as to the competence-competence clauses of the arbitration rules promulgated by other institutions for which the evidence shows a different subjective intent of the drafters. Nevertheless, as discussed above, most U.S. courts nonetheless treat competence-competence clauses in institutional arbitration rules as delegation clauses.

    Interestingly, the AAA and the ICDR both amended the competence-competence clauses of their arbitration rules in 2021/2022, with the stated aim of resolving the “potential controversy” cause by the Restatement’s interpretation of the clauses. The amendments added the phrase “without any need to refer such matters first to a court” to the end of the competence-competence clauses. But rather than contradicting the Restatement interpretation, the added language reinforces it: the language makes even more clear that the rules authorize the arbitrator to act without waiting for courts to decide jurisdictional issues (the Restatement interpretation), rather than making the arbitrators’ authority exclusive. Despite the stated intent of the drafters to make clear that the competence-competence clauses function as delegation provisions, the language of the amendments fails to accomplish that end.

    C. Possible Landmines in Arbitration Boilerplate

    Arbitration boilerplate also includes a variety of possible contractual landmines, as defined by Choi, Gulati, and Scott. This subsection begins by examining two examples of arbitration boilerplate in negotiated contracts that might be contractual landmines. It then looks at other possible landmines in arbitration boilerplate, this time in employment and consumer contracts rather than commercial contracts.

    1. Arbitral Seat Versus Venue of Hearing

    The seat, or legal home, of the arbitration is a critical component of any international arbitration. By choosing an arbitral seat, the parties agree to the arbitration law that governs their arbitration and specify the national courts in which any action to vacate the arbitral award must be brought. The arbitral seat also determines whether the New York Convention applies to the award, at least in countries that have made the reciprocity reservation under the Convention. Often the seat is also the venue for the arbitral hearing, but that is not necessarily so: the hearing can be held elsewhere than the seat for the convenience of the parties and the arbitrator. Accordingly, specifying a hearing venue is not the same as specifying the arbitral seat.

    Because of the importance of the seat, arbitration institutions and commentators strongly recommend that the parties agree on the arbitral seat in their international arbitration agreement. Yet in a sample of arbitration clauses in international sales contracts, only 37.2 percent (29 of 78) clauses expressly named a location as the arbitral seat (or identified the location as the place where the award was made, which is largely to the same effect). Instead, the clauses used language that identified the venue of the hearing or (much more commonly) were ambiguous about whether the location was the seat or the hearing venue. Courts or arbitration institutions might construe an ambiguous clause as specifying the arbitral seat. But if not, the parties could end up with an arbitral seat that differs from their expectations. And, regardless, the added expense and uncertainty due to the ambiguity seems to make it appropriate to treat this sort of provision as a contractual landmine.

    2. Carve-Outs and Delegation Clauses

    Another example of a possible landmine arises out of the intersection of carve-outs from arbitration clauses with the competence-competence clauses of institutional arbitration rules. Carve-outs are provisions in arbitration clauses that exclude specified disputes from arbitration. Common carve-outs are for emergency relief, injunctive or equitable relief, and intellectual property actions—types of claims or actions as to which parties sometimes prefer court litigation to arbitration. Parties frequently use carve-outs not only in standard form contracts in the United States (e.g., franchise agreements) but also in international sales contracts.

    As described in the previous subsection, institutional arbitration rules regularly provide arbitrators with jurisdiction to rule on their own jurisdiction (competence-competence clauses), and U.S. courts regularly construe those provisions to act as delegation clauses. But what if a contract with a carve-out also chooses a set of institutional rules with a competence-competence clause? Have the parties agreed to delegate to the arbitrators issues in interpreting the carve-out—which are designed to keep certain types of disputes in court and out of arbitration—or do courts have authority to address those issues? As an example, what if the parties carve out claims for emergency relief from arbitration so that courts can act quickly on those claims if they arise? If the arbitration clause specifies a set of institutional arbitration rules to govern the arbitration (as most do), does any dispute over the carve-out first have to go to arbitration before the court can rule on the requested emergency relief? If so, the resulting delays could undercut the reason the parties agreed to the carve-out in the first place.

    U.S. courts are divided on how to resolve this issue. The Supreme Court has dismissed as improvidently granted a case that seemed to present the issue. The parties can avoid the issue by making clear in the arbitration clause that courts rather than arbitrators have the authority to rule on the scope of any carve-outs. But an arbitration clause that does not address that question includes a landmine that, when triggered, might lead to outcomes contrary to the parties’ evident intent.

    3. Arbitration Costs and Mass Arbitration

    Perhaps the most significant landmine, at least in recent years, is found in arbitral boilerplate in standard form employment and consumer contracts rather than in commercial contracts. The landmine comprises contract provisions and institutional arbitration rules providing that businesses (rather than employees or consumers) are to pay most or all of the administrative costs of arbitration. Relying on these provisions and rules, attorneys have made mass arbitration filings on behalf of employees (including workers that the business classifies as independent contractors) and consumers that require the business to pay sizable arbitration fees at the outset of the case. As just one example, a federal judge ordered DoorDash to pay $1900 in arbitration fees for each of 5010 arbitration demands filed on behalf of its delivery drivers for a total of over $9.5 million. David Horton describes mass arbitration filings as “an existential challenge for companies.”

    Much already has been written about what has given rise to mass arbitration filings as well as descriptions of the proceedings themselves and business responses to the filings. For purposes of this article, I will trace only briefly the development of mass arbitration.

    The story begins with businesses using arbitration clauses to reduce the risk that they would be subject to class actions. Once they started doing so, the businesses sought to protect their arbitration clauses from other legal challenges. One basis for challenging the enforceability of a consumer or employment arbitration clause is to argue that the higher upfront costs in arbitration (as compared to litigation) either were unconscionable or prevented parties from vindicating their federal statutory rights. Businesses sought to protect against this sort of challenge by agreeing themselves to pay most of the administrative costs of arbitration (typically any costs above court filing fees). Arbitration institutions likewise began to implement their Due Process Protocols—private fairness guidelines the institutions promulgated—by setting out fee schedules for consumer and employment arbitrations that required businesses to pay most of the filing fees.

    For businesses subject only to sporadic arbitration filings, agreeing to pay a disproportionate share of arbitration costs would essentially eliminate a ground for challenging the enforceability of the arbitration agreement (thereby preserving their protection from class actions) while imposing minimal costs on the business. But entrepreneurial and well-funded plaintiff-side lawyers, utilizing state-of-the-art case management and communications technology, recently have begun filing individual arbitrations on behalf of large numbers of claimants. The filings trigger the requirement that the business pay its (larger) share of arbitration costs, giving the mass claimants substantial settlement leverage over and above the value of their claims. Needless to say, neither businesses nor arbitration institutions anticipated this consequence of their boilerplate arbitration cost provisions (or at least did not believe it likely to happen). Responses by businesses and arbitration institutions to this contractual landmine are ongoing, and the ultimate resolution is as yet uncertain. But it provides a good illustration of a landmine in a consumer and employment contract rather than a commercial contract.

    4. Choice of Arbitration Law and the FAA’s Employment Exclusion

    Finally, David Horton has identified a possible landmine in arbitration boilerplate in employment contracts based on the interaction of choice-of-law language in some arbitration clauses and the FAA’s employment exclusion. In Volt Information Sciences v. Board of Trustees of Leland Stanford Junior University, the U.S. Supreme Court accepted a California court’s interpretation of a general choice-of-law clause in a contract as incorporating by reference California arbitration law. The effect of the decision was to provide a loophole to FAA preemption: under Volt, a state statute that otherwise might be preempted could be saved from preemption if the parties’ general choice-of-law clause was construed as incorporating the provision by reference into the parties’ contract.

    In response to Volt, businesses began adding language in their arbitration boil-erplate to limit the decision’s effect by making clear that the FAA governs the arbitration. For example, the American Arbitration Association suggests language providing that “[t]he United States Arbitration Act [an archaic name for the FAA] shall govern the interpretation, enforcement, and proceedings pursuant to the arbitration clause in this agreement.” Other clauses go a step further and provide for the FAA to govern to the exclusion of any state arbitration law.

    The landmine results from the fact that the FAA does not cover all employment contracts. Instead, FAA § 1 excludes from the scope of the FAA “contracts of employment of seamen, railroad employees, or any other class of workers engaged in foreign or interstate commerce.” The U.S. Supreme Court has construed section 1 as excluding from the scope of the FAA “transportation workers”—workers, whether employees or independent contractors, who are “actively ‘engaged in transportation’ of . . . goods across borders via the channels of foreign or interstate commerce.” But what if a business includes an arbitration clause in its contract with a transportation worker, and the clause provides that only the FAA and not state law applies? According to Horton, “[s]everal courts, including the Ninth Circuit, held that these drafters had created a contractual abyss in which ‘no law . . . governs the arbitration provision’ and thus ‘did not enter into a valid agreement to arbitrate.’”

    Not all courts agree with the Ninth Circuit’s interpretation. And, as Horton points out, there is an “easy fix” to this problem: businesses could use contract language providing that “the FAA governs but that state law applies if the FAA does not.” Nonetheless, the fact that at least “one variety of FAA choice-of-law clause” has begun to “backfire,” to use Horton’s words, identifies it as a possible landmine.

    D. Implications for the Analysis of Contractual Boilerplate

    Neither scope provisions nor delegation provisions (or more precisely, the competence-competence clauses in institutional rules that courts generally treat as delegation provisions) are contractual black holes as defined by Choi, Gulati, and Scott. Both are frequently litigated, and neither has been in existence long enough for its original meaning to have been lost. That said, both types of provisions share some characteristics with contractual black holes and provide insights into how courts might respond when faced with contractual black holes.

    Scope provisions in arbitration clauses illustrate how boilerplate is subject to rote usage and becomes encrusted with slight variations of contractual language. As noted, the eighty-six arbitration clauses from international supply contracts contained at least seventy variations in scope language. The model arbitration clauses promulgated by arbitration institutions would provide much greater standardization of these boilerplate terms, but only if used frequently by parties, which they are not. Courts typically have responded by classifying the scope provisions as either broad or narrow, rather than parsing the language of individual clauses. Given the institutional limitations faced by courts in interpreting encrusted contractual boilerplate, this simplified classification approach might be the best alternative available, but it is not without costs.

    By comparison, competence-competence clauses in institutional arbitration rules do not reflect the same degree of encrustation as the scope provisions (if for no other reason than that there are far fewer sets of institutional rules than there are arbitration clauses). Nor has the original meaning of the contractual language been lost. Yet with the (somewhat) rote repetition of the language of the competence-competence clauses, its meaning—as construed by courts—has shifted from that suggested by the drafting history. Moreover, once courts (incorrectly, in my view) interpreted the competence-competence clause in one set of institutional rules, that interpretation quickly spread to similar provisions in other sets of rules, even though their drafting history does not support such an interpretation. Notably, as discussed previously, the AAA and ICDR have recently responded to criticism of these interpretations by revising their rules—but did so in a way that likely does not accomplish their stated aim. So not only can third-party boilerplate providers be slow in revising contractual boilerplate, they might also sometimes fail to resolve interpretative issues that arise.

    Nor has the availability and use of contract provisions from third-party boilerplate providers prevented contractual landmines from appearing in arbitration boilerplate. In some cases, the availability of arbitration boilerplate from third-party arbitration providers did not prevent the creation of contractual landmines. Despite guidance from arbitration institutions and commentators, international arbitration clauses often do not clearly identify the arbitral seat. In other cases, institutional arbitral rules and drafting guidance contribute (or at least potentially contribute) to the creation of the contractual landmine. Court interpretations of institutional arbitration rules as delegation clauses create a potential landmine when applied to arbitration carve-outs. The cost provisions of institutional arbitration rules (and the cost provisions in business standard form arbitration agreements) provided the breeding ground for mass arbitration. And drafting guidance from institutional arbitration rules contributes to (or at least does not eliminate) the potential landmine resulting from arbitral choice-of-law clauses in employment arbitration agreements.

    Third-party boilerplate providers thus do not avoid the problems with contractual boilerplate identified by Choi, Gulati, and Scott—at least not for arbitration boilerplate. Whether third-party boilerplate providers reduce those problems (as opposed to eliminating them) is an empirical question that the examples and case studies here cannot answer.

    V. Conclusion

    This article has offered some preliminary thoughts about how the problems with contractual boilerplate identified by Choi, Gulati, and Scott might apply to boilerplate contract terms provided by third-party providers, such as arbitration institutions. Third-party boilerplate providers are more likely to be not-for-profit entities and provide industry-standard or specialized boilerplate than private law firms. Each of these differing characteristics would seem to make it less likely that a contractual black hole or landmine will form and more likely that the third-party provider would revise the boilerplate in response to an erroneous court decision. The examples of arbitration boilerplate here provide evidence that third-party boilerplate providers do not prevent black holes and contractual landmines from forming and do not necessarily promptly correct problems that do occur. Whether third-party boilerplate providers reduce, rather than prevent, the risks identified by Choi, Gulati, and Scott is not something the examples and case studies in this article can resolve but remains an issue worth future research.

    I appreciate helpful comments from John Coyle, Mitu Gulati, David Horton, Mark Weide Maier, and an anonymous reviewer, and discussions with participants at the Conference on Contractual Black Holes at Duke Law School and the 34th Sokol Colloquium: Landmines, Blackholes and the Contract Production Paradox at the University of Virginia School of Law.

      Author