Black Hole Apparitions

Volume 67 December 2017
Black Hole Apparitions
Lisa Bernstein†. Wilson-Dickinson Professor of Law and Aaron Director Research Scholar, The University of Chicago Law School, and International Research Fellow, Center for Corporate Reputation, Said School of Business, University of Oxford. I would like to thank Douglas Baird, Patrick Barry, Edward Bernstein, Brian Bix, Sadie Blanchard, Robin Effron, Roger Ford, Anna Gelpern, William Hubbard, Todd Ito, Avery Katz, Kate Lindgren, Bill Schwesig, Mark Ramseyer, Steve Ware, and participants at the Duke Conference on Contractual Black Holes (2017) for helpful conversations. Special thanks are due to the hole theorists themselves for engaging conversations about their piece over the past year.

Introduction

In their article The Black Hole Problem in Commercial Boilerplate [1][1]. Stephen J. Choi, G. Mitu Gulati & Robert E. Scott, The Black Hole Problem in Commercial Boilerplate, 67 Duke L.J. 1 (2017). and its companion pieces, [2][2]. The other two papers in the Black Hole corpus are Stephen J. Choi, Mitu Gulati, & Robert E. Scott, Contractual Arbitrage, in Oxford Handbook of International Governance (Eric Brosseau, Jean-Michel Galmant & Jerome Sgaard eds., forthcoming 2017) http://scholarship.law.duke.edu/cgi/ viewcontent.cgi?article=6304&context=faculty_scholarship [https://perma.cc/VC3Y-QLKN] and Stephen J. Choi, Mitu Gulati & Robert E. Scott, Variation in Boilerplate: Rational Design or Random Mutation, 19 AM. L. & ECON. REV. (forthcoming 2017) [hereinafter, Choi et al., Variation in Boilerplate]. Stephen Choi, Mitu Gulati, and Robert Scott [hereinafter “CG&S”] introduce and explore the concept of “contractual black holes,” boilerplate contract provisions that have been “emptied of any recordable meaning” [3][3]. Choi et al., supra note 1, at 2. through rote repetition, [4][4]. Id. at 6. the introduction of essentially “random variations in language,” [5][5]. Id. and the absence of “any validation [of their meaning] from courts or industry institutions.” [6][6]. Id. The article presents a detailed case study of the sovereign debt market’s slow response to the Second Circuit’s errant interpretation of the pari passu clause at issue in NML v. Argentina, [7][7]. NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also focuses on an earlier interpretation of the pari passu clause rendered by a Brussels court in 2000. See Elliott Assocs. v. Republic of Peru, 948 F. Supp. 1203 (S.D.N.Y. 1996). a common clause in sovereign debt contracts that they view as the “prototypical exemplar” [8][8]. Choi et al., supra note 1, at 7. of a black hole. Noting that the NML decision put a multi-trillion dollar market at risk by making it more difficult to restructure sovereign debt, [9][9]. Interestingly, the Second Circuit seemed to reject the idea that it was jeopardizing the market as a whole by making restructuring more difficult. See NML Capital, Ltd., 699 F.3d at 263 (explaining why this effect is unlikely to occur.) and that it took years for market transactors to draft clauses eliminating the problem, [10][10]. Choi et al., supra note 1, at 70 (concluding that “the inherently greater inertia costs [and social costs] that result from an aberrant interpretation of a black hole term,” as compared to an ordinary unclear term, and the “greater difficulty market players face in overcoming the resulting collective action problem[s]” slow the adoption of new and better contractual terms). the authors conclude that changes in contract doctrine are needed to deal with the problems caused by black holes. As they explain, “while black holes often remain for many years as relatively harmless surplusage, they can [also] generate substantial social costs once litigation results in an interpretation that introduces inefficiencies into the market.” [11][11]. Id.

This Comment focuses on the authors’ proposed doctrinal solution to the black hole problem, which seeks to eliminate any inquiry into “subjective intent” when courts are faced with the task of interpreting a black hole. It explores the conceptual and practical challenges of implementing the authors’ proposal and then questions whether legal reform is really needed to deal with the black hole problem. Part I identifies several common ways that standardized contract provisions that are often indistinguishable from true black holes may arise. [12][12]. Although it is possible that some of the types of clauses this Comment views as “black hole apparitions” would be classified as genuine black holes by CG&S, their search for additional examples of black holes by holding a conference on the subject at Duke Law School suggests that these ubiquitous types of clauses with uncertain meaning are unlikely to come within the purview of their definition. As Gulati explained, “black holes where meaning has been lost are different from provisions where there is a general sense of what they mean, but there is some vagueness or lack of clarity.” E-mail from Mitu Gulati, Professor of Law, Duke University, to author (Sept. 23, 2017) (on file with author). It suggests that any doctrinal solution to the black hole problem will have to either reliably distinguish real black holes from these relatively common “black hole apparitions,” or be desirable when applied to both types of provisions. Part II describes the proposed reform and explores the practical barriers to implementing it. Along the way it also sketches out several alternative avenues for solving the problems evidenced in the pari passu saga. It suggests that these avenues, while more limited in scope than the authors’ proposed reform, may turn out to be more feasible and less costly to implement. Part III questions whether any doctrinal or other solution to the so-called black hole problem can be justified on the basis of the lessons learned from the pari passu saga alone, given that other markets have been able to overcome collective action problems and adopt and amend standard-form contracts through processes that have tended to work more quickly and less contentiously over time. Finally, Part IV concludes by suggesting that the interpretive approach adopted by the Second Circuit in NML might be a passably good response to the black hole problem writ large.

I. A Typology of Black Hole Apparitions

CG&S suggest that contract doctrine should treat the interpretation of black holes differently from the interpretation of other unclear or essentially meaningless contractual provisions. Yet both conceptually and practically, it may be extraordinarily difficult to distinguish true black holes from what might be called “black hole apparitions.” Black hole apparitions are contract provisions that are indistinguishable on their face from black holes yet will not necessarily give rise to the drafting inertia and social costs that led CG&S to conclude that a doctrinal response to the black hole phenomenon was needed. Recognizing the existence of these apparitions suggests that any doctrinal reform that turns on a provision’s status as a black hole will, [13][13]. In order for a doctrine that conditions on whether a clause is or is not a black hole to be workable, a clause’s status must be verifiable by a court—that is, a court must be able to determine with reasonable accuracy at a cost the transactors consider reasonable from an ex ante perspective whether or not something is a black hole. See generally Alan Schwartz, Relational Contracts and the Courts: An Analysis of Incomplete Agreements and Judicial Strategies, 21 J. Legal Stud. 271 (1992). in practice, bring many black hole apparitions into its orbit. As a consequence, any such change is likely to add significant costs to ordinary commercial litigation and create and/or exacerbate opportunities for strategic behavior. Against this background, and to understand how common these apparitions are likely to be, it is useful to look more closely at three common types of black hole apparitions, those whose origins are procedural, relational or rational.

A. Procedural Apparitions

One category of black hole apparitions is what might be termed procedural apparitions—contract provisions that are at a high risk of losing their meaning due to state-supplied procedural rules or transactors’ procedure-related drafting choices.

For example, black holes are especially likely to arise in markets where most contracts provide for arbitration under the American Arbitration Association’s (AAA) Commercial Arbitration Rules. [14][14]. Commercial Arbitration Rules and Mediation Procedures (Am. Arbitration Ass’n 2013), https://www.adr.org/sites/default/files/Commercial%20Rules.pdf. [https://perma.cc/E8EW-CBBP]. Although it is increasingly common for AAA arbitrators to produce reasoned opinions rather than mere awards, [15][15]. The AAA discourages written opinions but permits parties to request them. See Charles A. Cooper, Harry Kaminsky & Neil Carmichael, Manual for Commercial Arbitration 162 (Am. Arbitration Ass’n 1999) (“Currently in domestic arbitrations, the AAA does not encourage commercial arbitrators to write opinions which give their reasons for the award. However, in instances where both parties request an opinion prior to the appointment of the arbitrator, the arbitrator should comply.”). While transactors might opt to revise a clause in face of a seemingly aberrant arbitration ruling, they might also surmise, much as the sovereign debt community did after the ruling of the Brussels court, that a more sensible panel (or as in the case of the pari passu clause the Second Circuit Court of Appeals) would rule differently. See Choi et al., supra note 1, at 12. these opinions are typically kept private and are of no precedential value. [16][16]. But see generally W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (suggesting that in some contexts arbitration decisions that are not technically precedent nevertheless have an effect on the outcome of future disputes). In addition, there are indications that companies fear that arbitral decisions will be given some weight if they have future disputes with the same counter-party. For example, Verizon’s consumer agreement provides for AAA commercial arbitration, includes a class action waiver, and also contains a clause stating that “[a]n arbitration award and any judgment confirming it apply only to that specific case; it can’t be used in any other case except to enforce the award itself.” Customer Agreement, Verizon Wireless (Oct. 20, 2017), https://www.verizonwireless.com/legal/notices/customer-agreement/ [https://perma.cc/3A2L-6YGT]; see also What’s Going on? Arbitration Explanation, Waste Mgmt., http://mediaroom.wm.com/legal-notice/ [https://perma.cc/TZ3C-X4GB] (“The award of the arbitrator may be entered and enforced in any court having jurisdiction. An arbitration award and any judgment confirming it apply only to that specific case; it cannot be used in any other case except to enforce the arbitration award itself.”). As a consequence, the terms of these contracts (whether standard or bespoke) are correspondingly less likely to be interpreted in written opinions with precedential value and are at a heightened risk of gradually losing their meaning over time. The routine inclusion of a AAA arbitration clause, therefore, increases the risk that some of an agreement’s provisions will devolve into black hole apparitions. [17][17]. For example, trade association-run private legal systems are designed in ways that will strongly discourage the emergence of black holes. At the National Grain and Feed Association, for example, arbitrators write reasoned opinions that are posted on the association’s website. When a rule is unclear, or appears to lead to an undesirable or unanticipated result, the arbitrators will state this explicitly, apply the rule as it is stated, and note that the association’s rules committee should consider revising the rule, something that is typically done. See Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1780 & nn.50¬–51 (1996).

Similarly, the standards of appellate review and the fact that most state and many federal trial courts do not routinely publish opinions [18][18]. Steven M. Barken, Barbara A. Bintliff & Mary Whisner, Fundamentals of Legal Research 41 (10th ed. 2015) (“Ordinarily, cases decided by state trial courts are not reported. . . . Only a few states, such as New York, Ohio, Pennsylvania, and Virginia, publish some trial court opinions, but those selected are few in number and represent only a very small portion of the total cases heard by trial courts.”). A similar problem may exist with respect to cases decided in federal court. See David A. Hoffman, Alan J. Izenman & Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash. U. L. Rev. 681, 727 (2007) (“[S]tarkly expos[ing] how little trial court work is explained through written opinions.”). might also play a role in the emergence of some types of contractual black holes. The vast majority of appellate courts provide greater deference to lower-court decisions on question of fact than on questions of law, or in some jurisdictions, mixed questions of law and fact. As a consequence, the meaning of standard-like provisions that are interpreted using a highly fact-specific inquiry is less likely to be appealed than the meaning of rule-like provisions whose interpretation is more often considered a matter of law. Standard-like provisions are correspondingly less likely to be interpreted in a written opinion of any kind. They are therefore more likely than rule-like provisions to devolve into meaningless provisions that look like classic black holes. [19][19]. It is, however, possible that the doctrinal contours of the black hole or no black hole determination would become clearer over time because under the proposed change this would be determined as a “matter of law.” See Choi et al., supra note 1, at 68.

B. Relational Apparitions

A second type of black hole apparition, which might be called a relational apparition, is particularly likely to emerge: (1) where there is a written or standard-form template that is used in most deals in a particular market; or (2) where seemingly bespoke contracts are used, [20][20]. CG&S focus mostly on standard commercial provisions in standardized contracts. However, both black holes and black hole apparitions can arise in seemingly bespoke contracts as well. To the extent this phenomenon exists, the proposed reform, in as much as it seems to apply only to standard contracts with boilerplate provision, would be underinclusive with respect to the types of holes found in these seemingly bespoke agreements. but those who negotiate them (either the business people or the contract lawyers) share a rough but deeply ingrained understanding about the types and content of terms that are commonly included in such agreements. In both of these situations, transactors may be especially hesitant to propose changes to commonly used terms. When a transactor proposes a change in the usual way of doing things, even when that usual way is not terribly important or when the variant creates value for both transactors, her counter-party might well interpret it as a signal that she is either more likely to engage in strategic behavior or more focused on her legal rights than the average transactor. The counter-party might therefore ask for protective changes in response. This, in turn, might trigger a cascade of change requests that could greatly increase contracting costs, reduce the likelihood of post-signing cooperation, [21][21]. The way that a contract is negotiated has an effect on how well the business deal can be operationalized once the contract is signed. See Danny Ertel, Getting Past Yes: Negotiating as if Implementation Mattered, 85 Harv. Bus. Rev. 60, 62 (2004). or destroy the deal completely. [22][22]. For a more complete discussion of this bargaining dynamic and the way that it effects the terms of agreements, see generally Lisa Bernstein, Social Norms & Default Rules Analysis, 3 S. Cal. Interdisc. L.J. 59 (1993). As long as transactors are aware of or intuit this dynamic, they might be reluctant to propose changing or eliminating common provisions, even if these provisions appear to be meaningless or irrelevant and standardization, across either the market as a whole or particular subsets of market participants, has no independent value.

Relational black hole apparitions may also be common in markets where transactors trust one another or where the force of reputation-based network governance is strong. In some markets where genuine interpersonal trust is present, it remains common for transactors to put the contract in the drawer. As a consequence, transactors will likely spend little or no time negotiating its terms. In these situations, contract language is quite likely to be unthinkingly recycled. [23][23]. Gillian K. Hadfield & Iva Bozovic, Scaffolding: Using Formal Contracts To Build Informal Relations to Support Innovation, 2016 Wis. L. Rev. 981, 982, 987 (noting that Stewart Macaulay’s finding that in commercial contracting relationships problems tended to be worked out informally and “written contracts . . . were often highly-standardized documents that were often confined to the drawer,” held in a modern day sample of non-innovative firms where informal methods of dispute resolution remained common and firms either “did not generate [formal contracts] or relied only on standardized documents,” but did not hold in a sample of innovative firms). Similarly, in markets where transactors rely on reputation-based network governance to support exchange, using similar contracts across the market makes these reputation-based forces stronger. [24][24]. See generally Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561 (2015) (explaining how widely known and standardized contract provisions may affect the strength of network governance). When contracts are relatively standardized and/or there is widespread consensus about what constitutes cooperation and good-faith behavior, [25][25]. In some markets where reputation-based network governance plays an important role in contracting, some large companies post their quality requirements and their standard terms and conditions on their website so that the contours of what is and is not expected from their counterparties is known throughout the market. See, e.g., Supplier Quality Manual, John Deere, https://jdsn.deere.com/wps/portal/jdsn/Applications?WCM_GLOBAL_CONTEXT=/wps/wcm/connect/jdsn_website/jdsn/business+processes/quality/supplier_quality_manual/supplier_quality_manual_index [https://perma.cc/57SG-BQUK]; Terms and Conditions for the Purchase of Goods and Services, John Deere, https://www.johndeerestore.com/jdb2cstorefront/JohnDeereStore/en/terms [https://perma.cc/B7DH-QPQU]. the network need only transmit information about how a transactor behaved in order for other market participants to assess the desirability of dealing with him. In contrast, when contracts are more varied, information about both the scope of the relevant obligations and actual behavior may need to be transmitted through the relevant network in order for reputation to play an important role in governing exchange—a process that is bound to be more expensive, less effective, and more error prone than in markets where relatively standard agreements are used. [26][26]. In such markets, general and impressionistic information about whether a transactor behaved properly or poorly could still be easily transmitted, but in the absence of standard terms, a consensus about what constitutes good behavior, or information about both what was promised and what was done, this information is likely to be given less weight (especially when it travels more than one step from its source) than information that circulates in markets where contracts and/or expectations are either relatively standardized or widely known. Bernstein, supra note 24, at 578–89 (discussing how John Deere achieves this). Since lawsuits are less common in markets where genuine trust is present or network governance is operational, fewer terms in commercial contracts are likely to be interpreted by courts and both black holes and black hole apparitions are correspondingly more likely to arise.

C. Rational Apparitions

A third and final type of black hole apparition, and one that may also emerge in both standard-form and bespoke contracts, is a rational black hole. A rational black hole is a term that is left deliberately vague or without meaning at the time of contracting. [27][27]. Sometimes the line between relational and rational black holes may be blurry. For example, Scott and Triantis note that it is often economically rational to use very standard-like terms like “good faith,” “reasonable efforts,” or “best efforts,” and leave it for the court to give these terms meaning ex post. Robert E. Scott & George Triantis, Anticipating Litigation in Contract Design, 115 Yale L.J. 814, 835 (2006). However, once those terms become common in an industry’s contracts (or between a pair of long-term repeat dealers), their presence in contracts may be exceptionally sticky. If, for example, someone sends their counterparty a draft obligating the counterparty (and themselves) to act in good faith, the relational costs of saying to one’s counter party, “well let’s spell that out” or “let’s leave that out” might be high. The economic literature has identified a variety of reasons that the inclusion of such terms might be rational. [28][28]. For a discussion of rational reasons why many clauses are left vague or undefined, see Scott & Triantis, supra note 27, at 814–15. The literature conceptualizes these terms as agreements-to-agree or agreements to determine meaning of a term through negotiation or litigation if an interpretive dispute arises. Among them: transactors believe that they will be better able to devise a response to a contingency when it actually arises, since additional facts about its implications will be known; transactors conclude that the probability a contingency will arise is so low (or the variety of related contingencies is so large) that it is not worth dealing with any or all of them ex ante; and finally, transactors have divergent views of the bargaining power they will each have when and if the issue arises, leading each to favor dealing with it later. In addition, in some contexts, clauses may be left meaningless or vague simply because transactors realize that in the event of a dispute over the meaning of the clause, the aggrieved party is unlikely to have a credible threat to sue for any of a number of reasons, including: litigation costs, the reputational cost of litigation, the information they would have to divulge in discovery, [29][29]. For a complete discussion of the ways that transactors’ “secrecy interest” in certain types of business information may affect the credibility of their threat to sue in a variety of contexts, see generally Omri Ben Shahar & Lisa Bernstein, The Secrecy Interest in Contract Law, 109 Yale L.J. 1885 (2000). or the fact that litigation would likely end their otherwise valuable contracting relationship. [30][30]. In contracting relationships like those between an Original Equipment Manufacturer and its suppliers of component parts where a lawsuit tends to end the parties contracting relationship, many contractual provisions that are formally enforceable may not operate in the shadow of the law at all. The reason is simple: if one of these provisions were breached in isolation, the promisee would not have a credible threat to sue since the expected recovery would be very unlikely to exceed the future value of the relationship as a whole. These “interior contract provisions” operate outside of the shadow of the law, unless and until they are breached in combination with enough other provisions that the counterparty concludes that there has been a breach of the contracting relationship that makes it worthwhile to end the transactors’ business relationship and sue. See Bernstein, supra note 24, at 570–71.

D. Conclusion

In sum, the existence of black hole apparitions complicates any attempt to deal with the black hole problem through changes in general contract doctrine that are contingent on a term’s status as a black hole. [31][31]. At one point in their analysis CG&S hint that it might be desirable to apply a knock-out rule to black hole terms akin to the rule adopted by some courts applying U.C.C. § 2-207. However, this rule will also require black holes to be identified by courts and will therefore be unworkable for many of the same reasons discussed herein. There are likely to be many contexts where black holes cannot be distinguished from black hole apparitions. In addition, depending on precisely how a black hole is defined, there are likely to be some instances where black holes exist yet the type of redrafting inertia that beset the sovereign debt market is unlikely to arise since the need for the type of market wide uniformity that is desirable when contracts are akin to financial instruments will not exist. As a consequence, any reform that conditions on a clause’s status as a black hole and applies to commercial contracts generally is likely to be overinclusive. By bringing commercial contracts dealing with the routine purchase and sale of goods and provision of services within its purview, any such reforms will increase litigation costs and create substantial social and private costs of their own.

Indeed, the pari passu black hole itself can be understood as arising from a mix of the procedural, relational, and rational considerations that give rise to black hole apparitions. The pari passu is partly a procedural black hole. Historically, the law of sovereign immunity made it extraordinarily difficult to sue or collect judgments [32][32]. See Sadie Blanchard, Courts Without Enforcement: Adjudicating Reputation in the Sovereign Debt Market 6–7 (Oct. 27, 2017) (unpublished manuscript) (concluding that “[e]ven though [the holders of sovereign debt] cannot enforce debtor states’ obligations through conventional judicial means, courts play a key role in the sovereign debt market . . . because they provide information that has the power to provoke reactions by third parties that are costly for the debtor or its agents,” so that “[c]reditors litigate because producing such information through the courts strengthens their leverage in settlement negotiations,” and makes reputation a more powerful force in the market). from other nations. [33][33]. See Julian Schumacher, Christoph Trebesch & Henrik Enderlein, Sovereign Defaults in Court 10–11 (May 6, 2014) (unpublished manuscript), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2189997 [https://perma.cc/T2Q7-LNGN], (finding that only 5 percent of sovereign defaults resulted in litigation until the 2000s, when the proportion rose to nearly half). Although in recent years it has become easier to sue a sovereign, [34][34]. See W. Mark C. Weidermaier, Sovereign Immunity and Sovereign Debt, 2014 U. Ill. L. Rev. 67, 68–70 (2014). many collection barriers remain. These legal and procedural barriers may be one reason why prior to a Brussels court’s 2000 decision in Elliott Associates v. Republic of Peru, [35][35]. Cours d’Appel de Bruxelles, 8th Chamber, 2000/QR/92, Elliott Assocs. v. République de Peru, 2000/QR/92, Sept. 26, 2000. there were no contemporary judicial decisions interpreting the meaning of pari passu provisions. The rote inclusion of the pari passu clause and the market’s slow reaction to the NML decision [36][36]. NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also looks at the market’s nonresponse to a pari passu decisions of a Brussels court, but attributes it to market players’ beliefs that other, more important courts for the sovereign debt community would not follow such an aberrant decision. can also be understood as the product of relational costs. As CG&S point out, as long as rating agencies did not downgrade bonds because the pari passu clause remained unchanged, individual issuers were wary of changing the clause and potentially triggering a negative rating-agency response. [37][37]. Choi et al., supra note 1, at 49. Finally, market players’ failure to revise the pari passu clause after the aberrant court judgments may or may not have been rational. [38][38]. As CG&S point out, it might not have been individually rational for any one issuer to make these changes to their past and future indentures, even though it would have been rational for the market as a whole to move to a different term. Id. at 13. However, the decision not to attend to its meaning more precisely at the time indentures were drafted may have been rational. After all, if the clause had not been interpreted in court for such a long period, the expected benefit of tinkering with it in a purposeful way might properly have been seen as small. Together, these considerations suggest that even on a purely conceptual level the line between a contract provision whose meaning has been eroded as a result of procedural, relational, and rational considerations and a true black hole is quite difficult to demarcate with analytic precision.

II. The Proposed Doctrinal Response

CG&S acknowledge that at present we have only a limited understanding about the frequency of black holes and the contexts in which courts’ erroneous interpretations of widely used terms will be met with drafting inertia. Nevertheless, they conclude that immediate legal reform to deal with the black hole problem is needed. In their view, the “vexing collective action problems” that market participants face when attempting to respond to errant court interpretations are so significant that the potential social costs of not dealing with the black hole problem are too significant to ignore. [39][39]. Id. at 37 (“We use both qualitative and quantitative data to support the claim that courts searching for shared intent in the case of black holes in standardized contracting can result in substantial social costs.”).

CG&S suggest that the best doctrinal way to deal with the black hole problem is for courts interpreting “boilerplate terms in commercial contracts” [40][40]. Id. at 67. to be “open to arguments that as a matter of law, the clause in question has been emptied of meaning and functions as a black hole in boilerplate.” [41][41]. Id. at 68. In cases in which the parties meet the burden of establishing the existence of a black hole, the court would then be required to adopt an interpretive approach that avoids any inquiry into subjective intent and looks to other interpretive principles to decide the case. [42][42]. See id. at 66–67, 69 Exactly what this alternative interpretive approach would be is not specified in great detail. See id. at 54–56. If, instead, some form of knock-out rule were contemplated, similar problems would arise. See supra note 31. Under the proposed approach, the hole or non-hole determination would turn on a variety of highly fact-specific considerations and types of evidence. These include the types of evidence adduced in the pari passu case study, [43][43]. Choi et al., supra note 1, at 38. as well as the types of information needed to answer the following questions:

Has the clause been repeated by rote over many years, without having been tested in litigation, where repetition has robbed the term of any obvious conventional meaning? Has the term been embedded in layers of legal jargon such that its intelligibility is substantially reduced and variations in the formulation of the term across contracts have no apparent significance? Is a historic or original meaning of the term accessible in a fashion that makes sense in the contemporary context and are contemporary commercial actors aware of that meaning? Is there credible evidence that the particular provision was priced at the original issue stage? [44][44]. Id. at 68.

Notwithstanding its theoretical appeal, the implementation of this approach faces significant practical challenges. First, the test will be widely overinclusive even with respect to standard provisions in standard-form contracts because answers to these questions will not reliably distinguish black holes from black hole apparitions. [45][45]. Indeed, the methodology CG&S used to reach the conclusion that the pari passu provision was a black hole, rather than an efficient contract term, relied on extensive information about the market’s response to the decision over a period of years. See id. at 11. Moreover, because the new test applies to all standard commercial agreements, it gives a significant advantage to the party with deeper pockets who might claim a black hole exists simply to increase putative litigation costs and thereby lever a better settlement.

Second, and relatedly, as CG&S themselves acknowledge, introducing this inquiry would open the door to moral hazard and other types of strategic behavior. [46][46]. The types of strategic behavior it would introduce are analogous to those identified in Alan Schwartz and Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 587 (2000) and Lisa Bernstein, Custom in the Courts, 110 Nw. L. Rev. 63, 106–08 (2015). Although they suggest that moral hazard can be adequately dealt with by adopting “an initial presumption against the existence of a black hole,” [47][47]. Choi et al., supra note 1, at 68. even if they are correct the risk of strategic behavior will remain substantial.

Third, even abstracting from overinclusiveness and strategic behavior, as a practical matter the type of information CG&S view as relevant to the black hole or non-black hole determination, will often be either unavailable to the parties or prohibitively costly to obtain. [48][48]. Id. at 55–56, 64. For example, in contexts where the relevant contracts are not disclosed under the securities laws, transactors’ reluctance to share this information will make it impossible to get the contracts used in similar transactions—an evidentiary problem that will prevent courts from answering many of the questions that CG&S view as directly relevant to the hole or no-hole determination. Similarly, there is no obvious way to determine whether a particular clause has been priced into a contract. In many settings, price and other terms are negotiated separately, and even when they are not, determining the connection between price and a particular term may be difficult either because the transactors did not explicitly think about the connection or because reconstructing the sequence of a negotiation from the testimony of witnesses with conflicting agendas is likely to be an error-fraught process. In addition, market players, even those who willingly spoke to academics, might be reluctant to talk to litigators or testify in court. Thus, even if in theory the hole or no-hole determination could be accurately made on the basis of the types of evidence that CG&S view as relevant, in practice much of this information is likely to be either prohibitively expensive or entirely unavailable.

Fourth, while CG&S want the courts to intervene when “parties exploit” either the absence of meaning or the presence of “random variations in language [to advance] an interpretation the market disavows,” [49][49]. Id. at 1. they offer little guidance on how random variants can be distinguished from what they call “rational design,” [50][50]. Choi et al., Variation in Boilerplate, supra note 2, at 4. a determination that would likely create both evidentiary problems and additional interpretive uncertainty. [51][51]. CG&S undertake just this inquiry in another part of the black hole corpus. See id. at 6. There, the authors seek to demonstrate that the observed variations in the pari passu clauses used in sovereign debt contracts are the result of random mutation rather than rational design. The information and analysis they present is a good proxy for how difficult it would be for a transactor to litigate this distinction in court. Moreover, not all of the information they refer to could be known at trial—for example, how the market would respond if the court gave weight to a small drafting difference.

Finally, without a more detailed description of the interpretive rule that the court would apply when a hole is found and a better sense of the extent to which the chosen rule would result in predictable outcomes, it is difficult to asses how frequently the disputing parties will have the proper incentives to take advantage of the proposed reform and argue that black holes exist, even in contexts where they both believe that a true black hole actually exists.

Consider, for example, a context in which both parties think that the best subjective interpretation of a clause that they can advance is better than the expected interpretation they think a court would give the clause if a black hole were found to exist. [52][52]. When real black holes actually exist, parties are likely to be largely unconstrained with respect to the range of plausible subjective intents they can proffer. As a consequence, the assumptions made in the text are likely to characterize a meaningful subset of the cases that might arise. In such a situation, both parties might opt not to argue for a hole’s existence even if they both genuinely believe that a hole exists. To see why, consider a plaintiff who is trying to decide whether to plead that a hole exists or to advance her own subjective interpretation of a clause. If she pleads that a hole exists, the defendant faces a strategic choice. If he pleads that a hole exists, then the court will likely find that a hole exists, deem the subjective intent of the parties irrelevant, and interpret the clause using its own interpretive principles. In contrast, if the defendant pleads his subjective interpretation, the court will either accept his subjective interpretation or conclude that a hole exists, deem subjective intent irrelevant, and decide the case using its own interpretive principles. Because (by assumption) both parties believe that their subjective interpretation is better than the court’s interpretation, if the plaintiff pleads that a hole exists, the defendant is better off pleading that his subjective interpretation should govern. This response creates at least some chance that the court will accept his preferred subjective interpretation. Alternatively, if the plaintiff pleads her own subjective interpretation of the clause, the defendant will reason that he is always better off pleading his subjective interpretation than claiming that a hole exists. If he argues for his subjective interpretation, there is at least some possibility that the court will accept his interpretation, whereas if he pleads a hole, the court will find either find that a hole exists or accept the plaintiff’s subjective interpretation. Thus, regardless of the plaintiff’s choice, the defendant is always better off arguing for his subjective interpretation of the clause. Given this, the plaintiff too is better off arguing for her own subjective interpretation of the clause, since doing so creates at least some chance that her interpretation will be accepted by the court. As a consequence, at least in the run of situations where both parties think that the best subjective interpretation of the clause they can advance is better than the expected interpretation the court would give the clause, they are unlikely to avail themselves of the proposed doctrinal step even when a hole exists. Indeed, in the NML case itself neither party asserted that the pari passu clause was without meaning. Rather each side advanced an argument that it meant something different. [53][53]. See, e.g., Brief of Defendant-Appellant the Republic of Argentina, NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2012 WL 6777133; Joint Response Brief of Plaintiffs-Appellees NML Capital, LTD and Olifant Fund, LTD, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2013 WL 388621.

A. Other Directions for Reform

In sum, proposals to alter contract doctrine to respond to the black hole problem may be difficult to implement in practice. As a consequence, to the extent that the problems revealed in the pari passu saga are wide-spread, it is worthwhile to explore whether there are approaches that do not rely on the hole-or-no-hole distinction that could prove to be a more workable response to the black hole problem.

If, for example, the inertia problem were found to exist more broadly in particular contexts—such as markets where the contracts are closer to the pole of financial instruments than ordinary contracts for services or the sale of goods [54][54]. CG&S are careful to delineate a number of reasons that the collective action problems they observe in the sovereign debt markets where the contracts at issue—bond indentures—are in practice closer to the pole of financial instruments, may be particularly severe compared to the collective action problems that might or might not impede the revision of ordinary contracts for sales of goods or provision of services when a court makes an aberrant interpretation of one of their terms. See Choi et al., supra note 1, at 59–66.—a separate article of the UCC could be added that would apply different interpretive standards as a matter of course. [55][55]. The reform could take the form of a new sub-chapter of Article 2 akin to Article 2A on leases, or an additional subset of rules in Article 2 akin to its merchant rules. This change would have the advantage of eliminating the costly and uncertain black hole inquiry. [56][56]. While such a change would raise the issue of how to decide which contracts would come within the new Article’s purview, it could be made available to the parties on an opt-in basis at the time of contracting, subject to rejection by a court that found the choice to be inappropriate. Nevertheless, given the well-known problems that impede the American Law Institute (ALI) and National Conference of Commissioners on Uniform State Laws (NCCUSL) rules creation processes, such an approach, however desirable, might also prove infeasible. [57][57]. For an overview of the problems with these private uniform law-making bodies, see generally Alan Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. Pa. L. Rev., 595–654 (1995). For an example of the difficulties of creating new Articles of the UCC in particular, consider the largely failed effort to get states to adopt the Uniform Computer Information Transactions Act, which has been adopted by only two states. See The Uniform Computer Information Transactions Act (UCITA) Is a Proposed State Contract Law, UCITA Online, http://www.ucitaonline.com/ [https://perma.cc/8SUP-W2L8] (discussing the impediments to adopting the model law).

Alternatively, if the black hole problem turns out to be widespread in certain types of well-defined markets or types of contracts, eliminating any inquiry into subjective intent in those contexts would achieve most of the benefits of the proposed black hole reform without increasing litigation costs or the frequency of strategic behavior. Black holes aside, eschewing the legal fiction of subjective intent in the interpretation of commercial contracts and adopting a largely textualist plain-meaning oriented interpretive approach is a reform that would likely prove highly beneficial for sophisticated commercial transactors [58][58]. The best available, though imperfect, empirical evidence suggests that sophisticated commercial parties prefer textualist adjudication. See Lisa Bernstein, Merchant Law in a Modern Economy, in Philosophical Foundations of Contract Law 251 (Gregory Klass, George Letsas & Prince Saprai eds., Oxford Univ. Press 2014) (noting the common use of plain meaning clauses in large commercial contracts providing for arbitration); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001) (documenting the cotton industry’s preference for a textualist/formalist adjudicative approach); see also Bernstein, Merchant Law in a Modern Economy, supra (documenting the grain and feed industry’s preference for a textualist/formalist legal approach); Theodore Eisenberg & Geoffrey P. Miller, The Flight to New York: An Empirical Study of Choice of Law and Choice of Forum Clauses in Publicly-Held Companies’ Contracts, 30 Cardozo L. Rev. 1475 (2009) (demonstrating commercial actors’ strong preferences for relatively formalist New York Law); Stuart Popham, The View of European Business: Survey Results Presentation, Clifford Chance LLP (Mar. 14, 2008), http://denning.law.ox.ac.uk/news/events_files/Popham_-_presentation.PPT [https://perma.cc/DRB3-KXAL] (finding that in business contracts that provided for arbitration at the International Chamber of Commerce, transactors preferred British law, the most formalistic and textualist of the EU alternatives). for reasons that have long been advanced by the neo-formalist school of contract interpretation. [59][59]. See generally Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 Yale L.J. 926 (2010) (defending a formalist approach to contract interpretation).

Indeed, the neo-formalist approach reflected in the Second Circuit’s NML decision can be understood as taking aim at one of the causes of black holes—namely, encrustation. According to CG&S, encrustation is the end result of the meaningless drafting changes that lawyers introduce either without thinking or to show their clients they are doing something. By signaling that the court may well attach radically different meanings to clauses with small wording differences, the approach taken by the NML court does two key things: it increases the cost of meaningless amendments and should therefore discourage them; and it increases the return to meaningful amendments since the court will take them into account in interpreting the clause which should, in turn, encourage meaningful drafting. It can therefore be understood as an interpretive penalty default rule that should over time change drafting behavior in ways that may reduce the amount of meaningless encrustation in commercial agreements and with it the likelihood that black holes will emerge.

Nevertheless, despite the potentially significant problems that black holes may create and the many possible directions legal reform might take, [60][60]. Another potential direction for reform that could be implemented without changing the common law, but that depends critically on the ability of groups—like state legislatures, the ALI, the NCCUSL, trade associations, and a variety of private entities and intermediaries—to revise contractual language when courts make errant decisions, would be to selectively adopt the reform proposals of the legal choice theorists. See Hanoch Dagan & Michael Heller, The Choice Theory of Contracts 102–13 (Cambridge Univ. Press 2017). These theorists explore the benefits that might accrue when states and other private and public entities provide more “contract types” that transactors could use to structure their affairs—much as Delaware law offers those structuring a business the option of incorporating as a public entity, a close corporation, an LLC or a partnership. If these “types” were carefully developed, and proved easier to change in the event of an errant judicial interpretation of their terms than individually structured agreements, heeding the call for the creation of these types may help solve the black hole problem in some contracting contexts. See infra notes 61–65 and accompanying text (noting that trade associations have proved remarkably able to amend their trading rules and standard form contracts in response to arbitral interpretations of these terms). But see Schwartz & Scott, infra note 59 (discussing the problems that groups like the ALI have in revising their model laws). the current understanding of black holes may be far too limited to devise sufficiently nuanced reforms. As discussed further below, there are reasons to question whether the saga of the pari passu, which is the empirical foundation of CG&S’s call for immediate reform, is, on its own, sufficient to justify the need for an immediate and generally applicable legal response, especially given the many differences—differences CG&S carefully delineate—between bond indentures and the other types of contracts the proposed reform would govern.

III. Is Immediate Reform Needed?

Given the many difficulties that stand in the way of a legal solution to the black hole problem, it is important to explore whether legal reform is really needed. The answer to this question turns, in large part, on whether the inertia in revising contract terms after an aberrant decision interpreting a black hole is likely to be a frequent occurrence across a variety of markets. It also depends on whether those black holes that do emerge are likely to be enduring phenomena or a transitory stage in the development of the relevant market.

There is no systematic empirical evidence about the frequency of black holes. However, the experiences of American trade associations that created standard form contracts and trading rules to govern transactions among their members suggest that the inertia costs that stem from collective action problems and impede the adoption of new contract provisions, may decrease over time as industry participants and/or market institutions learn how to more quickly respond to the need for market-wide contractual change.

In the textile industry, for example, the first edition of The Worth Street Textile Trading Rules, [61][61]. See Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710, 730–32 (1999) (providing a short history of the Textile Trading Rules). which consisted of both a set of trading rules and a standard textile sales note, took eighteen years of committee work to draft. The rules creation process “was fraught with conflict, [and] involved negotiations among numerous trade associations.” [62][62]. Id. at 732. Yet once adopted, the rules were revised and even entirely rewritten numerous times far more quickly and with a great deal less infighting. Similarly, when the Silk Association of America set out to create a set of trading rules for raw silk, many controversies arose. The rules-creation process took seven years to complete because “[t]he get-together spirit was not sufficiently pronounced to override the differences that arose . . . .” [63][63]. Id. at 736 (quoting SAA, Thirty-Fifth Annual Report 23 (1907)). Nevertheless, despite these initial difficulties, the Raw Silk Trading Rules, like the Worth Street Rules, were subsequently amended and revised many times with far less difficulty. A 1921 amendment, for example, was adopted after only “a year of careful study on the part of the [rules] Committee,” [64][64]. Id. (quoting SAA, Fortieth Annual Report 31 (1912), Revision, 3 Silkworm 73 (May 1921)). and a 1924 amendment was adopted after mere “months of intensive effort.” [65][65]. Id. (quoting SAA, Fifty-Third Annual Report 26 (1925)). Similarly, the National Grain and Feed Association’s effort to adopt the first set of Grain Trading Rules was also hard-fought and filled with disagreement. Yet since their adoption, these rules have been successfully amended seventy-seven times, also with little infighting and only rare controversy. The experience of these and other industries suggests that while the initial costs of agreeing on standard contract terms and trading rules may be high, over time these costs tend to gradually decrease as market players and institutions get used to working with one another to respond to technological changes, market changes, and other types of disruptions that require market-wide changes in contractual rules or forms.

The experience of these and other groups is far from determinative; yet it provides a reason to be cautiously optimistic that the sovereign debt community will be able to respond more quickly in the future to any adverse court decisions that might once again threaten to impose large social costs on their market. There are even indications that this may already be happening. [66][66]. For a detailed discussion of the institutional response to the pari passu problem which suggests that institutions “learn” each time they attempt responses, see Anna Gelpern, Ben Heller & Brad Setser, Count the Limbs: Designing Robust Aggregation Clauses in Sovereign Bonds, in Too Little, Too Late: The Quest To Resolve Sovereign Debt Crises (Martin Guzman, José Antonio Ocampo & Joseph E. Stiglitz eds., Columbia Univ. Press 2016). The pari passu case study reveals that when key market players first met at Columbia Law School to discuss the NML court’s interpretation of the pari passu clause in Argentine debt, the meeting was marked by so much “dissension and disagreement” [67][67]. Choi et al., supra note 1, at 40. that the prospect of “any significant movement towards wholesale revision of the clause [appeared] unlikely in the near term.” [68][68]. Id. at 39. Nevertheless, when the Federal Reserve convened a meeting of many of the very same people just a few weeks later, it became clear that “everyone involved needed, and was willing to, cooperate in trying to solve the systemic problem caused by the rogue interpretation of the New York courts.” [69][69]. Id. at 40. Legal change accelerated shortly thereafter. [70][70]. However, as CG&S point out, these revisions occurred only in sovereign debt—not quasi-sovereign debt—which in their view left the market facing significant social costs. Id. at 24–25.

The experience of American trade associations and the sovereign debt community’s own response when they came together at the Federal Reserve suggest that prudence is warranted in abstracting from the saga of the pari passu. It is far from certain that any future errant court decisions about the meaning of black holes in sovereign debt indentures will remain unremedied by collective action for a significant period of time. Only time and additional research on this and other markets will reveal whether the saga of the pari passu is best understood as: (1) an illustration of the transition problems that particular types of markets face when they are confronted with the need to overcome collective action obstacles to contractual change for the first time; (2) an enduring feature of markets where contracts are inter-dependent in ways that make the risk of judicial error more serious than in other markets; or (3) a problem that exists in a wider array of contracting contexts than standard contract theory would predict.

Conclusion

This Comment has explored some of the conceptual and practical problems that make it difficult to devise an effective legal strategy for dealing with black holes. It has also questioned whether or not a legal response to the black hole problem is needed. Still, nothing in the discussion has taken anything away from the core contribution of the black hole articles—namely, the identification of a distinct type of contract provision that poses unique interpretive challenges and may increase the consequences of judicial error in some contracting contexts. This contribution is particularly timely because black holes are likely to become increasingly common as a result of a number of technological changes that facilitate the creation of complex agreements with largely standardized yet slightly variant terms. [71][71]. See, e.g., KMStandards (2016), KMstandards.com [https://perma.cc/5KTY-Z2NJ]; IACCM (2017), IACCM.com [https://perma.cc/YH8G-8D73] (providing contract drafting software that permits the quick and inexpensive drafting of largely standard form contracts while giving transactors a variety of clauses to choose from), the website of the International Association of Contract and Commercial Managers [https://perma.cc/KKP7-3SMZ] (containing information about new contract drafting technologies).

Nevertheless, until more is known about black holes in both the sovereign debt market and other contexts it might be best to refrain from undertaking any general doctrinal legal reforms in response to the phenomenon, especially as the type and magnitude of the problems caused by black holes are likely to be vastly different depending on the market in question.

Moreover, the cost of not responding to the problem may be far lower than it seems. The interpretive approach adopted by the Second Circuit in NML v. Argentina may turn out to be the best or at least a passably good long-run common law response to the black hole problem—especially in markets where the players (individuals, governments, and institutions) are sophisticated and are advised by able counsel. By functioning as an interpretive penalty default rule, [72][72]. Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 91–130 (1989) (describing the theory of the penalty default rule). the Second Circuit’s adjudicative approach creates incentives for transactors and market intermediaries like rating agencies to pay more attention to drafting choices. It may also have the beneficial effect of encouraging new or existing market or quasi-governmental intermediaries to develop the institutional frameworks needed to provide the types of tailored responses to contracting problems that the common law cannot provide without introducing changes that will be vastly over- or underinclusive and will also be likely to increase both litigation costs and the incidence of strategic behavior.

Interestingly, while CG&S remain critical of the NML decision, they now acknowledge that some type of institutional response (whether induced by the common law or encouraged by other types of government action) may, at the end of the day, be the best response to the black hole problem in the long-run. As they explain in reference to the interchange of ideas that led to this Comment:

[T]he true lesson of our study may be that the IMF and other groups that constitute the ‘official sector’ may be better able than courts to solve these problems over time as they gain experience and become more confident in their methods. If this is so, then even though the pari passu case shows that the intervention of a public authority is sometimes required to solve contractual disputes that have third-party effects, relying on the courts rather than private ordering to craft the solution may not always be the best choice. [73][73]. Choi et al., supra note 1, at 69–70 (citing discussions with Bernstein).

Copyright © 2017 Lisa Bernstein.

Wilson-Dickinson Professor of Law and Aaron Director Research Scholar, The University of Chicago Law School, and International Research Fellow, Center for Corporate Reputation, Said School of Business, University of Oxford. I would like to thank Douglas Baird, Patrick Barry, Edward Bernstein, Brian Bix, Sadie Blanchard, Robin Effron, Roger Ford, Anna Gelpern, William Hubbard, Todd Ito, Avery Katz, Kate Lindgren, Bill Schwesig, Mark Ramseyer, Steve Ware, and participants at the Duke Conference on Contractual Black Holes (2017) for helpful conversations. Special thanks are due to the hole theorists themselves for engaging conversations about their piece over the past year.

[1] Stephen J. Choi, G. Mitu Gulati & Robert E. Scott, The Black Hole Problem in Commercial Boilerplate, 67 Duke L.J. 1 (2017).

[2] The other two papers in the Black Hole corpus are Stephen J. Choi, Mitu Gulati, & Robert E. Scott, Contractual Arbitrage, in Oxford Handbook of International Governance (Eric Brosseau, Jean-Michel Galmant & Jerome Sgaard eds., forthcoming 2017) http://scholarship.law.duke.edu/cgi/ viewcontent.cgi?article=6304&context=faculty_scholarship [https://perma.cc/VC3Y-QLKN] and Stephen J. Choi, Mitu Gulati & Robert E. Scott, Variation in Boilerplate: Rational Design or Random Mutation, 19 AM. L. & ECON. REV. (forthcoming 2017) [hereinafter, Choi et al., Variation in Boilerplate].

[3] Choi et al., supra note 1, at 2.

[4] Id. at 6.

[5] Id.

[6] Id.

[7] NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also focuses on an earlier interpretation of the pari passu clause rendered by a Brussels court in 2000. See Elliott Assocs. v. Republic of Peru, 948 F. Supp. 1203 (S.D.N.Y. 1996).

[8] Choi et al., supra note 1, at 7.

[9] Interestingly, the Second Circuit seemed to reject the idea that it was jeopardizing the market as a whole by making restructuring more difficult. See NML Capital, Ltd., 699 F.3d at 263 (explaining why this effect is unlikely to occur.)

[10] Choi et al., supra note 1, at 70 (concluding that “the inherently greater inertia costs [and social costs] that result from an aberrant interpretation of a black hole term,” as compared to an ordinary unclear term, and the “greater difficulty market players face in overcoming the resulting collective action problem[s]” slow the adoption of new and better contractual terms).

[11] Id.

[12] Although it is possible that some of the types of clauses this Comment views as “black hole apparitions” would be classified as genuine black holes by CG&S, their search for additional examples of black holes by holding a conference on the subject at Duke Law School suggests that these ubiquitous types of clauses with uncertain meaning are unlikely to come within the purview of their definition. As Gulati explained, “black holes where meaning has been lost are different from provisions where there is a general sense of what they mean, but there is some vagueness or lack of clarity.” E-mail from Mitu Gulati, Professor of Law, Duke University, to author (Sept. 23, 2017) (on file with author).

[13] In order for a doctrine that conditions on whether a clause is or is not a black hole to be workable, a clause’s status must be verifiable by a court—that is, a court must be able to determine with reasonable accuracy at a cost the transactors consider reasonable from an ex ante perspective whether or not something is a black hole. See generally Alan Schwartz, Relational Contracts and the Courts: An Analysis of Incomplete Agreements and Judicial Strategies, 21 J. Legal Stud. 271 (1992).

[14] Commercial Arbitration Rules and Mediation Procedures (Am. Arbitration Ass’n 2013), https://www.adr.org/sites/default/files/Commercial%20Rules.pdf. [https://perma.cc/E8EW-CBBP].

[15] The AAA discourages written opinions but permits parties to request them. See Charles A. Cooper, Harry Kaminsky & Neil Carmichael, Manual for Commercial Arbitration 162 (Am. Arbitration Ass’n 1999) (“Currently in domestic arbitrations, the AAA does not encourage commercial arbitrators to write opinions which give their reasons for the award. However, in instances where both parties request an opinion prior to the appointment of the arbitrator, the arbitrator should comply.”). While transactors might opt to revise a clause in face of a seemingly aberrant arbitration ruling, they might also surmise, much as the sovereign debt community did after the ruling of the Brussels court, that a more sensible panel (or as in the case of the pari passu clause the Second Circuit Court of Appeals) would rule differently. See Choi et al., supra note 1, at 12.

[16] But see generally W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (suggesting that in some contexts arbitration decisions that are not technically precedent nevertheless have an effect on the outcome of future disputes). In addition, there are indications that companies fear that arbitral decisions will be given some weight if they have future disputes with the same counter-party. For example, Verizon’s consumer agreement provides for AAA commercial arbitration, includes a class action waiver, and also contains a clause stating that “[a]n arbitration award and any judgment confirming it apply only to that specific case; it can’t be used in any other case except to enforce the award itself.” Customer Agreement, Verizon Wireless (Oct. 20, 2017), https://www.verizonwireless.com/legal/notices/customer-agreement/ [https://perma.cc/3A2L-6YGT]; see also What’s Going on? Arbitration Explanation, Waste Mgmt., http://mediaroom.wm.com/legal-notice/ [https://perma.cc/TZ3C-X4GB] (“The award of the arbitrator may be entered and enforced in any court having jurisdiction. An arbitration award and any judgment confirming it apply only to that specific case; it cannot be used in any other case except to enforce the arbitration award itself.”).

[17] For example, trade association-run private legal systems are designed in ways that will strongly discourage the emergence of black holes. At the National Grain and Feed Association, for example, arbitrators write reasoned opinions that are posted on the association’s website. When a rule is unclear, or appears to lead to an undesirable or unanticipated result, the arbitrators will state this explicitly, apply the rule as it is stated, and note that the association’s rules committee should consider revising the rule, something that is typically done. See Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1780 & nn.50¬–51 (1996).

[18] Steven M. Barken, Barbara A. Bintliff & Mary Whisner, Fundamentals of Legal Research 41 (10th ed. 2015) (“Ordinarily, cases decided by state trial courts are not reported. . . . Only a few states, such as New York, Ohio, Pennsylvania, and Virginia, publish some trial court opinions, but those selected are few in number and represent only a very small portion of the total cases heard by trial courts.”). A similar problem may exist with respect to cases decided in federal court. See David A. Hoffman, Alan J. Izenman & Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash. U. L. Rev. 681, 727 (2007) (“[S]tarkly expos[ing] how little trial court work is explained through written opinions.”).

[19] It is, however, possible that the doctrinal contours of the black hole or no black hole determination would become clearer over time because under the proposed change this would be determined as a “matter of law.” See Choi et al., supra note 1, at 68.

[20] CG&S focus mostly on standard commercial provisions in standardized contracts. However, both black holes and black hole apparitions can arise in seemingly bespoke contracts as well. To the extent this phenomenon exists, the proposed reform, in as much as it seems to apply only to standard contracts with boilerplate provision, would be underinclusive with respect to the types of holes found in these seemingly bespoke agreements.

[21] The way that a contract is negotiated has an effect on how well the business deal can be operationalized once the contract is signed. See Danny Ertel, Getting Past Yes: Negotiating as if Implementation Mattered, 85 Harv. Bus. Rev. 60, 62 (2004).

[22] For a more complete discussion of this bargaining dynamic and the way that it effects the terms of agreements, see generally Lisa Bernstein, Social Norms & Default Rules Analysis, 3 S. Cal. Interdisc. L.J. 59 (1993).

[23] Gillian K. Hadfield & Iva Bozovic, Scaffolding: Using Formal Contracts To Build Informal Relations to Support Innovation, 2016 Wis. L. Rev. 981, 982, 987 (noting that Stewart Macaulay’s finding that in commercial contracting relationships problems tended to be worked out informally and “written contracts . . . were often highly-standardized documents that were often confined to the drawer,” held in a modern day sample of non-innovative firms where informal methods of dispute resolution remained common and firms either “did not generate [formal contracts] or relied only on standardized documents,” but did not hold in a sample of innovative firms).

[24] See generally Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561 (2015) (explaining how widely known and standardized contract provisions may affect the strength of network governance).

[25] In some markets where reputation-based network governance plays an important role in contracting, some large companies post their quality requirements and their standard terms and conditions on their website so that the contours of what is and is not expected from their counterparties is known throughout the market. See, e.g., Supplier Quality Manual, John Deere, https://jdsn.deere.com/wps/portal/jdsn/Applications?WCM_GLOBAL_CONTEXT=/wps/wcm/connect/jdsn_website/jdsn/business+processes/quality/supplier_quality_manual/supplier_quality_manual_index [https://perma.cc/57SG-BQUK]; Terms and Conditions for the Purchase of Goods and Services, John Deere, https://www.johndeerestore.com/jdb2cstorefront/JohnDeereStore/en/terms [https://perma.cc/B7DH-QPQU].

[26] In such markets, general and impressionistic information about whether a transactor behaved properly or poorly could still be easily transmitted, but in the absence of standard terms, a consensus about what constitutes good behavior, or information about both what was promised and what was done, this information is likely to be given less weight (especially when it travels more than one step from its source) than information that circulates in markets where contracts and/or expectations are either relatively standardized or widely known. Bernstein, supra note 24, at 578–89 (discussing how John Deere achieves this).

[27] Sometimes the line between relational and rational black holes may be blurry. For example, Scott and Triantis note that it is often economically rational to use very standard-like terms like “good faith,” “reasonable efforts,” or “best efforts,” and leave it for the court to give these terms meaning ex post. Robert E. Scott & George Triantis, Anticipating Litigation in Contract Design, 115 Yale L.J. 814, 835 (2006). However, once those terms become common in an industry’s contracts (or between a pair of long-term repeat dealers), their presence in contracts may be exceptionally sticky. If, for example, someone sends their counterparty a draft obligating the counterparty (and themselves) to act in good faith, the relational costs of saying to one’s counter party, “well let’s spell that out” or “let’s leave that out” might be high.

[28] For a discussion of rational reasons why many clauses are left vague or undefined, see Scott & Triantis, supra note 27, at 814–15. The literature conceptualizes these terms as agreements-to-agree or agreements to determine meaning of a term through negotiation or litigation if an interpretive dispute arises.

[29] For a complete discussion of the ways that transactors’ “secrecy interest” in certain types of business information may affect the credibility of their threat to sue in a variety of contexts, see generally Omri Ben Shahar & Lisa Bernstein, The Secrecy Interest in Contract Law, 109 Yale L.J. 1885 (2000).

[30] In contracting relationships like those between an Original Equipment Manufacturer and its suppliers of component parts where a lawsuit tends to end the parties contracting relationship, many contractual provisions that are formally enforceable may not operate in the shadow of the law at all. The reason is simple: if one of these provisions were breached in isolation, the promisee would not have a credible threat to sue since the expected recovery would be very unlikely to exceed the future value of the relationship as a whole. These “interior contract provisions” operate outside of the shadow of the law, unless and until they are breached in combination with enough other provisions that the counterparty concludes that there has been a breach of the contracting relationship that makes it worthwhile to end the transactors’ business relationship and sue. See Bernstein, supra note 24, at 570–71.

[31] At one point in their analysis CG&S hint that it might be desirable to apply a knock-out rule to black hole terms akin to the rule adopted by some courts applying U.C.C. § 2-207. However, this rule will also require black holes to be identified by courts and will therefore be unworkable for many of the same reasons discussed herein.

[32] See Sadie Blanchard, Courts Without Enforcement: Adjudicating Reputation in the Sovereign Debt Market 6–7 (Oct. 27, 2017) (unpublished manuscript) (concluding that “[e]ven though [the holders of sovereign debt] cannot enforce debtor states’ obligations through conventional judicial means, courts play a key role in the sovereign debt market . . . because they provide information that has the power to provoke reactions by third parties that are costly for the debtor or its agents,” so that “[c]reditors litigate because producing such information through the courts strengthens their leverage in settlement negotiations,” and makes reputation a more powerful force in the market).

[33] See Julian Schumacher, Christoph Trebesch & Henrik Enderlein, Sovereign Defaults in Court 10–11 (May 6, 2014) (unpublished manuscript), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2189997 [https://perma.cc/T2Q7-LNGN], (finding that only 5 percent of sovereign defaults resulted in litigation until the 2000s, when the proportion rose to nearly half).

[34] See W. Mark C. Weidermaier, Sovereign Immunity and Sovereign Debt, 2014 U. Ill. L. Rev. 67, 68–70 (2014).

[35] Cours d’Appel de Bruxelles, 8th Chamber, 2000/QR/92, Elliott Assocs. v. République de Peru, 2000/QR/92, Sept. 26, 2000.

[36] NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also looks at the market’s nonresponse to a pari passu decisions of a Brussels court, but attributes it to market players’ beliefs that other, more important courts for the sovereign debt community would not follow such an aberrant decision.

[37] Choi et al., supra note 1, at 49.

[38] As CG&S point out, it might not have been individually rational for any one issuer to make these changes to their past and future indentures, even though it would have been rational for the market as a whole to move to a different term. Id. at 13.

[39] Id. at 37 (“We use both qualitative and quantitative data to support the claim that courts searching for shared intent in the case of black holes in standardized contracting can result in substantial social costs.”).

[40] Id. at 67.

[41] Id. at 68.

[42] See id. at 66–67, 69 Exactly what this alternative interpretive approach would be is not specified in great detail. See id. at 54–56. If, instead, some form of knock-out rule were contemplated, similar problems would arise. See supra note 31.

[43] Choi et al., supra note 1, at 38.

[44] Id. at 68.

[45] Indeed, the methodology CG&S used to reach the conclusion that the pari passu provision was a black hole, rather than an efficient contract term, relied on extensive information about the market’s response to the decision over a period of years. See id. at 11.

[46] The types of strategic behavior it would introduce are analogous to those identified in Alan Schwartz and Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 587 (2000) and Lisa Bernstein, Custom in the Courts, 110 Nw. L. Rev. 63, 106–08 (2015).

[47] Choi et al., supra note 1, at 68.

[48] Id. at 55–56, 64.

[49] Id. at 1.

[50] Choi et al., Variation in Boilerplate, supra note 2, at 4.

[51] CG&S undertake just this inquiry in another part of the black hole corpus. See id. at 6. There, the authors seek to demonstrate that the observed variations in the pari passu clauses used in sovereign debt contracts are the result of random mutation rather than rational design. The information and analysis they present is a good proxy for how difficult it would be for a transactor to litigate this distinction in court. Moreover, not all of the information they refer to could be known at trial—for example, how the market would respond if the court gave weight to a small drafting difference.

[52] When real black holes actually exist, parties are likely to be largely unconstrained with respect to the range of plausible subjective intents they can proffer. As a consequence, the assumptions made in the text are likely to characterize a meaningful subset of the cases that might arise.

[53] See, e.g., Brief of Defendant-Appellant the Republic of Argentina, NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2012 WL 6777133; Joint Response Brief of Plaintiffs-Appellees NML Capital, LTD and Olifant Fund, LTD, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2013 WL 388621.

[54] CG&S are careful to delineate a number of reasons that the collective action problems they observe in the sovereign debt markets where the contracts at issue—bond indentures—are in practice closer to the pole of financial instruments, may be particularly severe compared to the collective action problems that might or might not impede the revision of ordinary contracts for sales of goods or provision of services when a court makes an aberrant interpretation of one of their terms. See Choi et al., supra note 1, at 59–66.

[55] The reform could take the form of a new sub-chapter of Article 2 akin to Article 2A on leases, or an additional subset of rules in Article 2 akin to its merchant rules.

[56] While such a change would raise the issue of how to decide which contracts would come within the new Article’s purview, it could be made available to the parties on an opt-in basis at the time of contracting, subject to rejection by a court that found the choice to be inappropriate.

[57] For an overview of the problems with these private uniform law-making bodies, see generally Alan Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. Pa. L. Rev., 595–654 (1995). For an example of the difficulties of creating new Articles of the UCC in particular, consider the largely failed effort to get states to adopt the Uniform Computer Information Transactions Act, which has been adopted by only two states. See The Uniform Computer Information Transactions Act (UCITA) Is a Proposed State Contract Law, UCITA Online, http://www.ucitaonline.com/ [https://perma.cc/8SUP-W2L8] (discussing the impediments to adopting the model law).

[58] The best available, though imperfect, empirical evidence suggests that sophisticated commercial parties prefer textualist adjudication. See Lisa Bernstein, Merchant Law in a Modern Economy, in Philosophical Foundations of Contract Law 251 (Gregory Klass, George Letsas & Prince Saprai eds., Oxford Univ. Press 2014) (noting the common use of plain meaning clauses in large commercial contracts providing for arbitration); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001) (documenting the cotton industry’s preference for a textualist/formalist adjudicative approach); see also Bernstein, Merchant Law in a Modern Economy, supra (documenting the grain and feed industry’s preference for a textualist/formalist legal approach); Theodore Eisenberg & Geoffrey P. Miller, The Flight to New York: An Empirical Study of Choice of Law and Choice of Forum Clauses in Publicly-Held Companies’ Contracts, 30 Cardozo L. Rev. 1475 (2009) (demonstrating commercial actors’ strong preferences for relatively formalist New York Law); Stuart Popham, The View of European Business: Survey Results Presentation, Clifford Chance LLP (Mar. 14, 2008), http://denning.law.ox.ac.uk/news/events_files/Popham_-_presentation.PPT [https://perma.cc/DRB3-KXAL] (finding that in business contracts that provided for arbitration at the International Chamber of Commerce, transactors preferred British law, the most formalistic and textualist of the EU alternatives).

[59] See generally Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 Yale L.J. 926 (2010) (defending a formalist approach to contract interpretation).

[60] Another potential direction for reform that could be implemented without changing the common law, but that depends critically on the ability of groups—like state legislatures, the ALI, the NCCUSL, trade associations, and a variety of private entities and intermediaries—to revise contractual language when courts make errant decisions, would be to selectively adopt the reform proposals of the legal choice theorists. See Hanoch Dagan & Michael Heller, The Choice Theory of Contracts 102–13 (Cambridge Univ. Press 2017). These theorists explore the benefits that might accrue when states and other private and public entities provide more “contract types” that transactors could use to structure their affairs—much as Delaware law offers those structuring a business the option of incorporating as a public entity, a close corporation, an LLC or a partnership. If these “types” were carefully developed, and proved easier to change in the event of an errant judicial interpretation of their terms than individually structured agreements, heeding the call for the creation of these types may help solve the black hole problem in some contracting contexts. See infra notes 61–65 and accompanying text (noting that trade associations have proved remarkably able to amend their trading rules and standard form contracts in response to arbitral interpretations of these terms). But see Schwartz & Scott, infra note 59 (discussing the problems that groups like the ALI have in revising their model laws).

[61] See Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710, 730–32 (1999) (providing a short history of the Textile Trading Rules).

[62] Id. at 732.

[63] Id. at 736 (quoting SAA, Thirty-Fifth Annual Report 23 (1907)).

[64] Id. (quoting SAA, Fortieth Annual Report 31 (1912), Revision, 3 Silkworm 73 (May 1921)).

[65] Id. (quoting SAA, Fifty-Third Annual Report 26 (1925)).

[66] For a detailed discussion of the institutional response to the pari passu problem which suggests that institutions “learn” each time they attempt responses, see Anna Gelpern, Ben Heller & Brad Setser, Count the Limbs: Designing Robust Aggregation Clauses in Sovereign Bonds, in Too Little, Too Late: The Quest To Resolve Sovereign Debt Crises (Martin Guzman, José Antonio Ocampo & Joseph E. Stiglitz eds., Columbia Univ. Press 2016).

[67] Choi et al., supra note 1, at 40.

[68] Id. at 39.

[69] Id. at 40.

[70] However, as CG&S point out, these revisions occurred only in sovereign debt—not quasi-sovereign debt—which in their view left the market facing significant social costs. Id. at 24–25.

[71] See, e.g., KMStandards (2016), KMstandards.com [https://perma.cc/5KTY-Z2NJ]; IACCM (2017), IACCM.com [https://perma.cc/YH8G-8D73] (providing contract drafting software that permits the quick and inexpensive drafting of largely standard form contracts while giving transactors a variety of clauses to choose from), the website of the International Association of Contract and Commercial Managers [https://perma.cc/KKP7-3SMZ] (containing information about new contract drafting technologies).

[72] Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 91–130 (1989) (describing the theory of the penalty default rule).

[73] Choi et al., supra note 1, at 69–70 (citing discussions with Bernstein).

Black Hole Apparitions

Volume 67 December 2017
Black Hole Apparitions
Lisa Bernstein†. Wilson-Dickinson Professor of Law and Aaron Director Research Scholar, The University of Chicago Law School, and International Research Fellow, Center for Corporate Reputation, Said School of Business, University of Oxford. I would like to thank Douglas Baird, Patrick Barry, Edward Bernstein, Brian Bix, Sadie Blanchard, Robin Effron, Roger Ford, Anna Gelpern, William Hubbard, Todd Ito, Avery Katz, Kate Lindgren, Bill Schwesig, Mark Ramseyer, Steve Ware, and participants at the Duke Conference on Contractual Black Holes (2017) for helpful conversations. Special thanks are due to the hole theorists themselves for engaging conversations about their piece over the past year.

Introduction

In their article The Black Hole Problem in Commercial Boilerplate [1][1]. Stephen J. Choi, G. Mitu Gulati & Robert E. Scott, The Black Hole Problem in Commercial Boilerplate, 67 Duke L.J. 1 (2017). and its companion pieces, [2][2]. The other two papers in the Black Hole corpus are Stephen J. Choi, Mitu Gulati, & Robert E. Scott, Contractual Arbitrage, in Oxford Handbook of International Governance (Eric Brosseau, Jean-Michel Galmant & Jerome Sgaard eds., forthcoming 2017) http://scholarship.law.duke.edu/cgi/ viewcontent.cgi?article=6304&context=faculty_scholarship [https://perma.cc/VC3Y-QLKN] and Stephen J. Choi, Mitu Gulati & Robert E. Scott, Variation in Boilerplate: Rational Design or Random Mutation, 19 AM. L. & ECON. REV. (forthcoming 2017) [hereinafter, Choi et al., Variation in Boilerplate]. Stephen Choi, Mitu Gulati, and Robert Scott [hereinafter “CG&S”] introduce and explore the concept of “contractual black holes,” boilerplate contract provisions that have been “emptied of any recordable meaning” [3][3]. Choi et al., supra note 1, at 2. through rote repetition, [4][4]. Id. at 6. the introduction of essentially “random variations in language,” [5][5]. Id. and the absence of “any validation [of their meaning] from courts or industry institutions.” [6][6]. Id. The article presents a detailed case study of the sovereign debt market’s slow response to the Second Circuit’s errant interpretation of the pari passu clause at issue in NML v. Argentina, [7][7]. NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also focuses on an earlier interpretation of the pari passu clause rendered by a Brussels court in 2000. See Elliott Assocs. v. Republic of Peru, 948 F. Supp. 1203 (S.D.N.Y. 1996). a common clause in sovereign debt contracts that they view as the “prototypical exemplar” [8][8]. Choi et al., supra note 1, at 7. of a black hole. Noting that the NML decision put a multi-trillion dollar market at risk by making it more difficult to restructure sovereign debt, [9][9]. Interestingly, the Second Circuit seemed to reject the idea that it was jeopardizing the market as a whole by making restructuring more difficult. See NML Capital, Ltd., 699 F.3d at 263 (explaining why this effect is unlikely to occur.) and that it took years for market transactors to draft clauses eliminating the problem, [10][10]. Choi et al., supra note 1, at 70 (concluding that “the inherently greater inertia costs [and social costs] that result from an aberrant interpretation of a black hole term,” as compared to an ordinary unclear term, and the “greater difficulty market players face in overcoming the resulting collective action problem[s]” slow the adoption of new and better contractual terms). the authors conclude that changes in contract doctrine are needed to deal with the problems caused by black holes. As they explain, “while black holes often remain for many years as relatively harmless surplusage, they can [also] generate substantial social costs once litigation results in an interpretation that introduces inefficiencies into the market.” [11][11]. Id.

This Comment focuses on the authors’ proposed doctrinal solution to the black hole problem, which seeks to eliminate any inquiry into “subjective intent” when courts are faced with the task of interpreting a black hole. It explores the conceptual and practical challenges of implementing the authors’ proposal and then questions whether legal reform is really needed to deal with the black hole problem. Part I identifies several common ways that standardized contract provisions that are often indistinguishable from true black holes may arise. [12][12]. Although it is possible that some of the types of clauses this Comment views as “black hole apparitions” would be classified as genuine black holes by CG&S, their search for additional examples of black holes by holding a conference on the subject at Duke Law School suggests that these ubiquitous types of clauses with uncertain meaning are unlikely to come within the purview of their definition. As Gulati explained, “black holes where meaning has been lost are different from provisions where there is a general sense of what they mean, but there is some vagueness or lack of clarity.” E-mail from Mitu Gulati, Professor of Law, Duke University, to author (Sept. 23, 2017) (on file with author). It suggests that any doctrinal solution to the black hole problem will have to either reliably distinguish real black holes from these relatively common “black hole apparitions,” or be desirable when applied to both types of provisions. Part II describes the proposed reform and explores the practical barriers to implementing it. Along the way it also sketches out several alternative avenues for solving the problems evidenced in the pari passu saga. It suggests that these avenues, while more limited in scope than the authors’ proposed reform, may turn out to be more feasible and less costly to implement. Part III questions whether any doctrinal or other solution to the so-called black hole problem can be justified on the basis of the lessons learned from the pari passu saga alone, given that other markets have been able to overcome collective action problems and adopt and amend standard-form contracts through processes that have tended to work more quickly and less contentiously over time. Finally, Part IV concludes by suggesting that the interpretive approach adopted by the Second Circuit in NML might be a passably good response to the black hole problem writ large.

I. A Typology of Black Hole Apparitions

CG&S suggest that contract doctrine should treat the interpretation of black holes differently from the interpretation of other unclear or essentially meaningless contractual provisions. Yet both conceptually and practically, it may be extraordinarily difficult to distinguish true black holes from what might be called “black hole apparitions.” Black hole apparitions are contract provisions that are indistinguishable on their face from black holes yet will not necessarily give rise to the drafting inertia and social costs that led CG&S to conclude that a doctrinal response to the black hole phenomenon was needed. Recognizing the existence of these apparitions suggests that any doctrinal reform that turns on a provision’s status as a black hole will, [13][13]. In order for a doctrine that conditions on whether a clause is or is not a black hole to be workable, a clause’s status must be verifiable by a court—that is, a court must be able to determine with reasonable accuracy at a cost the transactors consider reasonable from an ex ante perspective whether or not something is a black hole. See generally Alan Schwartz, Relational Contracts and the Courts: An Analysis of Incomplete Agreements and Judicial Strategies, 21 J. Legal Stud. 271 (1992). in practice, bring many black hole apparitions into its orbit. As a consequence, any such change is likely to add significant costs to ordinary commercial litigation and create and/or exacerbate opportunities for strategic behavior. Against this background, and to understand how common these apparitions are likely to be, it is useful to look more closely at three common types of black hole apparitions, those whose origins are procedural, relational or rational.

A. Procedural Apparitions

One category of black hole apparitions is what might be termed procedural apparitions—contract provisions that are at a high risk of losing their meaning due to state-supplied procedural rules or transactors’ procedure-related drafting choices.

For example, black holes are especially likely to arise in markets where most contracts provide for arbitration under the American Arbitration Association’s (AAA) Commercial Arbitration Rules. [14][14]. Commercial Arbitration Rules and Mediation Procedures (Am. Arbitration Ass’n 2013), https://www.adr.org/sites/default/files/Commercial%20Rules.pdf. [https://perma.cc/E8EW-CBBP]. Although it is increasingly common for AAA arbitrators to produce reasoned opinions rather than mere awards, [15][15]. The AAA discourages written opinions but permits parties to request them. See Charles A. Cooper, Harry Kaminsky & Neil Carmichael, Manual for Commercial Arbitration 162 (Am. Arbitration Ass’n 1999) (“Currently in domestic arbitrations, the AAA does not encourage commercial arbitrators to write opinions which give their reasons for the award. However, in instances where both parties request an opinion prior to the appointment of the arbitrator, the arbitrator should comply.”). While transactors might opt to revise a clause in face of a seemingly aberrant arbitration ruling, they might also surmise, much as the sovereign debt community did after the ruling of the Brussels court, that a more sensible panel (or as in the case of the pari passu clause the Second Circuit Court of Appeals) would rule differently. See Choi et al., supra note 1, at 12. these opinions are typically kept private and are of no precedential value. [16][16]. But see generally W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (suggesting that in some contexts arbitration decisions that are not technically precedent nevertheless have an effect on the outcome of future disputes). In addition, there are indications that companies fear that arbitral decisions will be given some weight if they have future disputes with the same counter-party. For example, Verizon’s consumer agreement provides for AAA commercial arbitration, includes a class action waiver, and also contains a clause stating that “[a]n arbitration award and any judgment confirming it apply only to that specific case; it can’t be used in any other case except to enforce the award itself.” Customer Agreement, Verizon Wireless (Oct. 20, 2017), https://www.verizonwireless.com/legal/notices/customer-agreement/ [https://perma.cc/3A2L-6YGT]; see also What’s Going on? Arbitration Explanation, Waste Mgmt., http://mediaroom.wm.com/legal-notice/ [https://perma.cc/TZ3C-X4GB] (“The award of the arbitrator may be entered and enforced in any court having jurisdiction. An arbitration award and any judgment confirming it apply only to that specific case; it cannot be used in any other case except to enforce the arbitration award itself.”). As a consequence, the terms of these contracts (whether standard or bespoke) are correspondingly less likely to be interpreted in written opinions with precedential value and are at a heightened risk of gradually losing their meaning over time. The routine inclusion of a AAA arbitration clause, therefore, increases the risk that some of an agreement’s provisions will devolve into black hole apparitions. [17][17]. For example, trade association-run private legal systems are designed in ways that will strongly discourage the emergence of black holes. At the National Grain and Feed Association, for example, arbitrators write reasoned opinions that are posted on the association’s website. When a rule is unclear, or appears to lead to an undesirable or unanticipated result, the arbitrators will state this explicitly, apply the rule as it is stated, and note that the association’s rules committee should consider revising the rule, something that is typically done. See Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1780 & nn.50¬–51 (1996).

Similarly, the standards of appellate review and the fact that most state and many federal trial courts do not routinely publish opinions [18][18]. Steven M. Barken, Barbara A. Bintliff & Mary Whisner, Fundamentals of Legal Research 41 (10th ed. 2015) (“Ordinarily, cases decided by state trial courts are not reported. . . . Only a few states, such as New York, Ohio, Pennsylvania, and Virginia, publish some trial court opinions, but those selected are few in number and represent only a very small portion of the total cases heard by trial courts.”). A similar problem may exist with respect to cases decided in federal court. See David A. Hoffman, Alan J. Izenman & Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash. U. L. Rev. 681, 727 (2007) (“[S]tarkly expos[ing] how little trial court work is explained through written opinions.”). might also play a role in the emergence of some types of contractual black holes. The vast majority of appellate courts provide greater deference to lower-court decisions on question of fact than on questions of law, or in some jurisdictions, mixed questions of law and fact. As a consequence, the meaning of standard-like provisions that are interpreted using a highly fact-specific inquiry is less likely to be appealed than the meaning of rule-like provisions whose interpretation is more often considered a matter of law. Standard-like provisions are correspondingly less likely to be interpreted in a written opinion of any kind. They are therefore more likely than rule-like provisions to devolve into meaningless provisions that look like classic black holes. [19][19]. It is, however, possible that the doctrinal contours of the black hole or no black hole determination would become clearer over time because under the proposed change this would be determined as a “matter of law.” See Choi et al., supra note 1, at 68.

B. Relational Apparitions

A second type of black hole apparition, which might be called a relational apparition, is particularly likely to emerge: (1) where there is a written or standard-form template that is used in most deals in a particular market; or (2) where seemingly bespoke contracts are used, [20][20]. CG&S focus mostly on standard commercial provisions in standardized contracts. However, both black holes and black hole apparitions can arise in seemingly bespoke contracts as well. To the extent this phenomenon exists, the proposed reform, in as much as it seems to apply only to standard contracts with boilerplate provision, would be underinclusive with respect to the types of holes found in these seemingly bespoke agreements. but those who negotiate them (either the business people or the contract lawyers) share a rough but deeply ingrained understanding about the types and content of terms that are commonly included in such agreements. In both of these situations, transactors may be especially hesitant to propose changes to commonly used terms. When a transactor proposes a change in the usual way of doing things, even when that usual way is not terribly important or when the variant creates value for both transactors, her counter-party might well interpret it as a signal that she is either more likely to engage in strategic behavior or more focused on her legal rights than the average transactor. The counter-party might therefore ask for protective changes in response. This, in turn, might trigger a cascade of change requests that could greatly increase contracting costs, reduce the likelihood of post-signing cooperation, [21][21]. The way that a contract is negotiated has an effect on how well the business deal can be operationalized once the contract is signed. See Danny Ertel, Getting Past Yes: Negotiating as if Implementation Mattered, 85 Harv. Bus. Rev. 60, 62 (2004). or destroy the deal completely. [22][22]. For a more complete discussion of this bargaining dynamic and the way that it effects the terms of agreements, see generally Lisa Bernstein, Social Norms & Default Rules Analysis, 3 S. Cal. Interdisc. L.J. 59 (1993). As long as transactors are aware of or intuit this dynamic, they might be reluctant to propose changing or eliminating common provisions, even if these provisions appear to be meaningless or irrelevant and standardization, across either the market as a whole or particular subsets of market participants, has no independent value.

Relational black hole apparitions may also be common in markets where transactors trust one another or where the force of reputation-based network governance is strong. In some markets where genuine interpersonal trust is present, it remains common for transactors to put the contract in the drawer. As a consequence, transactors will likely spend little or no time negotiating its terms. In these situations, contract language is quite likely to be unthinkingly recycled. [23][23]. Gillian K. Hadfield & Iva Bozovic, Scaffolding: Using Formal Contracts To Build Informal Relations to Support Innovation, 2016 Wis. L. Rev. 981, 982, 987 (noting that Stewart Macaulay’s finding that in commercial contracting relationships problems tended to be worked out informally and “written contracts . . . were often highly-standardized documents that were often confined to the drawer,” held in a modern day sample of non-innovative firms where informal methods of dispute resolution remained common and firms either “did not generate [formal contracts] or relied only on standardized documents,” but did not hold in a sample of innovative firms). Similarly, in markets where transactors rely on reputation-based network governance to support exchange, using similar contracts across the market makes these reputation-based forces stronger. [24][24]. See generally Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561 (2015) (explaining how widely known and standardized contract provisions may affect the strength of network governance). When contracts are relatively standardized and/or there is widespread consensus about what constitutes cooperation and good-faith behavior, [25][25]. In some markets where reputation-based network governance plays an important role in contracting, some large companies post their quality requirements and their standard terms and conditions on their website so that the contours of what is and is not expected from their counterparties is known throughout the market. See, e.g., Supplier Quality Manual, John Deere, https://jdsn.deere.com/wps/portal/jdsn/Applications?WCM_GLOBAL_CONTEXT=/wps/wcm/connect/jdsn_website/jdsn/business+processes/quality/supplier_quality_manual/supplier_quality_manual_index [https://perma.cc/57SG-BQUK]; Terms and Conditions for the Purchase of Goods and Services, John Deere, https://www.johndeerestore.com/jdb2cstorefront/JohnDeereStore/en/terms [https://perma.cc/B7DH-QPQU]. the network need only transmit information about how a transactor behaved in order for other market participants to assess the desirability of dealing with him. In contrast, when contracts are more varied, information about both the scope of the relevant obligations and actual behavior may need to be transmitted through the relevant network in order for reputation to play an important role in governing exchange—a process that is bound to be more expensive, less effective, and more error prone than in markets where relatively standard agreements are used. [26][26]. In such markets, general and impressionistic information about whether a transactor behaved properly or poorly could still be easily transmitted, but in the absence of standard terms, a consensus about what constitutes good behavior, or information about both what was promised and what was done, this information is likely to be given less weight (especially when it travels more than one step from its source) than information that circulates in markets where contracts and/or expectations are either relatively standardized or widely known. Bernstein, supra note 24, at 578–89 (discussing how John Deere achieves this). Since lawsuits are less common in markets where genuine trust is present or network governance is operational, fewer terms in commercial contracts are likely to be interpreted by courts and both black holes and black hole apparitions are correspondingly more likely to arise.

C. Rational Apparitions

A third and final type of black hole apparition, and one that may also emerge in both standard-form and bespoke contracts, is a rational black hole. A rational black hole is a term that is left deliberately vague or without meaning at the time of contracting. [27][27]. Sometimes the line between relational and rational black holes may be blurry. For example, Scott and Triantis note that it is often economically rational to use very standard-like terms like “good faith,” “reasonable efforts,” or “best efforts,” and leave it for the court to give these terms meaning ex post. Robert E. Scott & George Triantis, Anticipating Litigation in Contract Design, 115 Yale L.J. 814, 835 (2006). However, once those terms become common in an industry’s contracts (or between a pair of long-term repeat dealers), their presence in contracts may be exceptionally sticky. If, for example, someone sends their counterparty a draft obligating the counterparty (and themselves) to act in good faith, the relational costs of saying to one’s counter party, “well let’s spell that out” or “let’s leave that out” might be high. The economic literature has identified a variety of reasons that the inclusion of such terms might be rational. [28][28]. For a discussion of rational reasons why many clauses are left vague or undefined, see Scott & Triantis, supra note 27, at 814–15. The literature conceptualizes these terms as agreements-to-agree or agreements to determine meaning of a term through negotiation or litigation if an interpretive dispute arises. Among them: transactors believe that they will be better able to devise a response to a contingency when it actually arises, since additional facts about its implications will be known; transactors conclude that the probability a contingency will arise is so low (or the variety of related contingencies is so large) that it is not worth dealing with any or all of them ex ante; and finally, transactors have divergent views of the bargaining power they will each have when and if the issue arises, leading each to favor dealing with it later. In addition, in some contexts, clauses may be left meaningless or vague simply because transactors realize that in the event of a dispute over the meaning of the clause, the aggrieved party is unlikely to have a credible threat to sue for any of a number of reasons, including: litigation costs, the reputational cost of litigation, the information they would have to divulge in discovery, [29][29]. For a complete discussion of the ways that transactors’ “secrecy interest” in certain types of business information may affect the credibility of their threat to sue in a variety of contexts, see generally Omri Ben Shahar & Lisa Bernstein, The Secrecy Interest in Contract Law, 109 Yale L.J. 1885 (2000). or the fact that litigation would likely end their otherwise valuable contracting relationship. [30][30]. In contracting relationships like those between an Original Equipment Manufacturer and its suppliers of component parts where a lawsuit tends to end the parties contracting relationship, many contractual provisions that are formally enforceable may not operate in the shadow of the law at all. The reason is simple: if one of these provisions were breached in isolation, the promisee would not have a credible threat to sue since the expected recovery would be very unlikely to exceed the future value of the relationship as a whole. These “interior contract provisions” operate outside of the shadow of the law, unless and until they are breached in combination with enough other provisions that the counterparty concludes that there has been a breach of the contracting relationship that makes it worthwhile to end the transactors’ business relationship and sue. See Bernstein, supra note 24, at 570–71.

D. Conclusion

In sum, the existence of black hole apparitions complicates any attempt to deal with the black hole problem through changes in general contract doctrine that are contingent on a term’s status as a black hole. [31][31]. At one point in their analysis CG&S hint that it might be desirable to apply a knock-out rule to black hole terms akin to the rule adopted by some courts applying U.C.C. § 2-207. However, this rule will also require black holes to be identified by courts and will therefore be unworkable for many of the same reasons discussed herein. There are likely to be many contexts where black holes cannot be distinguished from black hole apparitions. In addition, depending on precisely how a black hole is defined, there are likely to be some instances where black holes exist yet the type of redrafting inertia that beset the sovereign debt market is unlikely to arise since the need for the type of market wide uniformity that is desirable when contracts are akin to financial instruments will not exist. As a consequence, any reform that conditions on a clause’s status as a black hole and applies to commercial contracts generally is likely to be overinclusive. By bringing commercial contracts dealing with the routine purchase and sale of goods and provision of services within its purview, any such reforms will increase litigation costs and create substantial social and private costs of their own.

Indeed, the pari passu black hole itself can be understood as arising from a mix of the procedural, relational, and rational considerations that give rise to black hole apparitions. The pari passu is partly a procedural black hole. Historically, the law of sovereign immunity made it extraordinarily difficult to sue or collect judgments [32][32]. See Sadie Blanchard, Courts Without Enforcement: Adjudicating Reputation in the Sovereign Debt Market 6–7 (Oct. 27, 2017) (unpublished manuscript) (concluding that “[e]ven though [the holders of sovereign debt] cannot enforce debtor states’ obligations through conventional judicial means, courts play a key role in the sovereign debt market . . . because they provide information that has the power to provoke reactions by third parties that are costly for the debtor or its agents,” so that “[c]reditors litigate because producing such information through the courts strengthens their leverage in settlement negotiations,” and makes reputation a more powerful force in the market). from other nations. [33][33]. See Julian Schumacher, Christoph Trebesch & Henrik Enderlein, Sovereign Defaults in Court 10–11 (May 6, 2014) (unpublished manuscript), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2189997 [https://perma.cc/T2Q7-LNGN], (finding that only 5 percent of sovereign defaults resulted in litigation until the 2000s, when the proportion rose to nearly half). Although in recent years it has become easier to sue a sovereign, [34][34]. See W. Mark C. Weidermaier, Sovereign Immunity and Sovereign Debt, 2014 U. Ill. L. Rev. 67, 68–70 (2014). many collection barriers remain. These legal and procedural barriers may be one reason why prior to a Brussels court’s 2000 decision in Elliott Associates v. Republic of Peru, [35][35]. Cours d’Appel de Bruxelles, 8th Chamber, 2000/QR/92, Elliott Assocs. v. République de Peru, 2000/QR/92, Sept. 26, 2000. there were no contemporary judicial decisions interpreting the meaning of pari passu provisions. The rote inclusion of the pari passu clause and the market’s slow reaction to the NML decision [36][36]. NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also looks at the market’s nonresponse to a pari passu decisions of a Brussels court, but attributes it to market players’ beliefs that other, more important courts for the sovereign debt community would not follow such an aberrant decision. can also be understood as the product of relational costs. As CG&S point out, as long as rating agencies did not downgrade bonds because the pari passu clause remained unchanged, individual issuers were wary of changing the clause and potentially triggering a negative rating-agency response. [37][37]. Choi et al., supra note 1, at 49. Finally, market players’ failure to revise the pari passu clause after the aberrant court judgments may or may not have been rational. [38][38]. As CG&S point out, it might not have been individually rational for any one issuer to make these changes to their past and future indentures, even though it would have been rational for the market as a whole to move to a different term. Id. at 13. However, the decision not to attend to its meaning more precisely at the time indentures were drafted may have been rational. After all, if the clause had not been interpreted in court for such a long period, the expected benefit of tinkering with it in a purposeful way might properly have been seen as small. Together, these considerations suggest that even on a purely conceptual level the line between a contract provision whose meaning has been eroded as a result of procedural, relational, and rational considerations and a true black hole is quite difficult to demarcate with analytic precision.

II. The Proposed Doctrinal Response

CG&S acknowledge that at present we have only a limited understanding about the frequency of black holes and the contexts in which courts’ erroneous interpretations of widely used terms will be met with drafting inertia. Nevertheless, they conclude that immediate legal reform to deal with the black hole problem is needed. In their view, the “vexing collective action problems” that market participants face when attempting to respond to errant court interpretations are so significant that the potential social costs of not dealing with the black hole problem are too significant to ignore. [39][39]. Id. at 37 (“We use both qualitative and quantitative data to support the claim that courts searching for shared intent in the case of black holes in standardized contracting can result in substantial social costs.”).

CG&S suggest that the best doctrinal way to deal with the black hole problem is for courts interpreting “boilerplate terms in commercial contracts” [40][40]. Id. at 67. to be “open to arguments that as a matter of law, the clause in question has been emptied of meaning and functions as a black hole in boilerplate.” [41][41]. Id. at 68. In cases in which the parties meet the burden of establishing the existence of a black hole, the court would then be required to adopt an interpretive approach that avoids any inquiry into subjective intent and looks to other interpretive principles to decide the case. [42][42]. See id. at 66–67, 69 Exactly what this alternative interpretive approach would be is not specified in great detail. See id. at 54–56. If, instead, some form of knock-out rule were contemplated, similar problems would arise. See supra note 31. Under the proposed approach, the hole or non-hole determination would turn on a variety of highly fact-specific considerations and types of evidence. These include the types of evidence adduced in the pari passu case study, [43][43]. Choi et al., supra note 1, at 38. as well as the types of information needed to answer the following questions:

Has the clause been repeated by rote over many years, without having been tested in litigation, where repetition has robbed the term of any obvious conventional meaning? Has the term been embedded in layers of legal jargon such that its intelligibility is substantially reduced and variations in the formulation of the term across contracts have no apparent significance? Is a historic or original meaning of the term accessible in a fashion that makes sense in the contemporary context and are contemporary commercial actors aware of that meaning? Is there credible evidence that the particular provision was priced at the original issue stage? [44][44]. Id. at 68.

Notwithstanding its theoretical appeal, the implementation of this approach faces significant practical challenges. First, the test will be widely overinclusive even with respect to standard provisions in standard-form contracts because answers to these questions will not reliably distinguish black holes from black hole apparitions. [45][45]. Indeed, the methodology CG&S used to reach the conclusion that the pari passu provision was a black hole, rather than an efficient contract term, relied on extensive information about the market’s response to the decision over a period of years. See id. at 11. Moreover, because the new test applies to all standard commercial agreements, it gives a significant advantage to the party with deeper pockets who might claim a black hole exists simply to increase putative litigation costs and thereby lever a better settlement.

Second, and relatedly, as CG&S themselves acknowledge, introducing this inquiry would open the door to moral hazard and other types of strategic behavior. [46][46]. The types of strategic behavior it would introduce are analogous to those identified in Alan Schwartz and Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 587 (2000) and Lisa Bernstein, Custom in the Courts, 110 Nw. L. Rev. 63, 106–08 (2015). Although they suggest that moral hazard can be adequately dealt with by adopting “an initial presumption against the existence of a black hole,” [47][47]. Choi et al., supra note 1, at 68. even if they are correct the risk of strategic behavior will remain substantial.

Third, even abstracting from overinclusiveness and strategic behavior, as a practical matter the type of information CG&S view as relevant to the black hole or non-black hole determination, will often be either unavailable to the parties or prohibitively costly to obtain. [48][48]. Id. at 55–56, 64. For example, in contexts where the relevant contracts are not disclosed under the securities laws, transactors’ reluctance to share this information will make it impossible to get the contracts used in similar transactions—an evidentiary problem that will prevent courts from answering many of the questions that CG&S view as directly relevant to the hole or no-hole determination. Similarly, there is no obvious way to determine whether a particular clause has been priced into a contract. In many settings, price and other terms are negotiated separately, and even when they are not, determining the connection between price and a particular term may be difficult either because the transactors did not explicitly think about the connection or because reconstructing the sequence of a negotiation from the testimony of witnesses with conflicting agendas is likely to be an error-fraught process. In addition, market players, even those who willingly spoke to academics, might be reluctant to talk to litigators or testify in court. Thus, even if in theory the hole or no-hole determination could be accurately made on the basis of the types of evidence that CG&S view as relevant, in practice much of this information is likely to be either prohibitively expensive or entirely unavailable.

Fourth, while CG&S want the courts to intervene when “parties exploit” either the absence of meaning or the presence of “random variations in language [to advance] an interpretation the market disavows,” [49][49]. Id. at 1. they offer little guidance on how random variants can be distinguished from what they call “rational design,” [50][50]. Choi et al., Variation in Boilerplate, supra note 2, at 4. a determination that would likely create both evidentiary problems and additional interpretive uncertainty. [51][51]. CG&S undertake just this inquiry in another part of the black hole corpus. See id. at 6. There, the authors seek to demonstrate that the observed variations in the pari passu clauses used in sovereign debt contracts are the result of random mutation rather than rational design. The information and analysis they present is a good proxy for how difficult it would be for a transactor to litigate this distinction in court. Moreover, not all of the information they refer to could be known at trial—for example, how the market would respond if the court gave weight to a small drafting difference.

Finally, without a more detailed description of the interpretive rule that the court would apply when a hole is found and a better sense of the extent to which the chosen rule would result in predictable outcomes, it is difficult to asses how frequently the disputing parties will have the proper incentives to take advantage of the proposed reform and argue that black holes exist, even in contexts where they both believe that a true black hole actually exists.

Consider, for example, a context in which both parties think that the best subjective interpretation of a clause that they can advance is better than the expected interpretation they think a court would give the clause if a black hole were found to exist. [52][52]. When real black holes actually exist, parties are likely to be largely unconstrained with respect to the range of plausible subjective intents they can proffer. As a consequence, the assumptions made in the text are likely to characterize a meaningful subset of the cases that might arise. In such a situation, both parties might opt not to argue for a hole’s existence even if they both genuinely believe that a hole exists. To see why, consider a plaintiff who is trying to decide whether to plead that a hole exists or to advance her own subjective interpretation of a clause. If she pleads that a hole exists, the defendant faces a strategic choice. If he pleads that a hole exists, then the court will likely find that a hole exists, deem the subjective intent of the parties irrelevant, and interpret the clause using its own interpretive principles. In contrast, if the defendant pleads his subjective interpretation, the court will either accept his subjective interpretation or conclude that a hole exists, deem subjective intent irrelevant, and decide the case using its own interpretive principles. Because (by assumption) both parties believe that their subjective interpretation is better than the court’s interpretation, if the plaintiff pleads that a hole exists, the defendant is better off pleading that his subjective interpretation should govern. This response creates at least some chance that the court will accept his preferred subjective interpretation. Alternatively, if the plaintiff pleads her own subjective interpretation of the clause, the defendant will reason that he is always better off pleading his subjective interpretation than claiming that a hole exists. If he argues for his subjective interpretation, there is at least some possibility that the court will accept his interpretation, whereas if he pleads a hole, the court will find either find that a hole exists or accept the plaintiff’s subjective interpretation. Thus, regardless of the plaintiff’s choice, the defendant is always better off arguing for his subjective interpretation of the clause. Given this, the plaintiff too is better off arguing for her own subjective interpretation of the clause, since doing so creates at least some chance that her interpretation will be accepted by the court. As a consequence, at least in the run of situations where both parties think that the best subjective interpretation of the clause they can advance is better than the expected interpretation the court would give the clause, they are unlikely to avail themselves of the proposed doctrinal step even when a hole exists. Indeed, in the NML case itself neither party asserted that the pari passu clause was without meaning. Rather each side advanced an argument that it meant something different. [53][53]. See, e.g., Brief of Defendant-Appellant the Republic of Argentina, NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2012 WL 6777133; Joint Response Brief of Plaintiffs-Appellees NML Capital, LTD and Olifant Fund, LTD, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2013 WL 388621.

A. Other Directions for Reform

In sum, proposals to alter contract doctrine to respond to the black hole problem may be difficult to implement in practice. As a consequence, to the extent that the problems revealed in the pari passu saga are wide-spread, it is worthwhile to explore whether there are approaches that do not rely on the hole-or-no-hole distinction that could prove to be a more workable response to the black hole problem.

If, for example, the inertia problem were found to exist more broadly in particular contexts—such as markets where the contracts are closer to the pole of financial instruments than ordinary contracts for services or the sale of goods [54][54]. CG&S are careful to delineate a number of reasons that the collective action problems they observe in the sovereign debt markets where the contracts at issue—bond indentures—are in practice closer to the pole of financial instruments, may be particularly severe compared to the collective action problems that might or might not impede the revision of ordinary contracts for sales of goods or provision of services when a court makes an aberrant interpretation of one of their terms. See Choi et al., supra note 1, at 59–66.—a separate article of the UCC could be added that would apply different interpretive standards as a matter of course. [55][55]. The reform could take the form of a new sub-chapter of Article 2 akin to Article 2A on leases, or an additional subset of rules in Article 2 akin to its merchant rules. This change would have the advantage of eliminating the costly and uncertain black hole inquiry. [56][56]. While such a change would raise the issue of how to decide which contracts would come within the new Article’s purview, it could be made available to the parties on an opt-in basis at the time of contracting, subject to rejection by a court that found the choice to be inappropriate. Nevertheless, given the well-known problems that impede the American Law Institute (ALI) and National Conference of Commissioners on Uniform State Laws (NCCUSL) rules creation processes, such an approach, however desirable, might also prove infeasible. [57][57]. For an overview of the problems with these private uniform law-making bodies, see generally Alan Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. Pa. L. Rev., 595–654 (1995). For an example of the difficulties of creating new Articles of the UCC in particular, consider the largely failed effort to get states to adopt the Uniform Computer Information Transactions Act, which has been adopted by only two states. See The Uniform Computer Information Transactions Act (UCITA) Is a Proposed State Contract Law, UCITA Online, http://www.ucitaonline.com/ [https://perma.cc/8SUP-W2L8] (discussing the impediments to adopting the model law).

Alternatively, if the black hole problem turns out to be widespread in certain types of well-defined markets or types of contracts, eliminating any inquiry into subjective intent in those contexts would achieve most of the benefits of the proposed black hole reform without increasing litigation costs or the frequency of strategic behavior. Black holes aside, eschewing the legal fiction of subjective intent in the interpretation of commercial contracts and adopting a largely textualist plain-meaning oriented interpretive approach is a reform that would likely prove highly beneficial for sophisticated commercial transactors [58][58]. The best available, though imperfect, empirical evidence suggests that sophisticated commercial parties prefer textualist adjudication. See Lisa Bernstein, Merchant Law in a Modern Economy, in Philosophical Foundations of Contract Law 251 (Gregory Klass, George Letsas & Prince Saprai eds., Oxford Univ. Press 2014) (noting the common use of plain meaning clauses in large commercial contracts providing for arbitration); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001) (documenting the cotton industry’s preference for a textualist/formalist adjudicative approach); see also Bernstein, Merchant Law in a Modern Economy, supra (documenting the grain and feed industry’s preference for a textualist/formalist legal approach); Theodore Eisenberg & Geoffrey P. Miller, The Flight to New York: An Empirical Study of Choice of Law and Choice of Forum Clauses in Publicly-Held Companies’ Contracts, 30 Cardozo L. Rev. 1475 (2009) (demonstrating commercial actors’ strong preferences for relatively formalist New York Law); Stuart Popham, The View of European Business: Survey Results Presentation, Clifford Chance LLP (Mar. 14, 2008), http://denning.law.ox.ac.uk/news/events_files/Popham_-_presentation.PPT [https://perma.cc/DRB3-KXAL] (finding that in business contracts that provided for arbitration at the International Chamber of Commerce, transactors preferred British law, the most formalistic and textualist of the EU alternatives). for reasons that have long been advanced by the neo-formalist school of contract interpretation. [59][59]. See generally Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 Yale L.J. 926 (2010) (defending a formalist approach to contract interpretation).

Indeed, the neo-formalist approach reflected in the Second Circuit’s NML decision can be understood as taking aim at one of the causes of black holes—namely, encrustation. According to CG&S, encrustation is the end result of the meaningless drafting changes that lawyers introduce either without thinking or to show their clients they are doing something. By signaling that the court may well attach radically different meanings to clauses with small wording differences, the approach taken by the NML court does two key things: it increases the cost of meaningless amendments and should therefore discourage them; and it increases the return to meaningful amendments since the court will take them into account in interpreting the clause which should, in turn, encourage meaningful drafting. It can therefore be understood as an interpretive penalty default rule that should over time change drafting behavior in ways that may reduce the amount of meaningless encrustation in commercial agreements and with it the likelihood that black holes will emerge.

Nevertheless, despite the potentially significant problems that black holes may create and the many possible directions legal reform might take, [60][60]. Another potential direction for reform that could be implemented without changing the common law, but that depends critically on the ability of groups—like state legislatures, the ALI, the NCCUSL, trade associations, and a variety of private entities and intermediaries—to revise contractual language when courts make errant decisions, would be to selectively adopt the reform proposals of the legal choice theorists. See Hanoch Dagan & Michael Heller, The Choice Theory of Contracts 102–13 (Cambridge Univ. Press 2017). These theorists explore the benefits that might accrue when states and other private and public entities provide more “contract types” that transactors could use to structure their affairs—much as Delaware law offers those structuring a business the option of incorporating as a public entity, a close corporation, an LLC or a partnership. If these “types” were carefully developed, and proved easier to change in the event of an errant judicial interpretation of their terms than individually structured agreements, heeding the call for the creation of these types may help solve the black hole problem in some contracting contexts. See infra notes 61–65 and accompanying text (noting that trade associations have proved remarkably able to amend their trading rules and standard form contracts in response to arbitral interpretations of these terms). But see Schwartz & Scott, infra note 59 (discussing the problems that groups like the ALI have in revising their model laws). the current understanding of black holes may be far too limited to devise sufficiently nuanced reforms. As discussed further below, there are reasons to question whether the saga of the pari passu, which is the empirical foundation of CG&S’s call for immediate reform, is, on its own, sufficient to justify the need for an immediate and generally applicable legal response, especially given the many differences—differences CG&S carefully delineate—between bond indentures and the other types of contracts the proposed reform would govern.

III. Is Immediate Reform Needed?

Given the many difficulties that stand in the way of a legal solution to the black hole problem, it is important to explore whether legal reform is really needed. The answer to this question turns, in large part, on whether the inertia in revising contract terms after an aberrant decision interpreting a black hole is likely to be a frequent occurrence across a variety of markets. It also depends on whether those black holes that do emerge are likely to be enduring phenomena or a transitory stage in the development of the relevant market.

There is no systematic empirical evidence about the frequency of black holes. However, the experiences of American trade associations that created standard form contracts and trading rules to govern transactions among their members suggest that the inertia costs that stem from collective action problems and impede the adoption of new contract provisions, may decrease over time as industry participants and/or market institutions learn how to more quickly respond to the need for market-wide contractual change.

In the textile industry, for example, the first edition of The Worth Street Textile Trading Rules, [61][61]. See Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710, 730–32 (1999) (providing a short history of the Textile Trading Rules). which consisted of both a set of trading rules and a standard textile sales note, took eighteen years of committee work to draft. The rules creation process “was fraught with conflict, [and] involved negotiations among numerous trade associations.” [62][62]. Id. at 732. Yet once adopted, the rules were revised and even entirely rewritten numerous times far more quickly and with a great deal less infighting. Similarly, when the Silk Association of America set out to create a set of trading rules for raw silk, many controversies arose. The rules-creation process took seven years to complete because “[t]he get-together spirit was not sufficiently pronounced to override the differences that arose . . . .” [63][63]. Id. at 736 (quoting SAA, Thirty-Fifth Annual Report 23 (1907)). Nevertheless, despite these initial difficulties, the Raw Silk Trading Rules, like the Worth Street Rules, were subsequently amended and revised many times with far less difficulty. A 1921 amendment, for example, was adopted after only “a year of careful study on the part of the [rules] Committee,” [64][64]. Id. (quoting SAA, Fortieth Annual Report 31 (1912), Revision, 3 Silkworm 73 (May 1921)). and a 1924 amendment was adopted after mere “months of intensive effort.” [65][65]. Id. (quoting SAA, Fifty-Third Annual Report 26 (1925)). Similarly, the National Grain and Feed Association’s effort to adopt the first set of Grain Trading Rules was also hard-fought and filled with disagreement. Yet since their adoption, these rules have been successfully amended seventy-seven times, also with little infighting and only rare controversy. The experience of these and other industries suggests that while the initial costs of agreeing on standard contract terms and trading rules may be high, over time these costs tend to gradually decrease as market players and institutions get used to working with one another to respond to technological changes, market changes, and other types of disruptions that require market-wide changes in contractual rules or forms.

The experience of these and other groups is far from determinative; yet it provides a reason to be cautiously optimistic that the sovereign debt community will be able to respond more quickly in the future to any adverse court decisions that might once again threaten to impose large social costs on their market. There are even indications that this may already be happening. [66][66]. For a detailed discussion of the institutional response to the pari passu problem which suggests that institutions “learn” each time they attempt responses, see Anna Gelpern, Ben Heller & Brad Setser, Count the Limbs: Designing Robust Aggregation Clauses in Sovereign Bonds, in Too Little, Too Late: The Quest To Resolve Sovereign Debt Crises (Martin Guzman, José Antonio Ocampo & Joseph E. Stiglitz eds., Columbia Univ. Press 2016). The pari passu case study reveals that when key market players first met at Columbia Law School to discuss the NML court’s interpretation of the pari passu clause in Argentine debt, the meeting was marked by so much “dissension and disagreement” [67][67]. Choi et al., supra note 1, at 40. that the prospect of “any significant movement towards wholesale revision of the clause [appeared] unlikely in the near term.” [68][68]. Id. at 39. Nevertheless, when the Federal Reserve convened a meeting of many of the very same people just a few weeks later, it became clear that “everyone involved needed, and was willing to, cooperate in trying to solve the systemic problem caused by the rogue interpretation of the New York courts.” [69][69]. Id. at 40. Legal change accelerated shortly thereafter. [70][70]. However, as CG&S point out, these revisions occurred only in sovereign debt—not quasi-sovereign debt—which in their view left the market facing significant social costs. Id. at 24–25.

The experience of American trade associations and the sovereign debt community’s own response when they came together at the Federal Reserve suggest that prudence is warranted in abstracting from the saga of the pari passu. It is far from certain that any future errant court decisions about the meaning of black holes in sovereign debt indentures will remain unremedied by collective action for a significant period of time. Only time and additional research on this and other markets will reveal whether the saga of the pari passu is best understood as: (1) an illustration of the transition problems that particular types of markets face when they are confronted with the need to overcome collective action obstacles to contractual change for the first time; (2) an enduring feature of markets where contracts are inter-dependent in ways that make the risk of judicial error more serious than in other markets; or (3) a problem that exists in a wider array of contracting contexts than standard contract theory would predict.

Conclusion

This Comment has explored some of the conceptual and practical problems that make it difficult to devise an effective legal strategy for dealing with black holes. It has also questioned whether or not a legal response to the black hole problem is needed. Still, nothing in the discussion has taken anything away from the core contribution of the black hole articles—namely, the identification of a distinct type of contract provision that poses unique interpretive challenges and may increase the consequences of judicial error in some contracting contexts. This contribution is particularly timely because black holes are likely to become increasingly common as a result of a number of technological changes that facilitate the creation of complex agreements with largely standardized yet slightly variant terms. [71][71]. See, e.g., KMStandards (2016), KMstandards.com [https://perma.cc/5KTY-Z2NJ]; IACCM (2017), IACCM.com [https://perma.cc/YH8G-8D73] (providing contract drafting software that permits the quick and inexpensive drafting of largely standard form contracts while giving transactors a variety of clauses to choose from), the website of the International Association of Contract and Commercial Managers [https://perma.cc/KKP7-3SMZ] (containing information about new contract drafting technologies).

Nevertheless, until more is known about black holes in both the sovereign debt market and other contexts it might be best to refrain from undertaking any general doctrinal legal reforms in response to the phenomenon, especially as the type and magnitude of the problems caused by black holes are likely to be vastly different depending on the market in question.

Moreover, the cost of not responding to the problem may be far lower than it seems. The interpretive approach adopted by the Second Circuit in NML v. Argentina may turn out to be the best or at least a passably good long-run common law response to the black hole problem—especially in markets where the players (individuals, governments, and institutions) are sophisticated and are advised by able counsel. By functioning as an interpretive penalty default rule, [72][72]. Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 91–130 (1989) (describing the theory of the penalty default rule). the Second Circuit’s adjudicative approach creates incentives for transactors and market intermediaries like rating agencies to pay more attention to drafting choices. It may also have the beneficial effect of encouraging new or existing market or quasi-governmental intermediaries to develop the institutional frameworks needed to provide the types of tailored responses to contracting problems that the common law cannot provide without introducing changes that will be vastly over- or underinclusive and will also be likely to increase both litigation costs and the incidence of strategic behavior.

Interestingly, while CG&S remain critical of the NML decision, they now acknowledge that some type of institutional response (whether induced by the common law or encouraged by other types of government action) may, at the end of the day, be the best response to the black hole problem in the long-run. As they explain in reference to the interchange of ideas that led to this Comment:

[T]he true lesson of our study may be that the IMF and other groups that constitute the ‘official sector’ may be better able than courts to solve these problems over time as they gain experience and become more confident in their methods. If this is so, then even though the pari passu case shows that the intervention of a public authority is sometimes required to solve contractual disputes that have third-party effects, relying on the courts rather than private ordering to craft the solution may not always be the best choice. [73][73]. Choi et al., supra note 1, at 69–70 (citing discussions with Bernstein).

Copyright © 2017 Lisa Bernstein.

Wilson-Dickinson Professor of Law and Aaron Director Research Scholar, The University of Chicago Law School, and International Research Fellow, Center for Corporate Reputation, Said School of Business, University of Oxford. I would like to thank Douglas Baird, Patrick Barry, Edward Bernstein, Brian Bix, Sadie Blanchard, Robin Effron, Roger Ford, Anna Gelpern, William Hubbard, Todd Ito, Avery Katz, Kate Lindgren, Bill Schwesig, Mark Ramseyer, Steve Ware, and participants at the Duke Conference on Contractual Black Holes (2017) for helpful conversations. Special thanks are due to the hole theorists themselves for engaging conversations about their piece over the past year.

[1] Stephen J. Choi, G. Mitu Gulati & Robert E. Scott, The Black Hole Problem in Commercial Boilerplate, 67 Duke L.J. 1 (2017).

[2] The other two papers in the Black Hole corpus are Stephen J. Choi, Mitu Gulati, & Robert E. Scott, Contractual Arbitrage, in Oxford Handbook of International Governance (Eric Brosseau, Jean-Michel Galmant & Jerome Sgaard eds., forthcoming 2017) http://scholarship.law.duke.edu/cgi/ viewcontent.cgi?article=6304&context=faculty_scholarship [https://perma.cc/VC3Y-QLKN] and Stephen J. Choi, Mitu Gulati & Robert E. Scott, Variation in Boilerplate: Rational Design or Random Mutation, 19 AM. L. & ECON. REV. (forthcoming 2017) [hereinafter, Choi et al., Variation in Boilerplate].

[3] Choi et al., supra note 1, at 2.

[4] Id. at 6.

[5] Id.

[6] Id.

[7] NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also focuses on an earlier interpretation of the pari passu clause rendered by a Brussels court in 2000. See Elliott Assocs. v. Republic of Peru, 948 F. Supp. 1203 (S.D.N.Y. 1996).

[8] Choi et al., supra note 1, at 7.

[9] Interestingly, the Second Circuit seemed to reject the idea that it was jeopardizing the market as a whole by making restructuring more difficult. See NML Capital, Ltd., 699 F.3d at 263 (explaining why this effect is unlikely to occur.)

[10] Choi et al., supra note 1, at 70 (concluding that “the inherently greater inertia costs [and social costs] that result from an aberrant interpretation of a black hole term,” as compared to an ordinary unclear term, and the “greater difficulty market players face in overcoming the resulting collective action problem[s]” slow the adoption of new and better contractual terms).

[11] Id.

[12] Although it is possible that some of the types of clauses this Comment views as “black hole apparitions” would be classified as genuine black holes by CG&S, their search for additional examples of black holes by holding a conference on the subject at Duke Law School suggests that these ubiquitous types of clauses with uncertain meaning are unlikely to come within the purview of their definition. As Gulati explained, “black holes where meaning has been lost are different from provisions where there is a general sense of what they mean, but there is some vagueness or lack of clarity.” E-mail from Mitu Gulati, Professor of Law, Duke University, to author (Sept. 23, 2017) (on file with author).

[13] In order for a doctrine that conditions on whether a clause is or is not a black hole to be workable, a clause’s status must be verifiable by a court—that is, a court must be able to determine with reasonable accuracy at a cost the transactors consider reasonable from an ex ante perspective whether or not something is a black hole. See generally Alan Schwartz, Relational Contracts and the Courts: An Analysis of Incomplete Agreements and Judicial Strategies, 21 J. Legal Stud. 271 (1992).

[14] Commercial Arbitration Rules and Mediation Procedures (Am. Arbitration Ass’n 2013), https://www.adr.org/sites/default/files/Commercial%20Rules.pdf. [https://perma.cc/E8EW-CBBP].

[15] The AAA discourages written opinions but permits parties to request them. See Charles A. Cooper, Harry Kaminsky & Neil Carmichael, Manual for Commercial Arbitration 162 (Am. Arbitration Ass’n 1999) (“Currently in domestic arbitrations, the AAA does not encourage commercial arbitrators to write opinions which give their reasons for the award. However, in instances where both parties request an opinion prior to the appointment of the arbitrator, the arbitrator should comply.”). While transactors might opt to revise a clause in face of a seemingly aberrant arbitration ruling, they might also surmise, much as the sovereign debt community did after the ruling of the Brussels court, that a more sensible panel (or as in the case of the pari passu clause the Second Circuit Court of Appeals) would rule differently. See Choi et al., supra note 1, at 12.

[16] But see generally W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (suggesting that in some contexts arbitration decisions that are not technically precedent nevertheless have an effect on the outcome of future disputes). In addition, there are indications that companies fear that arbitral decisions will be given some weight if they have future disputes with the same counter-party. For example, Verizon’s consumer agreement provides for AAA commercial arbitration, includes a class action waiver, and also contains a clause stating that “[a]n arbitration award and any judgment confirming it apply only to that specific case; it can’t be used in any other case except to enforce the award itself.” Customer Agreement, Verizon Wireless (Oct. 20, 2017), https://www.verizonwireless.com/legal/notices/customer-agreement/ [https://perma.cc/3A2L-6YGT]; see also What’s Going on? Arbitration Explanation, Waste Mgmt., http://mediaroom.wm.com/legal-notice/ [https://perma.cc/TZ3C-X4GB] (“The award of the arbitrator may be entered and enforced in any court having jurisdiction. An arbitration award and any judgment confirming it apply only to that specific case; it cannot be used in any other case except to enforce the arbitration award itself.”).

[17] For example, trade association-run private legal systems are designed in ways that will strongly discourage the emergence of black holes. At the National Grain and Feed Association, for example, arbitrators write reasoned opinions that are posted on the association’s website. When a rule is unclear, or appears to lead to an undesirable or unanticipated result, the arbitrators will state this explicitly, apply the rule as it is stated, and note that the association’s rules committee should consider revising the rule, something that is typically done. See Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1780 & nn.50¬–51 (1996).

[18] Steven M. Barken, Barbara A. Bintliff & Mary Whisner, Fundamentals of Legal Research 41 (10th ed. 2015) (“Ordinarily, cases decided by state trial courts are not reported. . . . Only a few states, such as New York, Ohio, Pennsylvania, and Virginia, publish some trial court opinions, but those selected are few in number and represent only a very small portion of the total cases heard by trial courts.”). A similar problem may exist with respect to cases decided in federal court. See David A. Hoffman, Alan J. Izenman & Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash. U. L. Rev. 681, 727 (2007) (“[S]tarkly expos[ing] how little trial court work is explained through written opinions.”).

[19] It is, however, possible that the doctrinal contours of the black hole or no black hole determination would become clearer over time because under the proposed change this would be determined as a “matter of law.” See Choi et al., supra note 1, at 68.

[20] CG&S focus mostly on standard commercial provisions in standardized contracts. However, both black holes and black hole apparitions can arise in seemingly bespoke contracts as well. To the extent this phenomenon exists, the proposed reform, in as much as it seems to apply only to standard contracts with boilerplate provision, would be underinclusive with respect to the types of holes found in these seemingly bespoke agreements.

[21] The way that a contract is negotiated has an effect on how well the business deal can be operationalized once the contract is signed. See Danny Ertel, Getting Past Yes: Negotiating as if Implementation Mattered, 85 Harv. Bus. Rev. 60, 62 (2004).

[22] For a more complete discussion of this bargaining dynamic and the way that it effects the terms of agreements, see generally Lisa Bernstein, Social Norms & Default Rules Analysis, 3 S. Cal. Interdisc. L.J. 59 (1993).

[23] Gillian K. Hadfield & Iva Bozovic, Scaffolding: Using Formal Contracts To Build Informal Relations to Support Innovation, 2016 Wis. L. Rev. 981, 982, 987 (noting that Stewart Macaulay’s finding that in commercial contracting relationships problems tended to be worked out informally and “written contracts . . . were often highly-standardized documents that were often confined to the drawer,” held in a modern day sample of non-innovative firms where informal methods of dispute resolution remained common and firms either “did not generate [formal contracts] or relied only on standardized documents,” but did not hold in a sample of innovative firms).

[24] See generally Lisa Bernstein, Beyond Relational Contracts: Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561 (2015) (explaining how widely known and standardized contract provisions may affect the strength of network governance).

[25] In some markets where reputation-based network governance plays an important role in contracting, some large companies post their quality requirements and their standard terms and conditions on their website so that the contours of what is and is not expected from their counterparties is known throughout the market. See, e.g., Supplier Quality Manual, John Deere, https://jdsn.deere.com/wps/portal/jdsn/Applications?WCM_GLOBAL_CONTEXT=/wps/wcm/connect/jdsn_website/jdsn/business+processes/quality/supplier_quality_manual/supplier_quality_manual_index [https://perma.cc/57SG-BQUK]; Terms and Conditions for the Purchase of Goods and Services, John Deere, https://www.johndeerestore.com/jdb2cstorefront/JohnDeereStore/en/terms [https://perma.cc/B7DH-QPQU].

[26] In such markets, general and impressionistic information about whether a transactor behaved properly or poorly could still be easily transmitted, but in the absence of standard terms, a consensus about what constitutes good behavior, or information about both what was promised and what was done, this information is likely to be given less weight (especially when it travels more than one step from its source) than information that circulates in markets where contracts and/or expectations are either relatively standardized or widely known. Bernstein, supra note 24, at 578–89 (discussing how John Deere achieves this).

[27] Sometimes the line between relational and rational black holes may be blurry. For example, Scott and Triantis note that it is often economically rational to use very standard-like terms like “good faith,” “reasonable efforts,” or “best efforts,” and leave it for the court to give these terms meaning ex post. Robert E. Scott & George Triantis, Anticipating Litigation in Contract Design, 115 Yale L.J. 814, 835 (2006). However, once those terms become common in an industry’s contracts (or between a pair of long-term repeat dealers), their presence in contracts may be exceptionally sticky. If, for example, someone sends their counterparty a draft obligating the counterparty (and themselves) to act in good faith, the relational costs of saying to one’s counter party, “well let’s spell that out” or “let’s leave that out” might be high.

[28] For a discussion of rational reasons why many clauses are left vague or undefined, see Scott & Triantis, supra note 27, at 814–15. The literature conceptualizes these terms as agreements-to-agree or agreements to determine meaning of a term through negotiation or litigation if an interpretive dispute arises.

[29] For a complete discussion of the ways that transactors’ “secrecy interest” in certain types of business information may affect the credibility of their threat to sue in a variety of contexts, see generally Omri Ben Shahar & Lisa Bernstein, The Secrecy Interest in Contract Law, 109 Yale L.J. 1885 (2000).

[30] In contracting relationships like those between an Original Equipment Manufacturer and its suppliers of component parts where a lawsuit tends to end the parties contracting relationship, many contractual provisions that are formally enforceable may not operate in the shadow of the law at all. The reason is simple: if one of these provisions were breached in isolation, the promisee would not have a credible threat to sue since the expected recovery would be very unlikely to exceed the future value of the relationship as a whole. These “interior contract provisions” operate outside of the shadow of the law, unless and until they are breached in combination with enough other provisions that the counterparty concludes that there has been a breach of the contracting relationship that makes it worthwhile to end the transactors’ business relationship and sue. See Bernstein, supra note 24, at 570–71.

[31] At one point in their analysis CG&S hint that it might be desirable to apply a knock-out rule to black hole terms akin to the rule adopted by some courts applying U.C.C. § 2-207. However, this rule will also require black holes to be identified by courts and will therefore be unworkable for many of the same reasons discussed herein.

[32] See Sadie Blanchard, Courts Without Enforcement: Adjudicating Reputation in the Sovereign Debt Market 6–7 (Oct. 27, 2017) (unpublished manuscript) (concluding that “[e]ven though [the holders of sovereign debt] cannot enforce debtor states’ obligations through conventional judicial means, courts play a key role in the sovereign debt market . . . because they provide information that has the power to provoke reactions by third parties that are costly for the debtor or its agents,” so that “[c]reditors litigate because producing such information through the courts strengthens their leverage in settlement negotiations,” and makes reputation a more powerful force in the market).

[33] See Julian Schumacher, Christoph Trebesch & Henrik Enderlein, Sovereign Defaults in Court 10–11 (May 6, 2014) (unpublished manuscript), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2189997 [https://perma.cc/T2Q7-LNGN], (finding that only 5 percent of sovereign defaults resulted in litigation until the 2000s, when the proportion rose to nearly half).

[34] See W. Mark C. Weidermaier, Sovereign Immunity and Sovereign Debt, 2014 U. Ill. L. Rev. 67, 68–70 (2014).

[35] Cours d’Appel de Bruxelles, 8th Chamber, 2000/QR/92, Elliott Assocs. v. République de Peru, 2000/QR/92, Sept. 26, 2000.

[36] NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012). The black hole corpus also looks at the market’s nonresponse to a pari passu decisions of a Brussels court, but attributes it to market players’ beliefs that other, more important courts for the sovereign debt community would not follow such an aberrant decision.

[37] Choi et al., supra note 1, at 49.

[38] As CG&S point out, it might not have been individually rational for any one issuer to make these changes to their past and future indentures, even though it would have been rational for the market as a whole to move to a different term. Id. at 13.

[39] Id. at 37 (“We use both qualitative and quantitative data to support the claim that courts searching for shared intent in the case of black holes in standardized contracting can result in substantial social costs.”).

[40] Id. at 67.

[41] Id. at 68.

[42] See id. at 66–67, 69 Exactly what this alternative interpretive approach would be is not specified in great detail. See id. at 54–56. If, instead, some form of knock-out rule were contemplated, similar problems would arise. See supra note 31.

[43] Choi et al., supra note 1, at 38.

[44] Id. at 68.

[45] Indeed, the methodology CG&S used to reach the conclusion that the pari passu provision was a black hole, rather than an efficient contract term, relied on extensive information about the market’s response to the decision over a period of years. See id. at 11.

[46] The types of strategic behavior it would introduce are analogous to those identified in Alan Schwartz and Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 587 (2000) and Lisa Bernstein, Custom in the Courts, 110 Nw. L. Rev. 63, 106–08 (2015).

[47] Choi et al., supra note 1, at 68.

[48] Id. at 55–56, 64.

[49] Id. at 1.

[50] Choi et al., Variation in Boilerplate, supra note 2, at 4.

[51] CG&S undertake just this inquiry in another part of the black hole corpus. See id. at 6. There, the authors seek to demonstrate that the observed variations in the pari passu clauses used in sovereign debt contracts are the result of random mutation rather than rational design. The information and analysis they present is a good proxy for how difficult it would be for a transactor to litigate this distinction in court. Moreover, not all of the information they refer to could be known at trial—for example, how the market would respond if the court gave weight to a small drafting difference.

[52] When real black holes actually exist, parties are likely to be largely unconstrained with respect to the range of plausible subjective intents they can proffer. As a consequence, the assumptions made in the text are likely to characterize a meaningful subset of the cases that might arise.

[53] See, e.g., Brief of Defendant-Appellant the Republic of Argentina, NML Capital, Ltd. v. Republic of Argentina, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2012 WL 6777133; Joint Response Brief of Plaintiffs-Appellees NML Capital, LTD and Olifant Fund, LTD, 699 F.3d 246, 257 (2d Cir. 2012) (Nos. 12-105cv(L) et al.), 2013 WL 388621.

[54] CG&S are careful to delineate a number of reasons that the collective action problems they observe in the sovereign debt markets where the contracts at issue—bond indentures—are in practice closer to the pole of financial instruments, may be particularly severe compared to the collective action problems that might or might not impede the revision of ordinary contracts for sales of goods or provision of services when a court makes an aberrant interpretation of one of their terms. See Choi et al., supra note 1, at 59–66.

[55] The reform could take the form of a new sub-chapter of Article 2 akin to Article 2A on leases, or an additional subset of rules in Article 2 akin to its merchant rules.

[56] While such a change would raise the issue of how to decide which contracts would come within the new Article’s purview, it could be made available to the parties on an opt-in basis at the time of contracting, subject to rejection by a court that found the choice to be inappropriate.

[57] For an overview of the problems with these private uniform law-making bodies, see generally Alan Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. Pa. L. Rev., 595–654 (1995). For an example of the difficulties of creating new Articles of the UCC in particular, consider the largely failed effort to get states to adopt the Uniform Computer Information Transactions Act, which has been adopted by only two states. See The Uniform Computer Information Transactions Act (UCITA) Is a Proposed State Contract Law, UCITA Online, http://www.ucitaonline.com/ [https://perma.cc/8SUP-W2L8] (discussing the impediments to adopting the model law).

[58] The best available, though imperfect, empirical evidence suggests that sophisticated commercial parties prefer textualist adjudication. See Lisa Bernstein, Merchant Law in a Modern Economy, in Philosophical Foundations of Contract Law 251 (Gregory Klass, George Letsas & Prince Saprai eds., Oxford Univ. Press 2014) (noting the common use of plain meaning clauses in large commercial contracts providing for arbitration); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 Mich. L. Rev. 1724 (2001) (documenting the cotton industry’s preference for a textualist/formalist adjudicative approach); see also Bernstein, Merchant Law in a Modern Economy, supra (documenting the grain and feed industry’s preference for a textualist/formalist legal approach); Theodore Eisenberg & Geoffrey P. Miller, The Flight to New York: An Empirical Study of Choice of Law and Choice of Forum Clauses in Publicly-Held Companies’ Contracts, 30 Cardozo L. Rev. 1475 (2009) (demonstrating commercial actors’ strong preferences for relatively formalist New York Law); Stuart Popham, The View of European Business: Survey Results Presentation, Clifford Chance LLP (Mar. 14, 2008), http://denning.law.ox.ac.uk/news/events_files/Popham_-_presentation.PPT [https://perma.cc/DRB3-KXAL] (finding that in business contracts that provided for arbitration at the International Chamber of Commerce, transactors preferred British law, the most formalistic and textualist of the EU alternatives).

[59] See generally Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 Yale L.J. 926 (2010) (defending a formalist approach to contract interpretation).

[60] Another potential direction for reform that could be implemented without changing the common law, but that depends critically on the ability of groups—like state legislatures, the ALI, the NCCUSL, trade associations, and a variety of private entities and intermediaries—to revise contractual language when courts make errant decisions, would be to selectively adopt the reform proposals of the legal choice theorists. See Hanoch Dagan & Michael Heller, The Choice Theory of Contracts 102–13 (Cambridge Univ. Press 2017). These theorists explore the benefits that might accrue when states and other private and public entities provide more “contract types” that transactors could use to structure their affairs—much as Delaware law offers those structuring a business the option of incorporating as a public entity, a close corporation, an LLC or a partnership. If these “types” were carefully developed, and proved easier to change in the event of an errant judicial interpretation of their terms than individually structured agreements, heeding the call for the creation of these types may help solve the black hole problem in some contracting contexts. See infra notes 61–65 and accompanying text (noting that trade associations have proved remarkably able to amend their trading rules and standard form contracts in response to arbitral interpretations of these terms). But see Schwartz & Scott, infra note 59 (discussing the problems that groups like the ALI have in revising their model laws).

[61] See Lisa Bernstein, The Questionable Empirical Basis of Article 2’s Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev. 710, 730–32 (1999) (providing a short history of the Textile Trading Rules).

[62] Id. at 732.

[63] Id. at 736 (quoting SAA, Thirty-Fifth Annual Report 23 (1907)).

[64] Id. (quoting SAA, Fortieth Annual Report 31 (1912), Revision, 3 Silkworm 73 (May 1921)).

[65] Id. (quoting SAA, Fifty-Third Annual Report 26 (1925)).

[66] For a detailed discussion of the institutional response to the pari passu problem which suggests that institutions “learn” each time they attempt responses, see Anna Gelpern, Ben Heller & Brad Setser, Count the Limbs: Designing Robust Aggregation Clauses in Sovereign Bonds, in Too Little, Too Late: The Quest To Resolve Sovereign Debt Crises (Martin Guzman, José Antonio Ocampo & Joseph E. Stiglitz eds., Columbia Univ. Press 2016).

[67] Choi et al., supra note 1, at 40.

[68] Id. at 39.

[69] Id. at 40.

[70] However, as CG&S point out, these revisions occurred only in sovereign debt—not quasi-sovereign debt—which in their view left the market facing significant social costs. Id. at 24–25.

[71] See, e.g., KMStandards (2016), KMstandards.com [https://perma.cc/5KTY-Z2NJ]; IACCM (2017), IACCM.com [https://perma.cc/YH8G-8D73] (providing contract drafting software that permits the quick and inexpensive drafting of largely standard form contracts while giving transactors a variety of clauses to choose from), the website of the International Association of Contract and Commercial Managers [https://perma.cc/KKP7-3SMZ] (containing information about new contract drafting technologies).

[72] Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 91–130 (1989) (describing the theory of the penalty default rule).

[73] Choi et al., supra note 1, at 69–70 (citing discussions with Bernstein).