Contract Interpretation Preferences: A Textualist Approach in Commercial Contracts, Study notes of Law

Download Contract Interpretation Preferences: A Textualist Approach in Commercial Contracts and more Study notes Law in PDF only on Docsity! 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 469 THE INTERPRETATION OF COMMERCIAL CONTRACTS: AN EMPIRICAL STUDY Uri Benoliel* ABSTRACT .................................................................................................. 470 INTRODUCTION .......................................................................................... 470 I. THE THEORETICAL DEBATE: TEXT VS. CONTEXT ................................. 472 II. EXISTING EMPIRICAL EVIDENCE ........................................................... 477 III. THE EMPIRICAL TEST ........................................................................... 480 A. Merger Clause—A Brief Overview .......................................... 481 B. The Theoretical Hypotheses .................................................... 482 C. Data ......................................................................................... 485 D. Methodology ............................................................................ 488 E. Results...................................................................................... 489 IV. DISCUSSION AND NORMATIVE IMPLICATIONS ..................................... 491 CONCLUSION .............................................................................................. 493 * Faculty of Law, College of Law & Business. J.S.D. (UC Berkeley); LL.M (Columbia University). I am grateful to Adi Ayal, Samuel Becher, Steven Burton, Hanoch Dagan, Omer Dekel, Sinai Deutch, Christopher Drahozal, Charles Goetz, Alon Harel, Eyal Katvan, Rinat Kitai, Pablo Lerner, Geoffrey Miller, Gideon Parchomovsky, Ariel Porat, Arie Reich, Boaz Sangero, and James White for invaluable comments on earlier drafts of this Article. This Article is dedicated, with love, to my father, Professor Ricardo Ben-Oliel. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 470 Alabama Law Review [Vol. 69:2:469 ABSTRACT The theoretical debate over contract interpretation revolves around one central question: what is the preference of most contracting parties for contract interpretation rules? Textualist theorists believe that most parties prefer textualist rules of interpretation, under which the contract interpreter must normally consider only the contract’s written text. In contrast, contextualist theorists believe that most parties prefer contextualist rules of interpretation, under which the interpreter should consider all relevant contextual evidence to interpret the contract, beyond the written text. Despite the widespread debate over contract interpretation, there has been very little empirical research on this topic. This Article aims to fill this research gap by empirically analyzing actual interpretation clauses of commercial contracts. Examining 1,521 commercial contracts that have been disclosed to the Securities and Exchange Commission, this Article finds that a clear majority (75.28%) of contracts include a textualist “merger clause,” which typically triggers a set of textualist rules of contract interpretation. In addition, the results of this study indicate that the merger clauses, included in the sample contracts, are not mere arbitrary boilerplates which were randomly added to the contracts. More specifically, the study found a significant statistical association between the contractual existence of a variety of textualist contractual clauses, other than a merger clause, and the existence of the textualist merger clause. The theoretical and practical implications of these results are discussed. INTRODUCTION Contract interpretation—that is, the undertaking by an adjudicator to identify the terms of the contract and give them a meaning1—plays a significant role in American law.2 It is one of the most common sources of 1. Richard A. Posner, The Law and Economics of Contract Interpretation, 83 TEX. L. REV. 1581, 1582 (2005) (“Contract interpretation is the undertaking by a judge or jury (or an arbitrator—more on arbitration later) to figure out what the terms of a contract are, or should be understood to be.”); see also Steven J. Burton, A Lesson on Some Limits of Economic Analysis: Schwartz and Scott on Contract Interpretation, 88 IND. L.J. 339, 341 (2013). 2. STEVEN J. BURTON, ELEMENTS OF CONTRACT INTERPRETATION 1 (2009) (“Issues of contract interpretation are important in American law.”); Avery Wiener Katz, The Economics of Form and Substance in Contract Interpretation, 104 COLUM. L. REV. 496, 496 (2004) (“Under the modern American law of contracts, almost all applications of legal doctrine turn on questions of interpretation . . . . ”); Joshua M. Silverstein, Using the West Key Number System as a Data Collection and Coding Device for Empirical Legal Scholarship: Demonstrating the Method Via a Study of Contract Interpretation, 34 J.L. & COM. 203, 204 (2016) (“Contract interpretation is one of the most significant areas of commercial law.”). 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 473 contract’s text.16 The adjudicator must specifically exclude the following major categories of contextual, nontextual, extrinsic evidence: (1) practice between the parties under prior contracts; (2) practice between the parties under the litigated contract; (3) precontractual oral statements or understandings; and (4) industry custom.17 According to the contextualist theory, the interpreter should consider all relevant contextual evidence to interpret the contract, beyond the written contractual text.18 The interpreter must look at “events before contract formation.”19 The interpreter must also consider events that occurred after contract formation.20 The interpreter should consider contextual evidence, even if it is oral or behavioral and nontextual.21 The adjudicator should determine “whether extrinsic evidence of the circumstances surrounding the contract . . . improves understanding of what parties intended regardless of the contractual text.”22 The adjudicator should consider context, even if the contract seems unambiguous.23 The written contractual language is “treated merely as establishing prima facie terms,” which the adjudicator can override by considering contextual, nontextual evidence if she believes 16. Gilson et al., supra note 15, at 171 n.1 (“In a textualist regime, and absent ambiguity, generalist courts cannot choose to consider context . . . .”); see also Gilson et al., supra note 3, at 25; Darius Palia & Robert E. Scott, Ex Ante Choice of Jury Waiver Clauses in Mergers, 17 AM. L. & ECON. REV. 566, 572 (2015) (“[T]he textualist approach bars context evidence . . . .”); Alan Schwartz & Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 YALE L.J. 541, 572 n.61 (2003) (a court applying a textualist approach “will admit extrinsic evidence only when the contract’s language is vague or ambiguous on its face”). 17. Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 YALE L.J. 926, 933 n.20 (2010); see also CATHERINE MITCHELL, INTERPRETATION OF CONTRACTS 123 (2007) (“Formalism may manifest itself in a desire for the documents to be taken as the primary evidence of what was agreed, without recourse to negotiations, trade customs, previous understandings or any other extrinsic material.”). 18. Charles J. Goetz & Robert E. Scott, The Limits of Expanded Choice: An Analysis of the Interactions Between Express and Implied Contract Terms, 73 CALIF. L. REV. 261, 308 (1985) (“[T]he contextualists have assumed that the purpose of interpretation is to uphold the expectations of the particular parties to the agreement by determining from an analysis of all relevant evidence what they ‘really meant.’”); Katz, supra note 2, at 498 (“A more ‘substantive’ approach to contract interpretation . . . would attempt to come to a more all-things-considered understanding, based on all of the materials reasonably available.”); Schwartz & Scott, supra note 16, at 572 (“[Contextualists’] theory lets courts consider all material evidence to resolve interpretive issues . . . .”). 19. Melvin Aron Eisenberg, The Emergence of Dynamic Contract Law, 88 CALIF. L. REV. 1743, 1770 (2000). 20. Id.; see also Bayern, supra note 15, at 1100 (“[C]ontextualists consider post-formation information . . . .”). 21. Scott, supra note 4, at 1 (“[C]ontextualist theories look beyond the writing, to . . . oral . . . evidence of what the parties intended.”). 22. Gilson et al., supra note 3, at 27. 23. Gilson et al., supra note 15, at 171 n.1 (“In a textualist regime, and absent ambiguity, generalist courts cannot choose to consider context; in a contextualist regime, these courts must consider it.”); Gilson et al., supra note 3, at 25–26. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 474 Alabama Law Review [Vol. 69:2:469 that doing so is needed in order to understand the parties’ actual intentions.24 A central theoretical argument that underlies the textualist theory is that given the potential benefits of a textual interpretation approach, most parties probably prefer this approach over the contextualist approach.25 The major benefits that arguably underlie the parties’ preference for textualism are the following: 1. The textualist regime has lower litigation costs compared to a contextualist regime.26 First, a contextualist approach is likely to increase litigation over contract interpretation compared with a textualist regime. Contextualism creates a new basis for dispute over the existence of each bit of context, its relative weight vis- à-vis other contextual bits, and its relative weight vis-à-vis the contract text.27 Second, under a contexualist approach, contrary to the textualist approach, courts must thoroughly examine all relevant contextual evidence.28 This task may require the costly hearing of expert testimony and witnesses.29 Conversely, the textualist approach allows enforcement of unambiguous contract terms by summary procedures.30 Accordingly, “litigation is more costly in a contextualist . . . regime because the parties more frequently will have full trials.”31 2. The textualist approach reduces the risk of judicial error created by the contextual approach.32 Under a contextualist regime, two judicial errors may occur. First, by considering infinitely elastic context, the interpreter might wrongly interpret the contract contrary to the true intention of the parties.33 Second, relying on 24. Scott, supra note 4, at 2. 25. Schwartz & Scott, supra note 17, at 941 (“[T]he rules should reflect the parties’ preferences; and . . . parties prefer textualist interpretive defaults.”). 26. Cf. MITCHELL, supra note 17, at 109–10; Schwartz & Scott, supra note 17, at 930. 27. Cf. MITCHELL, supra note 17, at 112 (“[L]itigation over terms and obligations is actually encouraged . . . by courts adopting a contextual approach . . . in relation to terms . . . .”). 28. See supra notes 18–21 and accompanying text. 29. MITCHELL, supra note 17, at 110 (under a contextualist regime, “[e]xpert testimony may have to be adduced, preliminary hearings may be required on matters of evidence and procedure and so on”). 30. Cf. Scott, supra note 15, at 376 (“Contextual interpretation . . . . prevents enforcement of even apparently clear obligations by summary procedures.”). 31. Schwartz & Scott, supra note 16, at 587. 32. Id. (“A plain-meaning linguistic default . . . would reduce the risk of judicial error.”). 33. MITCHELL, supra note 17, at 110, 115 (“[T]he greater the amount of contextual material, the greater the possibility for error. Decision-makers may easily become ‘bewildered by a large set of conflicting evidence’ . . . . The contextual approach arguably increases the chances for error by increasing the amount of information deemed relevant to the interpretation exercise.”); Schwartz & 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 475 contextual evidence may generate a contract misinterpretation, since the parties may actually have intended that their contract text will serve as the only interpretive tool.34 3. The textualist method, compared with the contextualist method, increases the ability of the parties to predict “how contract terms and language will be interpreted in [their] subsequent transactions.”35 By excluding contextual evidence when the contract text is unambiguous, the textualist approach preserves the linguistic clarity of existing unambiguous terms.36 4. The textualist approach prevents opportunistic behavior that might occur under a contextualist regime. Under the latter regime, a contract party might strategically dispute the meaning of a perfectly clear contract term, to which she freely agreed, in an effort to escape a bad bargain.37 The longer the contract, the easier it will be to strategically create disputes regarding its meaning.38 5. The textualist approach prevents the adjudicator from imposing his own set of beliefs on the contract by requiring him to follow only the contract text. Conversely, under a contextualist regime, “the interpreter necessarily imposes his own set of assumptions” on the parties’ contract by “selecting certain bits of context” and excluding others.39 Scott, supra note 16, at 587 (“[A] disappointed party may plausibly claim that the parties’ course of dealing or their oral negotiations showed that, in the parties’ language, ‘all’ meant ‘some’ . . . When such a claim is false but found to be true, the court necessarily will misinterpret the contract.”). 34. Scott, Text Versus Context, supra note 4, at 16 n.40 (“But sometimes the parties may actually have intended that their clear language should be read in the standard (plain meaning) way despite the fact that the language itself conflicts with the prior practices and negotiations of the parties. In such a case, a court that relies too heavily on context risks misinterpreting the parties’ actual intentions.”). 35. Cf. Scott, supra note 15, at 376. 36. Gilson et al., supra note 3, at 40–41; Scott, Text Versus Context, supra note 4, at 17 (“By insulating the standard meaning of terms from deviant interpretations, this strategy preserves a valuable collective good, namely a set of terms with a clear, unambiguous meaning that is already understood by the vast majority of commercial parties.”). 37. MITCHELL, supra note 17, at 113 (“One may use the ‘context’ to seek an unbargained for advantage in imposing terms after the parties are in a contractual relationship, even in circumstances where the written terms appear relatively complete.”); Scott, supra note 15, at 377 n.18 (“Here the risk is that, unless the court privileges the written agreement by excluding the contextual evidence, parties . . . will be motivated to dispute the meaning of perfectly communicative contract terms as a strategic response to a now disfavoured contract.”). 38. Schwartz & Scott, supra note 16, at 587 (“[T]he more complex the contract, the easier it will be to create disputes regarding what the contract says and what language it was written in.”). 39. Goetz & Scott, supra note 18, at 308 n.125. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 478 Alabama Law Review [Vol. 69:2:469 run private commercial law systems.51 In the first empirical research, Bernstein presents “a case study of the private legal system created by the National Grain and Feed Association (NGFA) to resolve contract disputes among its members.”52 The study shows that “NGFA arbitrators take a formalistic approach to adjudication.”53 They do not allow trade usage, course of dealing, and course of performance “to vary either trade rules or written contractual provisions.”54 In another study, Bernstein presents “a detailed case study of contractual relations in the cotton industry.”55 According to the study, most such relations are subject to arbitration in one of several cotton tribunals.56 The study furthermore shows that cotton arbitrators “use a relatively formalistic adjudicative approach that gives little explicit weight to elements of the contracting context.”57 The reliance of textualist scholars on Bernstein’s case studies suffers from one central limitation. The grain, feed, and cotton industries, empirically examined by Bernstein, may not be representative of other dominant commercial industries, such as health, construction, energy, and IT. Therefore, the results of Bernstein’s case studies do not allow generalizations about the interpretation preferences of parties to commercial contracts.58 Textualist scholars further base their theoretical assumption that parties to commercial contracts prefer a textualist approach on Professors Theodore Eisenberg and Geoffrey Miller’s quantitative empirical study on, inter alia, choice-of-law clauses.59 This important empirical study 51. For textualist scholars who rely on Prof. Bernstein’s empirical research, see, for example, Jody S. Kraus & Robert E. Scott, Contract Design and the Structure of Contractual Intent, 84 N.Y.U. L. REV. 1023, 1102 (2009); Geoffrey P. Miller, Bargains Bicoastal: New Light on Contract Theory, 31 CARDOZO L. REV. 1475, 1477 (2010); Schwartz & Scott, supra note 17, at 956; Schwartz & Scott, supra note 16, at 576 n.66 (referring to Lisa Bernstein’s empirical scholarship, Professors Schwartz and Scott argue that “[t]here is considerable evidence that firms prefer a formalist adjudicatory style”); Scott, supra note 15, at 378 & n.21; Silverstein, supra note 2, at 278–79 (“Textualism is frequently defended on the ground that businesses prefer that method of construction. This view finds support in the work of Lisa Bernstein.”). 52. Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. PA. L. REV. 1765, 1769 (1996). 53. Id. at 1769–70. 54. Id. 55. Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms and Institutions, 99 MICH. L. REV. 1724, 1725 (2001). 56. Id. at 1724 (“[M]ost such contracts are concluded under one of several privately drafted sets of contract default rules and are subject to arbitration in one of several merchant tribunals.”). 57. Id. at 1735. 58. Burton, supra note 1, at 347 n.64 (arguing that Bernstein’s “two case studies of arbitration practices in two commodities markets . . . cannot be easily generalized . . . .”). 59. For textualist scholars who rely on Professors Theodore Eisenberg and Geoffrey Miller’s empirical scholarship see, for example, Lisa Bernstein, Custom in the Courts, 110 NW. U. L. REV. 63, 109 (2015); Kraus & Scott, supra note 51, at 1102–03; Miller, supra note 51, at 1477–78; Schwartz & Scott, supra note 17, at 956–57; Bernstein, supra note 12, at 15–16. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 479 examines, among other things, choice-of-law clauses in a data set of “contracts contained as exhibits in Form 8-K filings by reporting corporations over [a] six month period in 2002 for twelve types of contracts and a seven month period in 2002 for merger contracts.”60 The results of the study show that the parties examined in the study chose New York law in approximately 46% of the contracts,61 while California was chosen for its law in less than 8% of the contracts.62 Since New York’s contract- interpretation law is inclined towards textualism,63 while California’s contract-interpretation law is inclined towards contextualism,64 textualist scholars believe that Eisenberg and Miller’s results indicate that parties to commercial contracts prefer the textualist approach of contract interpretation.65 The reliance of textualist scholars on Eisenberg and Miller’s empirical study suffers from one central limitation. There are many legal differences between New York law and California law, besides the differences in the rules of contract interpretation.66 Just within the realm of contract law, the differences between New York and California concern the application of many important noninterpretation doctrines, such as promissory estoppel, consideration, duress, unconscionability, public policy, and mistake.67 Hence, the dominance of New York choice-of-law clauses over California choice-of-law clauses—as witnessed in Eisenberg and Miller’s study— does not necessarily result from the parties’ preference for New York’s textualist interpretation rules.68 Generally put, the scant existing empirical literature assessing parties’ preferences of contract interpretation rules focuses on indirect indicators, 60. 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, 1475 (2009). 61. Id. at 1489. 62. Id. at 1490. 63. Miller, supra note 51, at 1478; Schwartz & Scott, supra note 17, at 956; Bernstein, supra note 12, at 15. 64. Miller, supra note 51, at 1478; Schwartz & Scott, supra note 17, at 956; Bernstein, supra note 12, at 15. 65. Miller, supra note 51, at 1478; Schwartz & Scott, supra note 17, at 956; Bernstein, supra note 12, at 15–16. 66. See, e.g., Miller, supra note 51, at 1479–1522. 67. Id. at 1482–84, 1485–1502, 1504–06. 68. Bayern, supra note 15, at 1122 (“[T]here are many provisions of substantive New York law that public firms might favor; an inference that they are specifically choosing textualism is unfounded.”); Burton, supra note 1, at 347 n.64 (arguing that “firms’ frequent use of choice-of-law clauses to select New York law . . . could be made for any of a variety of reasons” other than a preference for textualist contract interpretation rules); Silverstein, supra note 2, at 280 n.417 (“But Miller’s article identified roughly seventeen doctrinal differences between New York and California, only one of which was contract interpretation. Thus, it is far from clear that differences in interpretive regimes played an important role in the choice of law and forum decision-making that Eisenberg and Miller studied.” (citation omitted)). 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 480 Alabama Law Review [Vol. 69:2:469 such as the method of adjudication applied by arbitrators in a small number of industries (Professor Bernstein’s studies) or the choice of law clauses selected by public companies (Professors Eisenberg and Miller’s study). In order to avoid the pitfalls of indirect inference from a limited number of industries, this empirical study sought a database that would provide for more direct evidence of parties’ preferences for contract interpretation rules. This study, therefore, empirically examines the frequency with which textualist interpretation clauses are included in contracts without limiting the study to a small number of specific industries. III. THE EMPIRICAL TEST From a methodological perspective, it is difficult to “measure the extent of parties’ preferences for [textualist] adjudication by looking at their contracts.”69 This is due, in part, to the fact that most interpretive legal rules are mandatory;70 namely the parties normally “cannot contract directly for [a] textualist or [a] contextualist” interpretation approach.71 However, there is one central exception in which the parties can contractually choose a preferred interpretation rule: they can include a merger clause in their contract.72 Merger clauses in commercial contracts between sophisticated parties are normally enforceable by courts.73 This study, therefore, will empirically focus on this type of important interpretation clause. 69. Bernstein, supra note 12, at 12. 70. Bernstein, supra note 12, at 12 (“[T]he [U.C.C.’s] contextualist interpretive approach is, in practice, quasi-mandatory.”); Schwartz & Scott, supra note 16, at 583, 585 n.84 (“Courts in general . . . treat interpretation rules as mandatory . . . . The current interpretive rules are mandatory . . . .”); Scott, Text Versus Context, supra note 4, at 8 (“Contract interpretation rules are . . . mandatory . . . .”). 71. Scott, Text Versus Context, supra note 4, at 8, 21; see also Schwartz & Scott, supra note 16, at 583 (under a mandatory regime, “courts, not parties, should choose the rules that determine how contracts are read.”). 72. Bayern, supra note 15, at 1136 (“[T]he enforceability of strong merger clauses, suggests such a mandatory rule is not universal.”). 73. SCOTT & KRAUS, supra note 44, at 543 (merger clauses are “[i]n principle . . . enforceable at common law and under the Code”); JAMES J. WHITE & ROBERT S. SUMMERS, UNIFORM COMMERCIAL CODE § 2-12 (5th ed. 2000) (“‘Merger’ clauses . . . are generally valid.”); E. Allan Farnsworth, The Interpretation of International Contracts and the Use of Preambles, 2002 INT’L BUS. L.J. 271, 273 (parties who include a merger clause in their contract “can be confident that it will be respected by any judge or arbitrator applying a common law system”); Meredith R. Miller, Contract Law, Party Sophistication and the New Formalism, 75 MO. L. REV. 493, 503 (2010) (“If the parties are deemed sophisticated, the merger clause controls.”). 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 483 costs of a judicial error in the evaluation of precontractual evidence of commercial contracts, under a contextual approach, is likely to be high since the value of these contracts is normally significant.86 Most parties to commercial contracts between relatively sophisticated parties will include a merger clause in their contract for another reason. Merger clauses are likely to reduce commercial companies’ intrafirm agency costs.87 Sophisticated parties to commercial contracts often employ many contracting agents, who enter into numerous contracts every day.88 These agents may unintentionally give precontractual statements to the other party, which are not contained in their companies’ contract texts.89 A merger clause, by excluding precontractual evidence, relieves commercial companies of the need to monitor the infinite set of precontractual, unobservable statements by their employees.90 In addition, a merger clause reduces intrafirm learning costs. It allows the companies’ implementers of commercial contracts to avoid incurring the substantial cost of learning the infinite, elastic set of contextual evidence that already exists for each contract before the implementation time.91 This set of contextual evidence may be established by contracting agents who negotiated the contract and have already retired or moved on from the commercial company, thereby significantly increasing the learning costs of contract implementers.92 True, the contextualist approach may reduce the parties’ transaction costs93 by allowing the parties to write shorter contracts and leave it to the adjudicator to fill gaps with contextual evidence. However, this cost reduction is normally insignificant for commercial contract parties compared with the benefits of merger clauses.94 Sophisticated parties to frameworks of analysis whose conventions will be unfamiliar to them.”); Posner, The Parol Evidence Rule, supra note 77, at 553. 86. Posner, The Parol Evidence Rule, supra note 77, at 556 (“[B]ecause of the high value of the transaction, errors in enforcement are costly.”). 87. Bernstein, supra note 52, at 1817. 88. Id. 89. Id. 90. MITCHELL, supra note 17, at 113 (Parties to commercial contracts may prefer a textualist approach since it can “ensure that the things that their employees and representatives have said and done during negotiations do not bind the company”); Bernstein, supra note 52, at 1817. 91. BURTON, supra note 2, at 33; MITCHELL, supra note 17, at 113; Bernstein, supra note 52, at 1817. 92. Alan Berg, Thrashing Through the Undergrowth, 122 L.Q. REV. 354, 359 (2006) (“[O]ne of the main purposes in instructing lawyers to draft the contract is to ensure that its terms will be clear to those who have to deal with the contract in the future, and to the lawyers advising them, after the management who negotiated the contract have retired or moved on.”); see also BURTON, supra note 2, at 33. 93. MITCHELL, supra note 17, at 108–09 (“One of the arguments in favour of contextualism over literalism is that it lowers transaction costs . . . .”). 94. Id. at 109 (“Parties can write a simpler document, leaving it to the courts to fill gaps through the process of contextual interpretation.”). 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 484 Alabama Law Review [Vol. 69:2:469 commercial contracts, as opposed to unsophisticated parties, can draw upon their experience in order to easily move relevant contextual evidence into the written contract in anticipation of the likelihood of judicial error.95 In addition, parties to commercial contracts are often repeat players and therefore may be encouraged to bear the increased costs of moving contextual evidence into the written contract.96 By incurring these costs only once, commercial companies are able to use their written contracts repeatedly.97 Because parties to commercial contracts are able to embed the contractual context in a written contract, they “are more likely to resent than to welcome a court’s efforts to supplement or circumvent their” contract text by contextual evidence.98 Hence, most sophisticated parties to commercial contracts are expected to include a textualist merger clause in their contract.99 Thus, this paper proposes: H1: A merger clause is more likely than not to appear in commercial contracts between sophisticated parties In addition, this paper hypothesizes that contracts that include “textualist clauses” other than a merger clause are more likely to include a merger clause than contracts without textualist clauses. Textualist clauses, as defined in this paper, are contractual clauses that aim to prevent courts from considering contextual evidence that was not embedded formally in a written text, such as post- and pre-contractual oral statements or notices. Textualist clauses, as suggested in this paper, can be divided into two major categories: direct and indirect. Direct textualist clauses explicitly require courts to consider only written text, thereby ignoring nontextual contextual evidence. Relatively common examples of direct textualist clauses are: (1) a no-oral-modification clause, which states that the contract may be modified or amended only in writing;100 and (2) a notices clause, which states that all notices under the agreement shall be in writing.101 Indirect textualist clauses aim to limit courts from considering the infinite set of contextual evidence during the litigation process. These clauses embed relevant context, aiming to specify precisely the contextual “evidentiary base that will be made available to a court” during the litigation.102 Relatively common examples of indirect textualist clauses are: 95. Posner, The Parol Evidence Rule, supra note 77, at 553–54. 96. MITCHELL, supra note 17, at 110. 97. Id. 98. Gilson et al., supra note 3, at 26. 99. Id. 100. FARNSWORTH, supra note 74, at § 7.6a; Farnsworth, supra note 73, at 274. 101. See, e.g., Philip Morris USA, Inc. v. Appalachian Fuels, LLC, No. 3:08–CV–527, 2009 WL 1011650, at *6 (E.D. Va. Apr. 15, 2009). 102. Scott, supra note 3, at 23; see also Schwartz & Scott, supra note 17, at 961. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 485 (1) a “whereas clause,” also known as “recitals clause,” which includes “[a] preliminary statement in a contract . . . explaining . . . the [contextual] background of the transaction[s],”103 the reasons upon which the contract was formed,104 or the existence of particular contextual facts that surround the contract;105 and (2) a “definitions clause,” which normally “ascribe[s contextual] meanings to words and terms that may vary from their plain meaning.”106 Given their characteristics, textualist clauses, either direct or indirect, reflect the preferences of parties to a transaction for a textualist approach of interpretation, namely that the terms they write in text are enforced as written in the text.107 This paper assumes that if the parties indicate their preference for textualism by utilizing textualist clauses, they are more likely to utilize a merger clause in their contract. This is because a merger clause, given its legal textualist implications,108 supports the parties’ existing textualist preferences, as reflected in the textualist clauses. Thus, this paper proposes: H2: Contracts that include a no-oral-modification clause are more likely to include a merger clause than contracts without a no-oral- modification clause H3: Contracts that include a notices clause are more likely to include a merger clause than contracts without a notices clause H4: Contracts that include a whereas clause are more likely to include a merger clause than contracts without a whereas clause H5: Contracts that include a definitions clause are more likely to include a merger clause than contracts without a definitions clause C. Data The sample of this empirical study is based on commercial contracts contained as exhibits to Form 8-K filings with the SEC.109 Form 8-K 103. Recital, BLACK’S LAW DICTIONARY (10th ed. 2014); see also Edwards v. Doe, 331 F. App’x 563, 572 n.15 (10th Cir. 2009). 104. Blackstone Consulting, Inc. v. United States, 65 Fed. Cl. 463, 470 (2005); see Recital, BLACK’S LAW DICTIONARY (10th ed. 2014). 105. Edwards, 331 F. App’x at 572 n.15; see also Recital, BLACK’S LAW DICTIONARY (10th ed. 2014). 106. Gilson et al., supra note 3, at 59; see also Gilson et al., supra note 15, at 184; Scott, supra note 3, at 23–24. 107. Cf. Bernstein, supra note 12, at 13–14. 108. See supra notes 78–80 and accompanying text. 109. For the same methodological approach of analyzing contracts contained as exhibits to Form 8-K filings with the SEC, see Theodore Eisenberg, Geoffrey P. Miller & Emily Sherwin, Arbitration’s Summer Soldiers: An Empirical Study of Arbitration Clauses in Consumer and Nonconsumer Contracts, 41 U. MICH. J.L. REFORM 871, 880 (2008); Theodore Eisenberg & Geoffrey P. Miller, 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 488 Alabama Law Review [Vol. 69:2:469 The industries of the companies, which filed to the SEC the contracts of this study, are also very heterogeneous, including, for example: banking, education, electricity, agriculture, pharmaceutical, management and consulting, natural gas, hotels and motels, patents, personal services, telephone communication, television broadcasting services, motor vehicle parts and accessories, computer programming, retail, wholesale, tobacco, business services, and industrial organic chemicals.117 D. Methodology In order to locate contracts with a merger clause, I took the following central steps: First, I conducted an in-depth review of the full text of 100 random commercial contracts in the sample. The purpose of this review was to identify the terms commonly associated with a merger clause. Second, based on my in-depth review, I conducted a computerized search, via Westlaw’s terms-and-connectors search engine, for contracts that include the terms commonly associated with a merger clause.118 This search included the following terms: “entire agreement!”; “entire contract!”; “entire understanding!”; or (supersede! /s prior /s agreement!). The “!” symbol was used to search for words with multiple endings, and the “/s” symbol was used to search terms in the same sentence.119 Contracts with a merger clause were coded “1.” Finally, in order to verify that the search results for terms commonly associated with a merger clause were not overinclusive, I performed a human-coded audit of 100 random contracts which were coded “1.” The audit was successful.120 In order to locate contracts with textualist clauses, I took the following two steps for each textualist clause: First, I conducted an in-depth review of the full text of 100 random commercial contracts in the sample in order to identify the terms commonly associated with the textualist clause. Second, based on this review, I conducted a computerized search via Westlaw’s terms-and-connectors search engine for contracts that include the terms commonly associated with the textualist clause. For example, to determine whether a contract included a no-oral-modification clause, which states that the contract may be modified or amended only in writing, I searched via Westlaw’s terms-and-connectors search engine for terms such as “amend! / 117. The companies industries were located via the EDGAR company search engine. See EDGAR: Company Filings, U.S. SEC. & EXCH. COMM’N, https://www.sec.gov/edgar/searchedgar/ companysearch.html (last visited Sept. 6, 2017). 118. For a similar methodological approach applied on a “specific performance” clause, see Eisenberg & Miller, Damages Versus Specific Performance, supra note 109, at 44. 119. Terms and Connectors Searches, supra note 114. 120. Out of 100 results, no result was overinclusive. 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 2017] The Interpretation of Commercial Contracts 489 s writ!” or “modif! /s writ!.”121 Contracts with the textualist clause were coded “1.” Finally, in order to verify that my search results for terms commonly associated with textualist clauses were not overinclusive, I performed a human-coded audit of 100 random contracts which were coded “1.” The audit was successful.122 E. Results Out of 1,521 contracts, 1,145 (75.28%) included a merger clause and 376 did not include a merger clause. A chi-square test was performed, and it was statistically found, unsurprisingly, that a merger clause appears significantly more than not (χ2 [1, N=1,521]) = 388.80, p < .001), supporting H1. Interestingly, the results also show that contracts that have a choice-of- law clause of a state that has contextual rules of contract interpretation, such as California, had a significantly higher percentage of merger clauses.123 Specifically, out of 317 such contracts, 295 (93.06%) had a merger clause.124 These results may indicate that parties who choose to be governed by the laws of contextualist states are truly concerned by the liberal interpretation rules of these states and try to limit them by an inclusion of a textualist merger clause. The results also support H2–H5. A discriminant analysis was conducted to predict the dummy variable, “merger clause.”125 Predictor variables were four dummy variables: “no-oral-modification,” “notices,” “whereas,” and “definitions” clause. Table 2 shows the frequency and percentage of contracts with a merger clause for each textualist clause. 121. For all other textualist clauses, my search included the following terms: “notices /s in writ!”; “any notice /s writ!”; “whereas”; “recitals”; and definitions. I also searched, via Westlaw’s “Clause Title” search function, for contracts with the following clause titles: “amendment!”; “modification!”; “notices”; “recitals”; and definitions. 122. Out of 100 results, only one (1%) was overinclusive. 123. Other states that have contextualist rules of contract interpretation are Alabama, Alaska, Arizona, Michigan, New Jersey, New Mexico, Texas, Vermont, and Washington. See Palia & Scott, supra note 16, at 572 n.14. 124. Contracts with choice-of-law clauses of contextual states were located via Westlaw’s “Governing Law” function. This function allows searching for contracts by their governing-law clause. I performed a human-coded audit of 100 randomly selected contracts to check the accuracy of Westlaw’s Governing Law function. The audit was successful and all the function’s search results were accurate. 125. For more details on discriminant analysis see, for example, ROBERT B. BURNS & RICHARD A. BURNS, BUSINESS RESEARCH METHODS AND STATISTICS USING SPSS 589–608 (2008); ANDY FIELD, DISCOVERING STATISTICS USING IBM SPSS STATISTICS 654–60 (4th ed. 2013). 4 BENOLIEL 469-493 (DO NOT DELETE) 11/14/2017 8:31 PM 490 Alabama Law Review [Vol. 69:2:469 Table 2. Cross tabulation of a merger clause and other textualist clauses Type of Clause Total Number With Merger Clause Without Merger Clause Frequency Percentage Frequency Percentage No-Oral- Modification 1,240 995 80.24% 245 19.76% Without No-Oral- Modification 281 150 53.38% 131 46.62% Notices 1,133 918 81.02% 215 18.98% Without Notices 388 227 58.51% 161 41.49% Whereas 1,118 916 81.93% 202 18.07% Without Whereas 403 229 56.82% 174 43.18% Definitions 742 588 79.25% 154 20.75% Without Definitions 779 557 71.50% 222 28.50% The discriminate function was significant (Λ = .87, χ2 [4, N=1,521] = 216.38, p < .001; canonical R2 = .13), revealing a significant association between a merger clause and all predictors simultaneously: no-oral- modification, notices, whereas, and definitions clauses. Importantly, significant mean differences were observed for each of the predictors on the dependent-variable merger clause: no-oral-modification (Λ = .94, F[1,1519] = 94.20, p < .001), notices (Λ = .95, F[1,1519] = 82.95, p < .001), whereas (Λ = .93, F[1,1519] = 107.30, p < .001), and definitions (Λ = .99, F[1,1519] = 12.33, p < .001).126 126. In addition, a basic chi-square test was performed. It also showed a significant relationship between each of the predictors on the dependent variable merger clause: no-oral-modification ((χ2 [1, N=1521 ]= 88.82, p < .001, Kappa = .24), notices (χ2 [1, N=1521 ]= 78.76, p < .001, Kappa = .23),