Schwartz-Scott - Contract Theory and The Limits of Contract Law

The Yale Law Journal Company, Inc.
Contract Theory and the Limits of Contract Law Author(s): Alan Schwartz and Robert E. Scott Source: The Yale Law Journal, Vol. 113, No. 3 (Dec., 2003), pp. 541-619 Published by: The Yale Law Journal Company, Inc. Stable URL: http://www.jstor.org/stable/3657531 . Accessed: 16/12/2013 14:14
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
The Yale Law Journal Company, Inc. is collaborating with JSTOR to digitize, preserve and extend access to The Yale Law Journal.
http://www.jstor.org
This content downloaded from 189.125.126.40 on Mon, 16 Dec 2013 14:14:05 PM All use subject to JSTOR Terms and Conditions
Article
ContractTheory and the Limits of ContractLaw

Alan Schwartzt and Robert E. Scottt
CONTENTS I. II. INTRODUCTION .................................... 543
AN EFFICIENCY JUSTIFYING THEORY OFCONTRACT...................... 550
A. WhatFirms Maximize.................................................... ........... 550 555 B. Whythe State ShouldHelp Firms .....................................
III. THEENFORCEMENT FUNCTION ..................................... 556
A. EnforcementOftenIs Unnecessary........................... ............. 557 B. EncouragingRelation-SpecificInvestment................................ 559 C. ContractingTo Avoid Disruption: The Case of VolatileMarkets................................ 562 D. Enforcementand Duress............................. 565
IV. THE INTERPRETATION FUNCTION ..................................... 568
t SterlingProfessorof Law,Yale Law School;Professor,Yale Schoolof Management.

of Law;JustinW. D'AtriVisitingProfessor Law School. of Law,Businessand Society,Columbia This Article benefitedfrom commentsreceived at workshopsat the Law Faculty,Cambridge, England,at the 2003 AmericanLaw and EconomicsAssociationmeetings, and at the Harvard, Texas, Toronto,Virginia,and Yale Law Schools. We also are gratefulto Bruce Pennsylvania, John ScottBaker,Richard Ackerman, Brooks,JulesColeman,VictorGoldberg,Sam Issacharoff, Paul Mahoney,Tom Nachbar,Eric Posner,AndrewPostlewaite,Henry Jeffries,JasonJohnston, WilliamStuntz,andGeorgeTriantis for helpfulcomments. Smith,PaulStephan, Professorof Law, Universityof VirginiaSchool tt David andMaryHarrison Distinguished
541
542
The Yale Law Journal
[Vol. 113: 541
A. TheRelevantInterpretiveQuestion................................... 568 B. TwoInterpretiveIssues: Problems of Meaning and of Language ........................................ 570 C. TheParties' PreferencesRegardingInterpretiveStyles ............. 573 1. The Continuous-Payoff Case........................................... 574 2. TheDiscontinuous-PayoffCase ......................................... 578 3. Summary ..................................... 583 D. Private Languages,LinguisticDefaults, and the Parol Evidence R 584 ule.......................................................................... 1. ThePreferredLinguisticDefault .......................................... 584 2. TheParol EvidenceRule................................................... 590 3. Course-of-Performance Evidence ...................................... 592
V. THELEGAL DEFAULT PROJECT ........................ ............. 594
A. The Casefor Defaults ............................ . .................. ... 595 B. The Cost Concernand Default Rules............................... 598 C. TheMoral Hazard Concernand Default Standards .................... 601 D. TheAsymmetric InformationConcern...................................... 605 E. Summary..... o ............................... 608
VI. MANDATORY RULES ........ ................... ......... 609
A. Parties CannotBan Modifications....................... 611 B. Parties MustAccept SubstantialPerformance............. ......... 614 C. Parties CannotAgree to Penalties 616 ...........................................
VII. CONCLUSION .......................................... 618
2003]
Limitsof Contract Law

I. INTRODUCTION
543
Contract law hasneither a completedescriptive what theory,explaining the law is, nora complete normative what the law should theory,explaining be. Thesegaps are unsurprising definition of contract given the traditional as embracing all promisesthat the law will enforce.Even a theory of contractlaw that focuses only on the enforcement of bargains must still considerthe entirecontinuum fromstandard formcontracts betweenfirms and consumers to commercial contracts businesses. No among descriptive thatcomprehends such a broad theoryhas yet explaineda law of contract domain.Normative theoriesthat are grounded in a single norm-such as or have foundered over the heterogeneity of autonomy efficiency-also contractual contextsto which the theory is to apply.' Pluralisttheories to respond to the difficultythatunitary theories normative attempt pose by courts to and of urging pursue efficiency,fairness, goodfaith, theprotection individual Suchtheoriesneed,but so far lack, a meta-principle autonomy. that tells which of these goals should be decisive when they conflict.2
1. For a broaddiscussion of this problem,see MICHAEL J. TREBILCOCK, THELIMITS OF FREEDOM OFCONTRACT 241-68 (1993). Autonomytheoriesrequireelastic notionsof consentin orderto regulate the full scope of contracting behavior with one norm.See, e.g., RandyE. Barnett, A ConsentTheoryof Contract,86 COLUM. L. REV.269 (1986); RandyE. Barnett,TheSoundof Silence: Default Rules and ContractualConsent,78 VA. L. REV.821 (1992); Peter Benson, Abstract Right and the Possibility of a Nondistributive Conceptionof Contract:Hegel and ContractTheory,10 CARDOZO L. REV.1077 (1989); PeterBenson, Contract,in Contemporary 33 OSGOODE HALL 1996);PeterBenson,TheIdea of a Public Basis of Justification for Contract, L.J. 273 (1995). Efficiencytheoriestend to have a more limitedscope. Positive articlesanalyze broaddoctrinal in an attempt to find fundamental and patterns consistencybetweenthese patterns the efficiencynorm,but the authors do not purport to providea fully descriptive theoryof contract law. See, e.g., Ian Ayres & RobertGertner, Contracts: An Economic Filling Gaps in Incomplete L.J. 87 (1989); CharlesJ. Goetz & RobertE. Scott,Enforcing Theoryof DefaultRules, 99 YALE Promises: An Examinationof the Basis of Contract,89 YALEL.J. 1261 (1980). Normative economictheories,on the otherhand,typicallyevaluatediscretedoctrines by the efficiencynorm. See, e.g., CharlesJ. Goetz & Robert E. Scott, LiquidatedDamages, Penalties and the Just Modeland a Theory Compensation Principle:SomeNotes on an Enforcement Breach, of Efficient 77 COLUM. L. REV.554 (1977) [hereinafter Goetz & Scott,Liquidated Jolls, Damages];Christine Contracts as BilateralCommitments: A New Perspectiveon Contract 26 J. LEGAL Modification, STUD. 203 (1997); Alan Schwartz,TheCasefor SpecificPerformance, 89 YALE L.J. 271 (1979); RobertE. Scott, TheCasefor MarketDamages:Revisitingthe LostProfitsPuzzle,57 U. CHI.L. REV.1155 (1990). 2. The problemsthat pluralisttheories without meta-normspose are nicely illustratedin Melvin Eisenberg's effort, which purportsto solve the broad-scope-of-contract problem by sets of norms.See MelvinAron Eisenberg,TheBargainPrincipleand Its proposing overlapping L. REV.741 (1982); Melvin A. Eisenberg,The Theoryof Contracts,in THE Limits,95 HARV. THEORY OFCONTRACT LAW: NEW ESSAYS 206 (PeterBenson ed., 2001) [hereinafter Eisenberg, Theoryof Contracts].For example, Eisenberg'sschema restrictsthe domain of freedom of contract by normsof reciprocity, trust,and fairness.He recognizesthatthis multivalueapproach can generateconflictingsocial propositions. When conflicts actuallyoccur, "the lawmaker must makea legal rulethatgives a properweight androle to each of the conflictingvalues or goals in the context at hand." Eisenberg,Theoryof Contracts,supra, at 244. Further,"when social conflictthe Legislator propositions must exercisegood judgmentconcerning the weight androle
A COMPANION TO PHILOSOPHY OF LAW AND LEGALTHEORY 24, 33-43 (Dennis Patterson ed.,
544
[Vol. 113: 541
modest approach-to set out We attemptto make progresshere with a more. and defend a normativetheory to guide decisionmakersin the regulationof business contracts.3 The theory's affirmative claim, in brief, is that contract law should facilitate the efforts of contractingparties to maximize the joint gains (the "contractual surplus")from transactions.The theory's negative claim is that contract law should do nothing else. Both claims follow from the premise that the state should choose the rules that regulate commercial transactions accordingto the criterionof welfare maximization. A simple categorizationof the universe of bargainingtransactionswill clarify the domain of our theory. A transactioninvolves a seller (whetherof goods or services) and a buyer. Parties to transactionscan be partitioned into individuals and firms. This yields four transactionalcategories: (1) A firm sells to anotherfirm, (2) an individualsells to anotherindividual,(3) a firm sells to an individual,and (4) an individual sells to a firm. Category 2 contracts, between individuals, are primarily regulated by family law (antenuptial agreements and divorce settlements) and real property law (home sales and some leases). Few litigated contractsbetween individuals are regulatedby the rules of contractlaw. Category 3 contracts,between a firm as seller and an individual as buyer, are primarily regulated by consumerprotectionlaw, real propertylaw (most leases), and the securities laws. Category 4 contracts,between an individual as seller and a firm as buyer, commonly involve the sale of a person's labor, and are regulatedby laws governing the employment relation. That leaves Category 1 contracts (those between firms) as the main subject of what is commonly called contract law-namely, the rules in Article 2 of the Uniform Commercial Code (UCC) and the provisions of the Restatement(Second) of Contracts. Such provisions are primarily invoked to resolve disputes arising under Category 1 contracts.Our theory applies only to these contracts, and thus has importantimplicationsfor the content of the UCC and the common law of contracts.
to be given to eachproposition Id. Eisenberg in the issue at hand." recognizesthathis theorylacks a metricthatwouldtell the lawmaker "weightandrole"to each social just how to give the proper are likely to be involved or valuewhenconflictsoccur.Id. Sincecourtsor legislatures proposition or values arguablyfavormore thanone type of litigantor when the relevantsocial propositions interestgroup,pluralisttheoriessuch as Eisenberg'stend to be least helpfulwhen they are most needed. 3. In a thoughtful critique of autonomy and efficiency theories of contract, Michael butthatwithouta in theirown terms," Trebilcock concludesthatboththeorytypes are"legitimate thatweighs or ranksthese variousvalues,"bothshouldbe pursuedin varioussocial "meta-theory to perform contextsaccording of different to the relativecompetence effectively. legal institutions TREBILCOCK, supra note 1, at 248. This Articletakes up Trebilcock'sinvitationand proposesa normative the subsidiary relationships categoryof contractual theorythat fits businesscontracts, thatthe law most affects.
2003]
545
Category 1 contracts, however, can be partitioned into two Some partiesobviouslyare sophisticated economicactors subcategories. ElectricCompany). in commercial Other (e.g., the General partiesfunction contextsbut have manyof the characteristics of ordinary persons(e.g., a and owned run a to analyze effort gift shop by retiredteacher).Any contracts between"firms" thusconfronts a boundary issue-how to define of the analysis. a firmforpurposes We drawthisboundary hereby defining a Category1 firmas (1) an entitythatis organized form in the corporate or (3) a and that has five or more employees,(2) a limitedpartnership, suchas a law or accounting firm.Theseeconomic professional partnership entitiescanbe expected to understand and how to makebusinesscontracts, the theorywe developappliesonly to contracts betweentwo such firms. We do not address the extentto which our conclusions hold when one or both of the partiesto a commercial contract fall on the otherside of our boundary. Firmsthatmaximizeprofitsface the canonical"contracting problem" of ensuring both efficientex post tradeandefficientex anteinvestment in the subjectmatter of the contract.4 Partiestradeefficientlywhen,andonly to thebuyerexceedsthe cost when,thevalueof the exchanged performance of performance to the seller. Partiesinvest efficientlywhen their actions maximizea deal's expected surplus.Many observerswould agree that law shouldattempt contract to facilitate The efficienttradeandinvestment. of lies in our its the of novelty theory systematic implications development of this goal and in its claim that contract law shouldrestrictitself to the of efficiencyalone(forCategory 1 contracts). pursuit Four objectionsmay be made to the claim that contractlaw should restrict itselfto encouraging efficienttradeandinvestment. First,one could that firms sometimes do not maximize and, argue owing to the profits errors made the run who them,areincapable systematic cognitive by people of doing so shouldthey try. A law thatpresupposes profitmaximization would then be misguided. Second, one might claim that firms that maximizeprofitssometimesdo bad things-pollute the environment, for to deter.Third,one could contend example-that the law shouldattempt that the state should promote fairness in contractingin addition to thatthe state shouldpursue efficiency.And, finally, one might maintain distributional even if sometimes conflictwithefficiency. goals, theymay Theseobjections wouldbe troublesome that for an efficiencyapproach coveredall contract types. We will argue,however,that they have little forcewhen Category1 contracts Firmsandmarkets alone are considered. arestructured so as to minimize the likelihood of systematic cognitiveerror
4. Legal scholarscommonlyreferto investment in the contract's subjectmatteras "reliance." We use relianceandthe economist'sterm"investment" interchangeably.
546
TheYale LawJournal
[Vol. 113:541
withinthe firm.Cognitive decisionmakers error, then,is more by important and 4 to afflict contracts than 2, 3, Category Category1 contracts. likely the bad thingsthat firmsdo commonlyentailimposingcosts on Further, thirdparties,such as creatingenvironmental harmsor erectingbarriers to creation These behaviors-the of externalities-are entry. negative and antitrust laws. An analysisof contract regulated by the environmental can assumethe absenceof externalities. law as such therefore Finally,it fairness usuallyis futileto pursueeitherdistributional goals or contractual a largemeasureof contractual when firmsare permitted freedom.This is becausefirmswill contract or fairlegal rulesthat awayfromredistributive do not maximize In sum,efficiencyis the only institutionally joint surplus. feasible and normatively attractive goal for a contractlaw that regulates dealsbetweenfirms.5 An efficiencytheoryrestricted to contracts betweenfirms(as firmsare defined above) has four majorimplications for contractlaw. The first fact: Many contractswould be implicationfollows from an important even if therewereno legalsanction will be Contracts forbreach. performed when contracts and of a series "self-enforcing" parties contemplate making the gainsfrombreach arelowerthanthe expected fromfuture profitstream contracts thatbreach wouldcauseto vanish.Moreover, neither partywould breachif the gainsfrombreach the wereless thanthe reputational sanction market would exact.Whencontracts fall outsidethe self-enforcing range, enforcement is in to ensure however,legal necessary performance two in cases: volatilemarkets, whena party'sfailureto perform could principal its contract threaten would andwhencontractual survival; partner's surplus be maximized if one or both partiesmade relation-specific investments.6 "Enforcement" includesmore than simply requiring partiesto perform, however.It also entailsthe prevention of fraudandduress,as well as rules to encourage or facilitateperformance, such as damagesrules. Perhapsa thirdof the sections in UCC Article2 are enforcement rules underthe definitionhere. The initialimplication of our theoryis that enforcement, when needed, is by far the most important thing the state does. Put
5. As anotherexample of the criticism that we sidestep here, Professor Eisenberghas criticizedtheoriesholdingthat contractlaw shouldmaximizewelfare alone on the groundthat these theoriesare "impoverished ... becausethey excludeotherimportant policy values, such as the value of keeping intimateand other affective relationships free from the intrusionof state power."Eisenberg,Theoryof Contracts, supra note 2, at 238. This objectionmay have force as 2 contracts, betweenpersons,but seems irrelevant to the Category1 contracts appliedto Category we analyze.Contracts betweenGeneralElectricandGeneral or Motorsdo not involve"intimate" "affective" relationships. 6. A relation-specific investmentis not fully "redeployable." As an example,assumethat a seller purchasesstandard steel tubes to make a machinefor the buyer.The seller's investment wouldbe "general" if breachoccurred beforethe sellerbeganworkon the tubesbecausethe tubes could be resold on the market.The investmentwould become "relation-specific" if breach occurred afterthe tubeshadbeen fabricated into shapesthatonly the buyercoulduse, for thenthe transmuted tubescouldonly be resoldas scrap,probably for less thantheircost.
2003]
547
morestarkly,a modemcommercial economycan functionwell with little more than honest courtsand a set of enforcement rules. The rest is of second-order importance. A court cannot enforce contracts,however, without a theory of that"maps" fromthe semantic contentof the parties'writing interpretation Our second implication thus holds, in to the writing'slegal implications. contrast to the UCC and much modem scholarship,that textualist should be the default theory for Category 1 contracts. interpretation Business firms, that is, commonlyprefercourtsto adhereas closely as parol possible to the ordinarymeaningsof words, to apply a "hard" andto honor"merger evidencerule,7 clauses"(whichstatethatthe parties A textualist theirwritingto be interpreted as if it were complete). intended however,will not suit all partiesall of the time. theoryof interpretation, courtsshoulduse narrowevidentiary bases when interpreting Therefore, between but also should firms, agreements complywith partyrequeststo the base thatis applicable to them.This implication is at variance broaden with current is an issue for courtsto law, which holds that interpretation cannot decideandshouldbe conducted to rulesthatparties vary. according Contract law has more rules regulating various aspects of the thanare neededsolely to perform its enforcement contracting relationship functions. andinterpretation theserulesaredefaults,controlling Typically, out of them.Creating good defaultsis only when partiesdo not contract to the This believed be of a law of contracts. function widely principal belief is misguided. Ourtheory'sthirdimplication holds thatthe effective domain of business-contract law is much smaller than is commonly The statecan createdefaultsthatbusinessfirmswouldwantonly thought. undervery stringent To be useful,a defaultrulesmustapplyin conditions. few of states the world, be relativelysimple in form, be very possible efficientin a highly heterogeneous and not rely on set of circumstances, information that courtscannotconveniently recover.A defaultstandard shouldbe writtenwhen partiesdo not need, or it is too costly to provide them with, concrete guidance regardingthe performanceobligation. Becausestandards muchlatitude permit parties (e.g., the sellermustdeliver in a "reasonable" a will discretion standard confer time), good only when a party's likely actions under it will maximize joint-rather than draftersand courts, we argue, often adopt individual-gains. Statutory
default rules and standardsthat fail to satisfy these stringent conditions. This is itself inefficient because partiesrespondto bad rules or standards by
7. A "hard"parol evidence rule treats writings that appearto be complete contractsas See infratext accompanying notes 96-98. completecontracts. 8. The decisionmaker specifiesthe contentof a rule in advance.For example,driverscannot exceed a speed limit of fifty-five miles per hour. In contrast,the decisionmaker specifies the contentof a standard ex post. Thus,parties mustdrive"reasonably" underthe circumstances.
548
TheYale LawJournal
[Vol. 113:541
out of them. The creationof inefficientdefaultsthus raises contracting costs but does not otherwiseaffect their business parties' contracting behavior. The lack of good defaultsthus makesmuchof today'scontract life. law irrelevant to commercial In additionto its many defaults,contractlaw containsa numberof rules thatare appliedto contracts betweenfirmsas well as to mandatory contractsbetween firms and persons. The fourth implicationof our businesscontracts efficiencytheoryis that many of the rules regulating We discuss a numberof mandatory should not be mandatory. rules, rules modification and rules, rules, relatingto includinginterpretation clauses.The onlyjustification for theserulesis a form damages liquidated Theserulesdo not override of paternalism: contractual termsbecausethose or are unconscionable, termscreateexternalities but ratherthey seek to override terms that appear to conflict with parties' true substantive intentions.We argue, however, that business firms would have good reasons to adoptthetermsthattodayareprohibited. A commitment to party thatthosereasons be respected. sovereignty requires The need for a general efficiency theory of business contractsis salienttoday. The creationof a contractlaw has become an particularly thathave made a new (or renewed) important priorityin many countries commitment law is to markets, forthereis a consensus thata good contract a necessaryconditionfor a moderncommercial economy.It is less well understood how such a law is OurArticlethus to function. just supposed addresses concerns thathaveglobalimplications. A furtherreason motivatingour decision to develop a theory of businesscontracts is thatthe building blocksfor sucha theoryareonlynow available. Contract becoming theoryhasbecomeone of themostsignificant fields in modernmicroeconomics economics. and industrial organization Three recent Nobel Prizes were awardedto George Akerloff, Michael Spence, and Joseph Stiglitz, largely for work in contracttheory, even muchof the work thoughthe field is less thanthirtyyearsold.9Moreover, in the field takes a mathematical not been easily thus has and form, accessible to nonspecialists.We draw heavily on contracttheory to construct ournormative theoryof contracts.'o
9. The workof these scholarsis conciselysummarized in Karl-Gustaf Lbfgrenet al., Markets withAsymmetric The Contributions Information: of GeorgeAkerlof MichaelSpenceand Joseph J. ECON. 195 (2002). Stiglitz,104 SCANDINAVIAN 10. Law-and-economics scholarssuch as AaronEdlin, Ian Ayres, and JasonJohnstonhave used contracttheoryin an illuminating fashionwhen discussingparticular legal rules. See, e.g., Ayres & Gertner, supra note 1; Aaron S. Edlin, Cadillac Contractsand Up-FrontPayments: UnderExpectation 98 (1996); JasonScott & ORG. EfficientInvestment Damages, 12 J.L. ECON. Johnston,StrategicBargainingand the EconomicTheoryof Contract DefaultRules, 100 YALE L.J. 615 (1990). The genre of model that we and these scholars use has performedwell in THEORY: A SURVEY OF & B. SALANIE, CONTRACT TESTING empiricaltests. See P.A. CHIAPPORI
2003]
549
Finally, as we suggested earlier,the currentstate of contractlaw suffers from the absenceof a successfultheoryof contract. scholarship Thirty years ago, GrantGilmoredescribedwhat he called the classic Willistonian model.This model was grounded in formalist notionsof the of written between centrality agreements contracting voluntarily exchanged the limitedrole of the law in enforcingand parties,and it emphasized theseagreements." to Gilmore, this classicalmodel interpreting According owed moreto Holmes'simagination thanto a carefulreadingof the case this was so or not, Gilmorebelievedthatmodemcase law.12But whether the model.The disjunction law repudiated betweenthe dominant scholarly view and the lived doctrine,he thought,produced incoherence."3 Modem scholarscommonlyshareGilmore'srejection of Willistonbut have yet to thesis.14 We, too, lack a theoryof everything. disprovehis incoherence the theorywe develophereis Willistonian in spirit,butappliesin a Rather, limiteddomain-to contracts betweenfirmsthatdo not createexternalities. Thislimitedscopepermits ournormative to thesisto developaccording a particular logic. The market'ssocial functionis to maximizewelfare, and fairnessconstraints. Firms,we show below, subjectto distributional to choose the contracts have incentives andcontracting thatwill strategies maximizethe surplusfromtheirdeals. Further, firmsare betterable than courts or statutorydraftersto choose efficient terms and strategies.It follows that, when externalities are absent,a contractlaw that regulates firmsshouldbe the contract to apply law thatfirmswouldprefergenerally to theirtransactions. Thatis, the state shouldlet the preferences of firms control becausefirmscanbetter the objective thatboththe stateand pursue firmsshare.Thus,the centralorganizing of question this Articleis: What contract law wouldcommercial wantthestateto provide? parties We proceedas follows:PartII defendsthe welfare-maximization norm as appliedto the contracts of sophisticated actors.In PartIII,we describe commercialparties' first-orderpreferenceto have the state enforce contractsin orderto protectrelation-specific investmentsand to guard market movements. PartIV arguesthatfirms againstespeciallydisruptive
RECENT WORK 27 (Institut Nationalde la Statistique et des Etudes SOME lconomiques, Working No. 2002-11,2002), http://www.crest.fr/doctravail/document/2002-1 Paper l.pdf. THE DEATH OFCONTRACT 42-53 (1974). GILMORE, I11.See GRANT 12. See id. at 63 ("[T]hetheoryof contract, as formulated by HolmesandWilliston,seems to have gone into its protracted But see almostfromthe momentof its birth."). periodof breakdown Richard E. Speidel, An Essay on the ReportedDeath and Continued Vitalityof Contract, 27 STAN. L. REV.1161, 1182-83(1975) (reviewingGILMORE, supranote 11). 13. See GILMORE, supra note 11, at 67-68 ("[Consider]the Restatement'sdefinition of consideration(? 75) taken in connectionwith its most celebratedsection-? 90 [promissory cannotlive estoppel].... The one thing that is clear is thatthese two contradictory propositions in the end one mustswallowthe otherup."). comfortably together: 14. See, e.g., Eric A. Posner, EconomicAnalysis of ContractLaw After ThreeDecades: Successor Failure?, 112 YALE L.J. 829 (2003) (lamenting or the absenceof successfulnormative theoriesof contract descriptive law).
550
TheYale LawJournal
[Vol. 113:541
want the state to supply a theory of interpretation, but not the theory
currentlyadvancedby the UCC and the Restatement(Second) of Contracts.
we defenda textualist as the optimal default Rather, theoryof interpretation contracts. for business In Part we the restrictive V, approach develop conditionsunder which the state can create default terms that satisfy PartVI analyzesa set of unjustifiable typicalpartypreferences. mandatory rules-rules that rest on a misplacedview of the parties'interests.We conclude,in Part VII, that today's contractlaw is a series of category mistakes.Rules that are appropriate for contractsinvolving individuals 2 4 are too (Categories through above) frequently appliedto sophisticated law for centurieshas drawna distinctionbetween parties.Commercial mercantile contracts andothers.'5 Modern have not systematically scholars the normative of this ancient however.We distinction, pursued implications to curethis neglectby settingout the theoretical of a foundations attempt law merchant forourtime.
II. JUSTIFYING AN EFFICIENCY OFCONTRACT THEORY
A. WhatFirms Maximize
It has been traditionally assumedthat firms attemptto maximize the has been expectedprofits.16Recently, accuracyof this assumption challenged.There are two reasonswhy an economicactor may fail to maximizewealth:(1) She is maximizingsomethingother than her own becauseshe is concerned with fairness; or (2) she cannot wealth,perhaps maximizewealthin the contextunderstudy,perhaps becauseshe is prone to cognitiveerror. Thesereasonsapplymuchless to firmsthanto persons. A firmis directed Shareholders by its owners,who oftenare shareholders. then can prefertheir firms to maximizeprofits,which the shareholders
15. Vestiges of this distinctionexist in the few UCC sections that regulatedeals "between merchants" fromdealsbetweena merchant anda person.See, e.g., U.C.C. ?? 2-104(3), differently 2-201(2), 2-207(2),2-209(2),2-603, 2-609(2) (2003). 16. Individuals are assumedto be risk-averse The while firmsare assumedto be risk-neutral. utilityfunctionof a risk-neutral partyis linearin money-that is, the partyvalueseach additional dollarof wealthit may receiveas muchas it valuedall previousdollars.Becausemonetary gains arecoextensivewith utilitygains for risk-neutral firmswill maximizeprofits, parties,risk-neutral a monetary The risk-aversion measure. or profit-maximization for firmsfollows from assumption two facts. First,the amountat stakein any one contract commonlyis small in relationto the size of the firm,so firmsactuallyholdcontract in riskstendto offset one another portfolios.Individual a portfolio,so the portfolioholder-i.e., the firm--wantsonly to maximizethe monetary value of the portfolioas a whole, which is best done by maximizingthe value of each contractin it. who Second, as is consideredimmediatelyin the text above, firms are owned by shareholders themselveshold diversifiedportfolios.The value of a shareholder's portfoliois maximizedwhen each firm in it does as well financiallyas it can do. The firstgroundfor supposingthatfirmsare risk-averse sometimesdoes not hold. We pursuethe implications of this failurein Section III.C below.
2003]
551
consumeor save. Firmsthus will choose to maximizeprofitsunless the who runthemcannotbe controlled who own managers by the shareholders them.In these cases, the managers theirown earnings may be maximizing orperksat the expenseof profitmaximization. owners. No one doubtsthatmanagers sometimes successfully sabotage Fortwo reasons, however,we will assumethatmanagers obey shareholder instructions. shareholders either First, managerssabotage by diverting wealthto themselvesor by failing to take appropriate risks on corporate behalfof the firm.Managers, however,have no incentiveto degradethe thatthey write;afterall, these contracts createthe qualityof the contracts wealth that the managerslater can divert. Second, the legal rules that to deterbadmanager behavior fall in the domainsof the criminal, attempt corporate, and securities laws. Contract law should exploit this by assumingthat the agreementsit regulatesreflect the specialization choices. maximizing parties' Firmsthatattempt to maximize to do as well as profitscanbe expected is because the theircircumstances This to survive permit. promotes pressure Thispressure takestwo forms.First,firmsthatsystematically competence. make bad economic decisions lose out in competitionwith profitfirms.Hence,surviving firmsaregenerally the onesthatcando maximizing makebad whatthey set out to do. Second,employeeswho systematically economic decisions are unlikely to be promoted to positions of cangenerally do whattheyset outto Hence,seniormanagers responsibility. do.17This is not to say thatall firmsall the time pursue profit-maximizing But it to is that owners and the market strategies. say put systematic pressureon firms to behave optimally;hence, it is a plausibleworking assumptionthat firms rationallypursue the objective of maximizing
profits.18
17. In addition,many corporateexecutives have attendedbusiness school and also attend businessschool executiveprograms for workingmanagers.It is a functionof businesseducation to teach people to make optimizing(ratherthan cognitively erroneous)decisions. Studies also show that individualsubjects can be trained to performcomplex game-theoretic reasoning. See Miguel Costa-Gomes et al., Cognition and Behavior in Normal-Form Games: An 1193 (2001); EricJ. Johnsonet al., DetectingFailuresof Experimental Study,69 ECONOMETRICA BackwardInduction:MonitoringInformation Search in SequentialBargaining, 104 J. ECON. THEORY 16 (2002). 18. Psychologists and economists have shown that persons make systematic cognitive mistakesin laboratory when asked to solve specified individualdecisionproblems. experiments These experiments do not test a generaltheoryof how people makedecisions,andthusthey raise an issue of external validity: When will real-worldparties behave as did the experimental subjects?Two scholarsrecentlynoted a consequenceof this "lack of theoreticalfoundations": "[T]he policy influence of BE [behavioraleconomics] is limited by its inability to predict circumstances in which anomalous behaviorwill arise(otherthanin those sortsof circumstances in which it has been observedbefore)or how it will respondto policy changes." JessicaL. Cohen & WilliamT. Dickens,A Foundation REV.335, 335 for BehavioralEconomics,92 AM.ECON. to the law, see (2002). For a recentskepticalview of the relevanceof the psychologicalliterature
552
[Vol. 113: 541
The assumptionthat each partyto a contractwants to maximize its own profit does not itself imply that parties also want to maximize joint gains. Rather,a party may prefer a larger share of a smaller pie. Thus, one might think that parties will behave strategicallyat the expense of joint welfare maximization. On a deeper view, however, one can see that sophisticated parties at the negotiation stage preferto write contractsthat maximize total benefits."9To see why, assume that each party's share of the contractual surplusis set exogenously. This assumptionholds that a party cannot affect the size of its share of the parties' bargainby the (nonfraudulent) actions it takes duringa negotiation.On this assumption,the partieswill want only to maximize the total surplus.20 To put this point in a contractingcontext, let parties contemplatemaking a simple sales contractfor goods that the buyer values at $100 and that would cost the seller $80 to produce.Now assume that each party's share in the contractingsurplus ($100 - $80) is fixed in advance at one-half each. Then the price would be $90, and each party's profit would be $10. Of more importance,assume that the seller could make a $2 investment in the subject matter of the contract that would lower its production cost to $70. The seller would want to make this investment because then its share of the new $28 contractualsurplus($100 - $2 - $70) would be $14, a share that would be realized by a price reduction to $86.
TooSeriously?The Unwarranted Behavioralism Pessimismof theNew GregoryMitchell,Taking Behavioral Analysisof Law,43 WM.& MARY L. REV.1907 (2002). We provisionally view the individual decisionexperiments as not relevantto our projectfor three reasons. First, as we are in a world of speculation,we speculate that individualsin laboratories may performworse than officers of firms because experimental subjectshave not been trainedto make good decisions and are not subjectto the pressuresto maximizethat are describedabove.Second,recentevidencesuggeststhatbehavioral anomaliescan be substantially when individualsare asked to performas actors in firms, see mitigatedor made to disappear JenniferArlen et al., Endowment Effects WithinCorporate AgencyRelationships,31 J. LEGAL STUD.1 (2002), or when the applicableinstitutionspermitcommunication within a group of actorsandrequire betweengroups,see TILMAN SLEMBECK & JEAN-ROBERT TYRAN, competition
Do INSTITUTIONS PROMOTE RATIONALITY? AN EXPERIMENTAL STUDY OF THETHREE-DOOR
ANOMALY of St. Gallen, WorkingPaperNo. 21, 2002), http://www.vwa.unisg.ch/ (Universitat RePEc/usg/dp2002/dp022 Islembeckganz.pdf.The view thatorganizations composedof experts make better decisions than individualsubjects is coming to be acceptedin the psychological as well. See JeffreyJ. Rachlinski, TheUncertain community PsychologicalCasefor Paternalism, 97 Nw. U. L. REV.1165, 1214-16(2003). Third,experiments show thatwhenpersonsarerequired to functionin markets thatare rather thanto perform individual tasks,the personsreachequilibria consistent with individuallyoptimizing behavior. See Vincent P. Crawford,Introductionto GameTheory,104 J. ECON. THEORY Experimental 1, 3 n.8 (2002) ("[R]epeated play of the same K. Gode no matter whatsubjectsare thinking."); game often convergesto equilibrium Dhananjay & ShyamSunder, AllocativeEfficiencyof Marketswith Zero-Intelligence Traders:Marketas a Partial Substitute K. Gode 119 (1993); Dhananjay ECON. for Individual Rationality,101 J. POL. & Shyam Sunder,WhatMakes MarketsAllocationallyEfficient?,112 Q.J. ECON. 603 (1997); VernonL. Smith,RationalChoice:TheContrast BetweenEconomicsand Psychology,99 J. POL. 877 (1991) (reviewarticle). ECON. 19. Partiesneed the law's help to dealwithpostcontractual See infraPart1II. opportunism. 20. Put simply,if a partyis to receivea fixed twentypercentof a joint gain,it would always than$100. preferthejoint gainto be $200 rather
2003]
553
thebuyerhas an incentive to makecost-justified Similarly, value-increasing investments. forus to showthatparties'bargaining It remains sharesactually areset standard In is a function theory,bargaining bargaining exogenously. power of two factors.The first factoris the parties'relativepatience.The more will rejectoffers it dislikesto wait for more favorable patientbargainer while the less will acceptrelativelyunfavorable offers, patientbargainer offersjust to get a deal.21The secondfactoris each party'sdisagreement option).Disagreement powerin pointsaffectbargaining point(or next-best a subtleway. To see how, supposeinitiallythat the partiesare equally rate).Theythenwill dividethe surplus (theyhavethe samediscount patient froma deal equally,if they do the deal at all. Now let d, be the seller's point,andlet dhbe the buyer'sdisagreement payofffromits disagreement point. The expectedsurplusif the partiescontractis s. A deal would be efficientif it wouldgenerate a surplus thatexceededthe sumof theparties' outsideoptions(i.e., if s > dc + db). The parties'disagreement pointswill not determine the split of the surplus s from such a deal if one-halfs is greater thanbothd, anddb.Suppose, however,thatone-halfs is less thancd,the seller'sdisagreement point.Thesellerwouldnot contract with the buyerunless the sellerwould do at least as well in the deal as it could do elsewhere.In this variantof the example,the seller thus would receivea bargaining payoffthatis its disagreement pointd, plusone-halfof the surplus thatremains aftersubtracting the parties'disagreement payoffs. Hence,disagreement pointsonly affect bargaining powerin a deal when one of the partieswoulddo bettertakingits outsideoptionthanaccepting the surplus rateswouldotherwise discount splitthattheparties' imply.22
21. A party's discountrate measureshis patience:The higher a party's discountrate, the more highly the partyvalues currentdollarsthan futuredollars.Partieswith high discountrates thusare impatient bargainers: Theywanttheirshareof the surplusnow. A partyfor whomcurrent dollarsare relativelyless important--apartywith a low discountrate-suffers less from delay offers. Hence,patientpartiesdo well when and, as a result,is morewilling to rejectlow current with impatientparties,who will reduce their demandsin orderto reach agreement bargaining quickly. 22. Thereare two versionsof the Nash bargaining and "dealme game:"splitthe difference" out." In split-the-difference bargaining,if a deal would be profitable,each party receives its in deal-me-out each party disagreement payoffplus one-halfof the remaining surplus; bargaining, gets half the surplus(if they are equallypatient)unless one party'sdisagreement payoff exceeds half of an equal split. To illustrate the difference,assumethatthe seller's disagreement point is $2; the buyer'sdisagreement pointis $1; andthe surplusfroma deal betweenthemwouldbe $10. Theparties'payoffsunderthe two games (with split-the-difference set out first)are (1) Seller: $2 + V2 ($10 - $3)= $5.50 Buyer: $1 + ?2($10 - $3) = $4.50 (2) Seller: /2 ($10) = $5 Buyer: V2($10) = $5 Game theoristspredictthat deal-me-outbargaining is commonbecause a party'sthreatto exit unless it receivesits disagreement payoff plus one-halfthe remaining surplusseldom is credible. On the assumedpayoffshere,the buyerthus would not believe a seller's threatto exit unless the
554
TheYale LawJournal
[Vol. 113:541
A businessparty'spatienceis a functionof its abilityto financeits Firmsthathavecapitalor convenient accessto capitalcanbe more projects. patient than firms that need revenue immediatelyto survive. Parties to the cannotaffect the access of prospective contract partners ordinarily one to a cannot Moreover, party possiblecontract ordinarily capitalmarket. affectthe otherparty'salternative businessopportunities (its disagreement contract will realizethatits shareof the point).Thus,eachpotential partner the partiescouldgenerate been fixed maximum surplus jointlyhas already before any contractis signed. Hence, each partywill contractso as to the size of thepie. maximize This resultis contrary to the commonview. It is widely believedthat exercise weakercontracting bargaining parties powerby requiring partners terms.Thus,section2-302(1)of the UCCauthorizes to takeunfavorable a courtto strike"anyclauseof the contract" if the clauseis unconscionable.23 Termsthat superficially as the one-sided are commonlydescribed appear of product "unequal power is bargaining power."But when bargaining determined as is commonin businesscontexts, formation, priorto contract these views are incorrect. Bargaining power insteadis exercisedin the division of the surplus,which is determined by the price term. Parties whichthe termsso as to maximizethe surplus, jointlychoosethe contract then divide pricemay unequally.24
seller received $2 plus one-halfthe residualsurplus(the total $5.50 payoff) because the buyer would know thatthe seller neverwould reject$5 (its deal-me-out payoff) in favorof bargaining the buyer'sbelief, the seller would agreeto accept $5 rather $2, its outsideoption.Anticipating thanfutilelyattemptto get $5.50. Now let the seller's disagreement payoff increaseto $6. Then the partieswould switch from deal-me-outbargainingto split-the-difference The bargaining: $2.50. The switchin bargaining to occurbecausewhenthe sellerwouldrealize gamesis predicted $6 from its outside option, its threatto exit unless it gets at least $6 in the bargainbecomes credible.We use the deal-me-out this Article(exceptwhen the seller bargaining gamethroughout has a high outsideoption)becauseof the game theorists'logic andbecauseexperimental tests of F. bargainingbehavior reject "split the difference"in favor of "deal me out." See COLIN THEORY 175-82 (2003); Ken Binmoreet al., An OutsideOption CAMERER, GAME BEHAVORIAL 104 Q.J.ECON. 753 (1989). Experiment, 24. This conclusion applies even though parties may be uncertainabout the amountof bargaining powerthey actuallyhave. For example,partiescan use a maximinstrategywhen they knowthe set of possibledisagreement pointsbutdo not knowwhichmemberof the set appliesto their case. Partiesusing this strategywill evaluateuncertain gains by comparingtheir no-deal result to the smallest payoff possible under the set of potentialdisagreement points. If this minimumpotentialpayoff exceeds the surplus in a no-deal condition, parties will contract; otherwise,they will not. As a consequence,some efficientdeals will not be made,but those that are will be Paretoefficient.See WalterBossert& HansPeters,EfficientSolutionsto Bargaining Problemswith UncertainDisagreement 489 (2002). Our & WELFARE Points, 19 SOC.CHOICE conclusionin the text is unaffectedby this form of uncertainty because neitherpartycan affect any of the disagreement pointsif the full set is exogenous.This set of modelsrequires partiesto reachagreement on the basis of the possiblechoices availableto them.Whenpartiesare promptly optimisticabouttheirbargaining powerbut can learnthe truthby inferencefromthe sequenceof offers each of them makes, they will reachefficientbargains,thoughwith delay. See Muhamet Feb. 2004). We do not claim that Yildiz, WaitingTo Persuade, 119 Q.J. ECON. (forthcoming
seller would receive $6 + /2 ($10 - $7) = $7.50 and the buyer would receive $1 +
/2
($10 - $7) =
23. U.C.C. ? 2-302(1) (2003).
2003]
555
B. Why theStateShould HelpFirms We notedat the outsetthattherearefourmainobjections to the singleminded pursuitof welfare maximization for commercialcontractlaw. SectionII.A arguedthatthe cognitiveobjectionis weak, and SectionV.A and arguesagainstthe fairnessobjection.We discusshere the externality distributional objections.The externality objectionis weak because,as a affect only the partiesto matter,most commercialcontracts descriptive them. A single sales contractthat turns out badly is unlikely to put out of workor causeretailers in the firm'slocalityto suffer.It is employees a firm'ssystematic decisionsthatmay affectthirdpartiesin material ways. For example,a firmmay runa factorywith disregard for the environment or the rightsof its workers. inefficient or unfairbehavior of Systematically this kindis subjectto legal regulation underthe headings of environmental law. Relyingon this specialization and employment we assume principle, thatthe transactions law do not createexternalities, regulated by contract unlessthereis a particular reasonto believethattheydo. Commercialcontractrules seldom create systematicdistributional benefits for particular classes of parties.In the first place, commercial commonly occupyboththe rolesof sellerandbuyer(or licensorand parties a pro-seller rule wouldhurtfirmswhen licensee,etc.). As a consequence, rulewouldhurtthesesamefirmswhenthey sell. theybuy, anda pro-buyer In addition,because most commercialcontractlaw rules are defaults, distributional benefitsarehardto createevenfor firmsthatprimarily buy or sell. Suppose, for example, thata contract ruleallocates a riskto the selling side of the market in orderto createa distributional benefitfor the buying side. Supposealso thatcontractual if buyers surpluswouldbe maximized borethe riskat issue (because,say, they arethe cheapest A cost avoiders). contract allocatingthe risk to the buyerwould makeboth the seller and buyerbetteroff (becausethey would split a largersurplus). Consequently, the legal rule's allocationwould be unstable.Because business firms to maximize contractual the default rulesthatconstitute the attempt surplus, bulk of commercial law rarelycould systematically benefiteitherside of themarket. benefits for the Moreover,it is difficult to create distributional shareholders who own most business firms. Shareholders typicallyhold diversified A diversified shareholder oftenwill own some firms portfolios. thatbuy andsell, some firmsthatprimarily buy, andothersthatprimarily
parties always choose efficient contract terms. The existence of asymmetric information sometimes will cause parties to make constrainedefficient contracts;these contractsare not "first-best" but areefficientgiven the information structure facingthe parties.In PartV, we argue that the state seldom can improveon constrained that is efficient contractsbecause information unavailable to the partiesis unlikelyto be availableto the decisionmaker.
556
TheYaleLawJournal
[Vol. 113:541
is sell. An attemptto benefit either side of the marketdistributionally also is Diversification a for shareholder. such net to create unlikely gains ownerwantsthe value of his portfolio A diversified relevant. normatively firmsin his portfolioat the expense to increase, not the valueof particular of other firms in his portfolio.Indeed,investorsdiversifyprecisely to firm thatthe investor risk-the risk that a particular escape firm-specific owns will have an unusually bad outcome. Satisfying this investor acrossfirms. thusrequires surplus legalrulesthatmaximize preference claim normative In sum, cogent reasonsexist to justify our principal welfare maximize firms facilitate the of to should law contract that ability The reasonsset out here also imply, contracts. when makingcommercial thatit is unnecessary or futilefor courts for thisclass of contracting parties, The contractlaw of to distributional drafters pursue or statutory goals. is aboutefficiency. commercial parties
FUNCTION III. THEENFORCEMENT
A perennial law is why the stateshouldenforcea questionin contract contractagainstthe wishes of a partyto it. We excludeanswersto this questionthattake the followingform:The state shouldenforcea party's contractual promisesthe betterto permitpersonsto enlistotherpersonsin withinwhich theirprojects,and thus to increasethe sphereof autonomy the to or the stateshouldenforcepromises reinforce personscan operate;25 moralityof keepingthem. These answersare ruledout here becausethe contracts areartificialpersonswhose businessfirmsthatmakecommercial andwhosemorality the stateneednot respecton moralgrounds, autonomy for a normative The law. relevant is ordinarily question required by positive if the is of contract law commercial just when, ever, goal of welfare theory maximization of the contractsthat business requireslegal enforcement entitiesmake.
25. For excellent analyses of the strengthsand limitationsof the variousautonomy-based theoriesof contractlaw, see Richard Law,DefaultRules,and the Philosophy Craswell,Contract L. REV. Law, 489, 514 (1989);andJodyS. Kraus,Philosophy of Contract of Promising,88 MICH.
OF LAW687 (Jules Coleman in THEOXFORD OFJURISPRUDENCE AND PHILOSOPHY HANDBOOK
& ScottShapiro in the questionof when individual eds., 2001). We also arenot interested persons shouldkeep theirpromises,as we wantto identifythe cases in whichthe state shouldenforcethe thatfirmsmake.For an interesting questionas agreements analysisof the duty-to-keep-promises applied to persons, see Niko Kolodny & R. Jay Wallace, Promises and Practices Revisited, 31 PHIL. & PUB. AFF.119 (2003).
2003]
557
A. EnforcementOftenIs Unnecessary
A contracthas an intertemporal aspect: Partiesagree today to do something tomorrow.26 State enforcement of these agreements is whenthe agreements fall withinthe self-enforcing rangeor can unnecessary be enforcedwith reputational An agreementis said to be sanctions.27 when the threatby eitherpartyno longerto deal with the self-enforcing otheris sufficientin and of itself to induceperformance.28 in Reputation, will induce when a contract boycott turn, performance partner's single wouldnot.29For reputation to work,however,potential futurecontracting to learnwhy the originalparties'deal partiesmust be able conveniently broke down. Reputations, therefore,are difficult to establish in large economiesin which particular contracting partiesoften are anonymous. work best in small tradingcommunities, Rather,reputations especially thosewithethnically whereeverything thathappens members, homogenous soon becomescommonknowledge, andboycottsof bad actorsareeasy to sanctions also canbe effectivein industries enforce.30 thatcan Reputational the associations becomea form of collective establishtradeassociations; the contracting behavior of theirmembers.31 memory regarding This Article neverthelessfocuses on contractsthat fall outside the sanctions. self-enforcing rangeandthatcannotbe enforced by reputational We takethis focusfor two reasons. businessnetworks First,while informal
26. Agreements often arewritteneven when the partiesexpectnot to enforcethemlegally.A as over what the partieshad actuallyagreedto do. Disagreements writingreducesdisagreements to whatthe contract directsraiseinterpretation issues thatarediscussedin PartIV below. 27. For good, largely informal discussions of these issues, see Benjamin Klein, Why 444 34 ECON. INQUIRY Hold-UpsOccur:TheSelf-Enforcing Rangeof Contractual Relationships, 75 CAL. L. REV. in Long-Term Contracts, (1996); and RobertE. Scott, Conflictand Cooperation 2005,2039-50(1987). 28. SupposeS and B write a contract B later in a statethatdoes not legallyenforcecontracts. learns that it could make $100 more by breachingthe contractthan by performingit. If B breaches, however,S will no longerdeal withB. Let B's expectedprofitson these futurecontracts have a presentvalue of $200. ThenB will performthe contract,thoughit could not be sued for breach,becausebreachwould cause it to lose $100, thatis $100 - $200 = -$100. The contractis self-enforcing. 29. For example,supposethatS's laterrefusalto deal would imposeonly a $10 loss on the breachingbuyer, but other sellers also will refuse to deal, raising the buyer's total loss from breachin presentvalue termsto $200 ($10 plus a $190 reputational sanction).Again, B would voluntarily perform. 30. See Janet T. Landa,A Theoryof the EthnicallyHomogenousMiddlemanGroup:An Institutional Alternative to Contract STUD. 349 (1981). An excellentsurveyof Law, 10 J. LEGAL enforcement Lessons mechanisms is AvnerGreif,Informal Contract earlyinformal Enforcement: NEW THE LAW 287 from MedievalTrade,in 2 THE AND PALGRAVE DICTIONARY OFECONOMICS (PeterNewmaned., 1998). 31. For discussionon this point, see Lisa Bernstein,MerchantLaw in a MerchantCourt: the Code's Searchfor Immanent BusinessNorms, 144 U. PA. L. REV.1765, 1781-82 Rethinking (1996) [hereinafter Law in Bernstein,Merchant Law]; and Lisa Bernstein,Private Commercial the CottonIndustry: L. 99 MICH. Rules,Norms,and Institutions, CreatingCooperationThrough PrivateCommercial Bernstein, REV.1724(2001) [hereinafter Law].
558
TheYaleLawJournal
[Vol. 113:541
can be optimalfor the partiesto them, they often reducesocial welfare. Networksabsorbthe most reliablefirms,therebyreducingthe qualityof in the anonymous market. Hence,unlessa network exchanges encompasses muchof the economy,which is unlikely,the network's existencereduces efficiency in society generally.32Society is thereforebetteroff when it ratherthan when society adopts laws that improvemarketfunctioning on the ground eschewslegalreform will emergeas thatprivate associations substitutes. Second, marketexchangewill be commoneven satisfactory when businessnetworks exist. Both of these reasonssupport the utilityof the how state best facilitates in environments.33 asking contracting market Even so, the efficiency gains from enforcingcontractual promises the existence of and commercial transactions often contracts, presuppose are conducted withoutthem.Firmsoftenmakesimultaneous exchangesof cash for goodsor servicesrather thanexchange for the latertrade promises
32. For a generalanalysisof this issue, see, for example,Raja Kali, EndogenousBusiness & ORG. 615 (1999). Networks,15 J.L.ECON. 33. Recent theoreticalanalysesand economic experiments suggest that the domainof selfwould expandbeyondthatset out in the text if marketactorsbehavedfairly enforcingcontracts towardthose who hadbehavedfairlytowardthemandpunishedactorswho hadbehavedunfairly. In the experiments, to do so subjectsreciprocate good behavioreven thoughthey are not required andpunishbadbehavioreven thoughit is costly to do so. For a review of some of this evidence, see ErnstFehret al., Reciprocity as a Contract Device: Experimental Evidence,65 Enforcement ECONOMETRICA 833 (1997); and ERNST FEHR & KLAUS M. SCHMIDT, OFFAIRNESS THEORIES ANDRECIPROCITY-EVIDENCE ANDECONOMIC APPLICATIONS 2-3 (Univ. of Zurich,Inst. for Empirical Research in Econ., Working Paper No. 75, 2001), http://www.iew.unizh.ch/wp/ iewwp075.pdf. Analyses of how market contractingwould improve if parties engaged in fairnesscan be foundin YongminChen,Promises,Trustand Contracts,16 J.L.ECON. reciprocal & ORG. 209 (2000);andErnstFehr& KlausM. Schmidt, A Theory and ofFairness, Competition, 114 Q.J.ECON. 817 (1999). Cooperation, Whether this literature appliesto marketcontracting amongfirmsremainsan open question for two reasons.First, when individualsin the experimentswere put in marketcontexts and aboutthe payoffs of otheractors,reciprocity tendedto diminish.These deprivedof information resultscauseda leadingexperimenter to speculate: The effects of multiple players and limited information[in experiments]suggest a general conjecture about bargaining and markets. In two-person games with perfect informationabout how much each side is earning,fairness concerns loom largest.... The concern for fairness evident in two-player perfect information in such marketsdo games .., .disappearsin largemarkets.This does not meantraders not careaboutfairnessper se. Theymay care,butthey behaveself-interestedly because they aren't sure whetherothers are being fair and can't easily punish unfairness.A competitivemarketis simply a place in which it is hardto expressyour concernfor fairnessbecausebuyingor selling (or refusingto do so) will not generallychangeyour inequalitymuch. This... just means thatpeople will then express social preferences aboutunfairmarket outcomesthrough "voice"... , regulation, andlaw. were individual CAMERER, supranote 22, at 115. Second,the subjectsin the experiments persons who hadnot beentrained in market behavior.Suchsubjectsmay responddifferently fromofficers of firms,who commonlyhave marketexperienceand who are subjectto pressuresto maximize to applyresultsin the fairnessliterature to profits.These two reasonssuggestthatit is premature an analysis of contracting firms. For a discussionof the relevanceof the among sophisticated fairnessliterature to the enforcement of deliberately or indefiniteagreements between incomplete individual actors, see Robert E. Scott, A Theory of Self-EnforcingIndefiniteAgreements, 103 COLUM. L. REV.1641 (2003).
2003]
559
of these goods or services. Under such conditions,propertylaw is a sufficientencouragement to commerce becausea partywill only partwith or if the values more highly what is offered in party money goods to Thus, property supports efficiency.But, in contrast protecting exchange. simultaneous future exchanges,a contractis a set of promisesregarding behavior. Suchpromises arecostlyto makeandto memorialize. In orderto understand the role of the statein relation to contracting it thusis behavior, are necessaryto explainwhy partieswill incurthese costs when contracts enforceable but not otherwise. legally is helpful to contracting State enforcement partiesin a numberof in the two cases that SectionsIII.B contextsbut is particularly important is relation-specific andwhenthe and III.Cnext discuss-when investment of a badstateof the worldwouldcreateseriousdisruption costs. realization We analyze these cases both because of their intrinsic economic andbecauseof theirrelationto otheraspectsof contract law. significance For example,the need to cushionthe effects of realizingan adversestate thatwouldcreate and the need to avoid an adversecontract interpretation costs can causerisk-neutral act as if they were similardisruption to parties The recognitionof this shouldinfluencelegal doctrine.The risk-averse. cases. nexttwo Sectionsthustreatparadigmatic
B. EncouragingRelation-SpecificInvestment
We earlieridentifiedthe canonicalcontracting problemas ensuring efficientex post tradeand efficientex anteinvestment. We assumehere, the thatparties can ensureefficienttradeon theirown. per CoaseTheorem, Forexample,if the partieshad agreedto tradeten units,but it turnedout thattrading could twentyunitswouldmaximize joint gains,thenthe parties to providefor deliveryof the largerquantity. As we modifythe contract will see, ensuring efficientinvestment is moredifficult. we Theinvestments have in mind would include the productionof specializedgoods, the of humancapitalspecific to a particular to deal, or research development information about future costs or acquire prices. We developa simplemodelto explainwhy the stateplays an essential role in encouraging In the model, contractsare not legally investment. enforceable. Thesellersin ourstorycan function A markets. in two distinct seller can producea genericversionof a particular productand sell the genericin a competitivemarketat a price that equals cost (includinga return on the seller'sinvestment). The selleralso canproduce a specialized versionof the product for buyerswho arewillingto pay the extracost. To be precise,a buyer'svaluation forthe genericproduct is denoted v, andthe cost of the product is simplyg. Thus,the genericproduct will sell at the and generatea priceg (becauseprice equalscost in competitive markets)
560
TheYale LawJournal
[Vol. 113:541
we assumethatthe seller'sinvestment produce (wheres > g). Importantly, if the seller to make the specializedproductwould not be redeployable; wereto spends butthe dealwereto breakup,the sellerwouldlose all of s. The partiespreferto producethe specializedproductwhen that would maximizethe contractual This leadsto the efficientdecisionrule: surplus.
Producethe specialized version when v, - s > Vg- g.
contractual surplus of Vg - g. Suppose a particular buyer values the specializedversion of the productat v, and that the productcosts s to
To make our examplemore concrete,supposethe buyervalues the
specialized product at $80 (v, = $80); its cost is $50 (s = $50); the buyer values the generic version at $50 (vg = $50); and its cost is $40 (g = $40).
On these values, the parties'efficientdecisionrule requires the seller to It the a of produce specialized product: wouldgenerate surplus $30 while thegenericwouldgenerate a surplus of $10. If the partiescontract as the efficientdecisionrule requires, they will over how to dividethe expected$30 surplus. The divisionof the bargain will be determined the surplus by the bargaining play,andwe parties game assume for the reasons given above that they engage in deal-me-out Deal-me-out will generatean equal split in our bargaining.34 bargaining illustration if the parties areequallypatient becauseeachparty's bargainers for the the and for $0 seller) is less than disagreement payoff ($10 buyer one-half the $30 surplusfrom producingthe specializedproduct.We have becausefirmscommonly supposethatthe partiesare equallypatient, similar costs of capital, so thata contract the specialized to produce product wouldsplit the bargaining surplus equally.An equalsplit is achievedby a $65 price.35 Thepriceat whichthe parties will ultimately however,would transact, not be $65, becausethe buyer'sincentiveto cooperate vanishesafterthe sellerinvestss in the subject matter of thedeal.To see why,assumethatthe contract was madeas described. Afterthe sellerhad madeits investment, thebuyerwouldhavean incentive of the price.At to demand renegotiation thatpoint,the investment cost s would have been sunk and so would be in the new bargain: wouldbe whether Theonly issue fortheparties ignored to tradethe specialized would at to trade. some not or product Trading price a would of not the while produce grossgain $80, buyer'svaluation, trading produceno value. The partiesthus would trade,dividingthe $80 gain
34. See supranote 22. 35. The buyerearnsits valuation less the price,whichmustequalits payoff frombargaining. Since its valuationis $80 andits bargaining payoff is $15, the pricemustbe $65. The sellerearns the differencebetweenthe price and its cost, which must equal its bargaining payoff. Since the costs $50 to produceandthe seller'sbargaining specializedproduct payoff is $15, againthe price must be $65. Notice thata buyer'sthreatto exit unless it receives its disagreement payoff ($10) plus one-halfthe surplusafter this payoff is deductedwould not be credible:The seller would knowthatthe buyerwouldnot pass up $15, its payofffroma deal, to take$10, its outsideoption.
2003]
561
equally.If the price were reducedto $40, the buyerwould receive a $40 payoff and the seller would lose $10 ($40 less its cost of $50). Since the seller would lose its entire$50 investment if the partiesfailedto trade,it wouldagreeto thenew price. The lesson that this exampleteachesis not thatthe parties'ultimate transaction priceswoulddifferfromtheirinitialcontract prices;the point, is thatwhen contracts are unenforceable rather, a sophisticated seller will refuseto produce the specialized even thoughproducing it would product, maximizeexpectedsurplus. The sellerwouldanticipate $10 undera losing contract to produce the product rather thanearning$15 under renegotiated the initialcontract. the sellerwouldproduce the genericproduct. Therefore, This resultis inefficient. The genericproduct a social surplus of generates wouldhavegenerated a surplus of $30.36 $10, whilethe specialized product The parties would cooperateto produce the specialized product, however,if the buyer's promiseto pay the contractprice were legally enforceable. Underthe UCC(andthe commonlaw of contracts) the seller could treatthe buyer's demandto renegotiate ratherthan performas an breach."The sellerthenwouldbe entitled to recoverthe price anticipatory for couldnot be resoldat a reasonable if the goodscontracted Since price.38 the seller could not resell the specializedproduct for a positiveprice (its not to be redeployable), it thuscouldrecover investment, recall,is assumed the original wouldwrite $65 pricefromthebuyer.Knowing this,theparties the contract to producethe specializedproductand tradeit for $65. The sellerwouldanticipate andthebuyer for its investment, beingcompensated wouldprefer to havethe specialized a and realize $15 payoffrather product thanhavethe genericproduct andrealizea $10 payoff.
36. The Nash bargainingthat the text describesis ahistorical: Only the total surplus,the parties' discount rates, and (sometimes) their disagreementpoints determinethe bargaining outcome.Hence,the partiesin the examplesin the text areassumedto ignorethe seller'ssunk$50 investmentcost when renegotiating the contract.Individual and surveys personsin experiments sometimestake sunk costs into account,however, so that a party's payoff in a bargainwill increaseif the otherpartyknowsthatits partner Thiswill increasethe has spentmoneyto prepare. party's incentive to invest. See CAMERER, supra note 22, at 85-90; Lorne Carmichael& W. BentleyMacLeod,CaringAboutSunkCosts:A BehavorialSolutionto HoldupProblemswith SmallStakes,19 J.L.ECON. & ORG.106 (2003); JoepSonnemans et al., On the RelationBetween Asset Ownership and SpecificInvestments, entities J. 791 (2001). Whethercorporate 111 ECON. are motivated sunkcosts thatindividual by the fairnessconcernsregarding subjects experimental act uponis unknown.Whenpartiescan stage investments acrossperiods,thereis an equilibrium of the dynamicbargaining game they play in which partiesinvest more than the partiesin the modelwe use, even whenthose partiesareunableto enforcetheircontracts. See Yeon-KooChe & J6zsef Sikovics, A DynamicTheoryof Holdup(Apr.25, 2003) (unpublished on file manuscript, with authors).Whether the stage-investment assumption appliesin many cases is also unknown. Thus,we do not claim thatthe abilityof a partyto hold up its contract will always cause partner underinvestment-i.e., a refusal to make the specialized product-when contracts are not enforceable; rather,our claim is that underinvestment would occur often enough to make legal enforcement worthwhile. 37. See U.C.C.?? 2-610(b),2-703(e) (2003). 38. See id. ? 2-709.
562
TheYale LawJournal
[Vol. 113:541
This analysissupportstwo conclusions.First, contractremediesare thoughtto protectinjuredpromisees-the seller here-by awardingthe were not interest.This view is truebut shallow.If contracts expectation commercial seldom would themselves enforceable, sophisticated parties put in positionswherethey neededthe law's aid. They would insteadact as andsell it on wouldthe sellerhere,who wouldproduce the genericproduct the market rather thansubjectitself to exploitation. Enforcement actually empowerspromisorsby enabling them to make crediblepromises to performor to pay. The buyer in our example, when a contractis thuswantsthe powerto makea legallyenforceable-thatis, contemplated, a credible-promiseto pay the sellerthe $65 contract price.Enforcement, in sum, permitspartiesto makebelievablepromisesto each otherwhen or self-enforcement sanctions will not avail. reputational Second, and relatedly,our examplehelps to explainthe very small amountof foreigndirectinvestment thatprivatepartieshave madein the formerSoviet states and in many ThirdWorldcountries.Much of this investment would have been relation-specific (e.g., buildinga factoryfar from the home country,developinga mine or an oil field). Potential investors wouldnot deal unlessthe host country or local firmcouldmake than credible to the adhere to terms promises agreeduponrather originally of lack after The those terms been made. investments had renegotiate enforcement rulesandhonestcourtsin manyof these countries, however, the local from that believable are more prevents parties makingpromises thanwas the buyer'spromisein the exampleabove. In response,foreign reduceinvestment. Theabilityof a firmto makea credible parties promise, rare whichlawyersin developed is a regrettably countries takefor granted, powerin manypartsof theworld.39
C. ContractingToAvoid Disruption: The Case of VolatileMarkets
Partiesthat functionin "thick" have a choice of makinga markets40 contract for future delivery or making a spot purchase-that is, a simultaneous for future contract exchangeof cash for goods.A fixed-price to allocatethe riskof pricedeclinesto the buyerand deliveryis understood
39. For a vivid example of the second-best strategies used when attemptingcontract in weak states,see TimothyFrye& Ekaterina and enforcement Rackets, Regulation, Zhuravskaya, the RuleofLaw, 16 J.L.ECON. 478 (2000). As the authors & ORG. explain, [P]rivateprotectionracketsprimarilyprovidetwo services. First, they providebasic protectionfrom otherracketsand fromcriminals.Second,and to a lesser extent,they serves first as a These resultssuggestthatprivateprotection help enforceagreements. substitutefor the notoriouslyineffective Russianpolice forces. To a lesser extent, it also servesas a substitute for the notoriously ineffectiveRussiancourts. Id. at 491-92. 40. A thick market exists when there are many sellers and buyers trading a roughly homogenous product.
2003]
Law Limitsof Contract
563
to the seller.It is less well understood of priceincreases why partieswrite is costly andthe spot optionis available. these contracts when contracting andwe We set out another simplemodelthatstatesthis questionformally answer. to give a plausible thenattempt contract at timeTofor In themodel,theparties canmakean enforceable can wait until T, to makea spot of at time TI, or the buyer goods delivery The partiesbelieve at Tothatthe TImarket pricecouldtakeone purchase. of threevalues:(1) pk, (2) pk + z, or (3) pk - z. Eachof these outcomesis to be equallylikely;hence,viewedfromTo,the expectedTi priceis thought to deliverthe goods pk.41Thisalso will be thepricepaidat Tofor a promise would at T1.To see why,realizethatif the Topriceexceeded Pk, sellersat To for futuredeliveryat the high Toprice. enterthe marketto sell contracts The resultingincreasein supplywould cause the Toprice to fall. On the other hand, if the To price were below pk, buyers would enter to buy for futuredeliveryat the low Toprice.The resultingincreasein contracts wouldcausethe Topriceto rise. Hence,the uniqueequilibrium demand To a at if write contract As a the did be must To consequence, parties price pk. for T1delivery,the contract pricewouldbe pk. To completethe model,we or buy the goods is c ? pk andthat assumethatthe seller'scost to produce for the goods is v = pk + y. On these assumptions, the buyer'svaluation exceeds the goodswouldbe efficientbecausethe buyer'svaluation trading the seller's cost. The totalcost of writinga contract (insteadof the buyer > in a competitive is w 0. Since cost to until T, buy) priceequals waiting the seller's fractionof w will be includedin the price;the buyer market, of the price,plus the fraction thuswill bearall of w-the seller's fraction whichthebuyerpaysoutof pocket. Actualnumbers mightmakethis story a little more concrete.We let the w = $2. We assume furtherthat, for simplification, = $100 and pk in the valuation is to the market for buyer's good question, equal high price
so that z = y = $20 (and thus v = pk + y = $120). The question is whether
thebuyerwill makethe contract at Ti. at Toor makea spotpurchase In this model, the buyer would wait. If the buyer did contract,and contracts wereenforceable, the buyerwouldpay the pricepk andrealizeits net valuationy with certainty,either because the seller performedor a damage thuswould The buyer'sgain fromcontracting through recovery. On be its net valuationless contracting assumed our w. or costs, y
numbers,this gain would be $20 - $2 = $18. The buyer's expected gain as of Toif the buyer did not contract,however, would be the larger sum y, or $20. Withouta contract,the buyer would have to purchaseat the TI market price, whatever that price turned out to be. Since there are three equally likely futureprices, the buyer's expected returnfrom waiting until T1to buy
- = Pk. 41. The expectedT1priceis '/3(pk)+ ' (pk+ z) + / (Pk z)
564 is as of To), (measured

= E(Gnk)
/3
TheYaleLawJournal
[Vol. 113:541
[(Pk +
)-
(Pk+ z)] + 3 [(Pk +y) - Pk]+
/3
[(Pk+ Y)- (Pk-
) =y.
wouldeitherbe madeat the high T, market Thebuyer'sspotpurchase price to equalz, thaty is assumed Givena buyervaluation ofpk + y, andrecalling the buyer's expected gain from not contracting, and thus saving the cost w, is justthebuyer'snetvaluation contracting y, or $20.42 would not A risk-neutral a to ensurea certain gain; buyer pay premium a the would choose rather, party higher expectedbut risky gain. The here wouldbe the contracting cost (w = $2). Payingthatcost "premium" wouldguarantee the buyera sure$18 gain,whichis less thanthe preferred for the sellerwould rejecta contract expectedgain of $20. A risk-neutral samereason.In fact,spotpurchases would Andif parties areverycommon. not write contractsfor future delivery in thick marketseven if these were legally enforceable, contracts thereis no need to makethem legally enforceable. firms sometimeswrite Thus, the questionwhy risk-neutral forfuture contracts is not trivial. delivery We arguethatparties writethesecontracts (whenthey areenforceable) to reducethe risk of potentiallydisastrous outcomesthat would impose additional costs on the buyer.If the buyerin the exampledid not contract, thenone-third of the timeit wouldhaveto makea spotpurchase at the high + or also $120 Since the is valuation $120. buyer'sgross T1priceofpk z, + (that is, pk y), one-thirdof the time a purchaseof the goods would contribute toward A buyerwho failed thebuyer'sfixedobligations. nothing to pay rentor interest, costs.Once would incur serious however, disruption we incorporate costs (denoted asJ) andrealizethatthesewould disruption be incurred one-third of the time, the comparison in the model must be modified.The buyer'sactualdecisionrulewouldbe to contract when the net gain from contracting or exceededthe net gain fromnot contracting, when y - w > y - Af, or when w < '/ f In this example, if disruptioncosts exceeded$6, thebuyerwouldwritea contract forfuture delivery. This analysispredictsthat a buyerwho contemplates makingmany none of which will be in to relation the transactions, large buyer'sneed for cash,will act as woulda risk-neutral goods on the spot person,purchasing
market. When a bad realization on a single contract could seriously endanger the buyer's business, however, the buyer will act as if it were risk-averse,purchasingthe assuranceof performanceat the cost of writing a contract for future delivery. To be sure, some such buyers could guard against disruptionby maintainingsufficient cash reserves, but this strategy
42. Using our illustrative the buyer'sexpectedgain withouta contractis E(G,k)= numbers, ($120- $120) + ' ($120 -$100) + V ($120- $80) = $20.
(pk + z), the average T, marketprice (pk),or the low T1marketprice (Pk- z).
2003]
Limits of ContractLaw
565
is difficult for all businesses to follow. Also, if the buyer has good business prospects, the opportunity cost of hoarding cash likely would exceed contracting costs, especially for the simple fixed-price contracts we consider. The result of this analysis is that the state reduces social costs by giving parties that function in volatile markets the opportunityto make enforceablecontractsfor futuredelivery.43 In sum, when exchange is intertemporalrather than simultaneous, efficiency is enhanced when parties can make enforceable contractsin two principal situations: when at least one of the parties is requiredunder the contract to make an investment that is more profitable in the relationship than elsewhere, and when marketprices are volatile and an adverse market movement can have spillover effects.44 In the first case, in the absence of legal enforcementthe noninvestingpartyhas an incentive to renegotiatethe contractprice downwardratherthan to performunderthe original contract. In the second case, the party whom a market movement disadvantages may suffer disruptioncosts that would much exceed its expectation interest (as conventionally measured). Contracts do sometimes fall within the self-enforcing range and in some subeconomies reputation can make promises to performcredible. But nonlegal incentives can be ineffective in larger markets and in countries where social trust is low. Thus, without legal enforcement, private contractingparties cannot be expected often to createdeals that maximize social surplus. D. Enforcementand Duress Enforcemententails more than simply ordering a recalcitrantparty to perform. As an illustration, suppose that a seller sold its goods to a third party instead of delivering them to the contract buyer. A specific performanceorderwould thus be futile. Should the buyer be able to sue the seller for damages or to impose a constructivetrust on the proceeds of the sale? If the buyer makes a substitute purchase, is the seller's obligation discharged?Suppose instead that the buyer takes delivery of the goods but
43. We have arguedthat contractenforcementserves an insurancefunctionin the volatile markets case. Partiesalso can purchase but the transaction costs insurance, business-interruption of this alternativewould ordinarilyexceed the costs of the simple fixed-pricecontractswe consider.Companiesface pressureto settle "betthe ranch"lawsuitsfor reasonssimilarto those developedhere; settlementis a form of insuranceagainst being put out of business. See J.B. Pressure Heaton,Settlement 30-36 (Feb.2002) (unpublished on file with authors). manuscript, 44. Recallourassumption thatpartiesto businesscontracts A thirdmotiveto arerisk-neutral. contractis to transfer risk from more to less risk-averse of these parties.The legal enforcement contractssometimesis necessarybecausethe transferee of risk has an incentiveto breachwhen largerisksmaterialize. contracts are not considered here, in partbecauseone of the Risk-shifting partiesto them commonlyis an insurer, and insurance contracts are the subjectof a distinctand heavily regulatedlegal field. Moreover,althoughmany contractshave an insurancecomponent tendnot to give rise to litigation. (e.g., commodities contracts, currency hedging),thesecontracts
566
[Vol. 113: 541
claims that they do not conform to the qualitythe seller promisedto deliver. Must the buyer still pay the price and sue for damages or can the buyer cancel the sale? Finally, suppose that, after the contractis made but before delivery, a federal agency passes a regulation that prohibits productionof the product the buyer purchasedthe goods to make. The buyer no longer needs the goods. Must the buyer still pay the price? If not, does the seller have a remedy? These questions illustratethe complexity of the concept of "enforcinga contract."It is tempting to suggest, therefore, that a supplementaryset of publicly supplied "enforcement rules" also is needed. This suggestion would be premature,however, because parties can answer these questions in their own contracts. For example, parties can write a force majeure clause specifying the events that would excuse the seller's obligation to deliver or the buyer's obligation to pay; these events could include the possible passage of an unfavorable administrative regulation. That the contract laws of advanced nations commonly contain sets of enforcement rules thus requires explanation, a task that we address in Part V of this Article. The duress doctrine,however, is an enforcementrule thatpartiescannot create on their own. The law of duress applies in two contexts. Ex ante duress occurs when a party is wrongfully coerced to make a contract.45 Ex post duress occurs when a party is wrongfully coerced to modify an existing contract.46 Contractlaw applies the same legal standardin both cases: A contractor a modification is unenforceableif a party's consent thereto was obtained by an improperthreatthat left the party no reasonablealternative but to submit.47 Our focus here is on ex post duress, and we suggest that courts should ask a different question from those asked in ex ante duress cases. In an ex post duress case, the contact was fairly obtained and the parties could have provided for the situation that later arose had they thought about the issue. The court thus should ask whether parties with
45. An ex anteduressargument succeedsif the partyprovesthathe wouldnot have madethe coercive behaviorof the otherparty.The key is the challengedcontractabsentthe improperly law's focus on behaviorratherthan circumstances. Thus, it is not duress on the part of an employerwhen a poorpersonacceptsan offer to work at a low wage, nor is it duresson the part of a seller to chargea high price for gas when it is the only seller for manymiles. The employer does not createthe employee'slow wealth,nor does the seller lurethe buyerto the solitaryspot. Contractlaw thus requiresa duress claim to rest on the behaviorof the promisee,not on the of the promisor.Chouinard v. Chouinard, 568 F.2d 430, 434 (5th Cir. preexistingcircumstances 1978). This is becauseabsentcoercionby the promisee,the promisordoes betterby contracting thanby not contracting. See RobertE. Scott & WilliamJ. Stuntz,Plea Bargainingas Contract, 101 YALE L.J. 1909, 1919-20 (1992) (discussingthe generalproposition that"thewrongfulacts that constituteduress may be either physical force or an improper threat,but in any case the compulsionmust be producedby the promisee and not by exigent circumstances facing the promisor"). 46. See, e.g., AustinInstrument, Inc. v. LoralCorp.,272 N.E.2d533 (N.Y. 1971).
47. RESTATEMENT OFCONTRACTS (SECOND) ? 175 (1981).
2003]
567
sufficient foresight would have wanted the later modification agreementto be enforceable. Two examples show how this test should be applied. For the first, assume that the parties initially agreed to tradethirtyunits at a price of $10 each. Demand for the productturnsout to be higher than the parties initially believed; it becomes efficient to trade fifty units. The seller offers to transferan additionaltwenty units at a unit price, for all fifty, of $12. The buyer's profit would be higher underthe contractas modified, and it agrees to the new terms. The buyer cannot later claim that it was coerced to accept a price increase because, ex ante, the buyer would have wanted the court to enforce a modificationthatwould leave it betteroff thanperformanceunder the original contractwould have done. For our second example, return to the relation-specific investment model set out above. There, we claimed that permittingthe seller to sue for the price would deter the buyer's threatto renegotiate after the seller had invested. This claim is too strong because sellers in some cases could not make a credible threatto sue. The seller's threatwould be credible only if it had, or had convenient access to, the capital needed to sustain it until the buyer's performancecould be replaced. The buyer then would know that it would have to pay voluntarily or involuntarily. In contrast, a seller in a short-termbind (perhapsit had purchasedmaterialson credit and is facing demandsfor payment)may be betteroff accepting a low renegotiationprice than bringing a lawsuit. Recognizing this, the buyer may demand renegotiation even though the contract is enforceable. The renegotiated contract,however, would be a product of ex post duress. When the initial contractwas made, both partieswould have wanted a court not to enforce a purely redistributionalmodification-that is, a modification that would create no new wealth but rather would only redistributethe contractual surplus differently from the original contract. Parties dislike purely redistributional modificationsfor two reasons. As Section III.B showed, the of such modificationscan destroy the parties' incentive scheme anticipation for producingefficient specialized products.Further,the resourcesinvolved in negotiating the modification or guarding against it constitute a deadweightloss thatreduces the parties'joint gain fromthe contract.48 The duress doctrine thus permits the seller to perform under the renegotiated contract but later to reinstate the price term in the original contract. This is because modifications made under duress are not enforceable. As a consequence, the seller in our example could accept the
48. Ex post duresscases are largelyconsistentwith the test we propose,thoughthe courts standards. formallyapplythe Restatement's See, e.g., Wolf v. Marlton Corp., 154 A.2d 625 (N.J. 1959);AustinInstrument, 272 N.E.2d533. A moreextensiveanalysisof the ex post duresscase is RelationalContracts in Alan Schwartz, in the Courts:An Analysisof Incomplete and Agreements
Judicial Strategies, 21 J. LEGAL STUD.271, 308-13 (1992).
568
TheYaleLawJournal
[Vol. 113:541
$40 renegotiation price,deliverthe specialized goods,andlaterrecoverthe difference betweenthe $65 contract lowerrenegotiation and the price. price Sellerswho haveeasieraccessto cashin the long runthanin the shortrun, or who can sell theirlegal claim, will use the duressdoctrineto recover their original expectation.Buyers, therefore,will know that ex post states promisesby sellers thatdo not move the partiesto Pareto-superior Thelaw will permit themselves arenot credible. a sellerto renegeon sucha promise and sue for full payment.The unreliabilityof renegotiation coercedby duress reduces them(thatis, to the incentive to extract promises behaveas thebuyerin ourexample doctrine thusis The duress ex did). post an important function.49 enforcement aspectof a publiclysupplied In this Part,we haveargued thatfirmsneed stateenforcement in order to permitthemto makecrediblecommitments arenot whentheirpromises A courtcannotenforcea contract, however,withoutfirst self-enforcing. what the contract for state the determining says. Thus, parties'preference enforcement entailsa further of overthe set interpretive theories preference that courts could use to interpret We next attemptto their agreements. identify the interpretive theory in this set that typical firms will most commonly prefer.
IV. THEINTERPRETATION FUNCTION5o
A. TheRelevantInterpretiveQuestion
There is a consensus among courts and commentators that the is to have the enforcingcourt appropriate goal of contractinterpretation find the "correctanswer."The "correctanswer"is the solution to a
49. As may be obvious, the doctrineis symmetrical: It also applies to protectbuyerswho have made relation-specificinvestments from overreachingsellers. The doctrine would be if capitalmarkets wereperfect,or if damagesuitswereperfectlycompensatory. Then unnecessary parties could finance all good lawsuits or recover all losses. Capital markets,however, are anddamageawardsarenot alwaysfully compensatory. imperfect 50. PartIV concernscases in which the partieshave attempted to solve theirproblemwith writtenwords. The court,that is, is not called upon to fill gaps, but ratheris asked to discover what the partiesintendedtheirwrittenwordsto do. A theoryof interpretation has two aspects:a set of rules for determining the semanticcontentof a party'sutterances, and a set of rules for the legal significancethat should attachto the semanticcontent.We focus on the determining formerset of rules here becausewe have alreadydevelopeda normative theory:Courtsshould enforce businesscontractsas the partiesto them would want the contractsenforced.Our view of the rules determining regardingthe separability legal significancefrom the rules governing is not free fromdifficulty,however.Courtsdoing interpretation interpretation may be influenced outcomeis when the parties'meaningis not apparent. by theirview of what a good substantive See RichardCraswell,Offer,Acceptanceand EfficientReliance,48 STAN. L. REV.481, 551-52 on the assumption thatif the semanticcontentrulesimprove, (1996). We ignorethis complication the courts' need to give legal answersto factualintent questionswill lessen. A full theory of contractinterpretation would explore more seriously the actual and ideal relationbetween a court'ssubstantive andits choice of an interpretive preferences style.
2003]
569
contractingproblem that the parties intended to enact. Intention,however, is determinedobjectively and prospectively:A party is taken to mean what its contract partner could plausibly believe it meant when the parties contracted. There are two justifications for the goal of finding the correct answer. The first follows from an autonomy-based view of contract law. This justification holds that the exercise of state coercion against a person must be justified. A sufficient justification is that the court is making the person do what he had agreed to do." Hence, the court must ascertainjust what the person had agreed to do. The second justification is consistent with an efficiency-based view of contract law. On this view, parties contract to maximize the surplusthat their deal can create. This goal is unattainableif courts fail to enforce the parties' solution but rather impose some other solution. Thus, the court must ascertainthe solution that the parties actually adopted. There is a dispute in the literatureas to whether the rules that courts use when attemptingto find the correct answer are mandatoryin the formal sense, but there is a consensus that the rules are difficult for parties to escape.52 In our view, the current consensus asks the wrong question. A commitmentto party sovereignty regardingthe contract'ssubstantiveterms implies a further commitment to party sovereignty regarding the interpretivestyle an adjudicatorshould use to find the substantive terms. Partypreferencesregarding judicial interpretivestyles can differ. Therefore, should be defaults. The relevant question, then, is what interpretivestyles should be the majoritarian default. Put anotherway, the issue is not what interpretivestyle is best calculated to yield the correct answer. Rather,the issue is what interpretive style would typical parties want courts to use when attemptingto find the correct answer. We will argue here that the majoritariandefault is Willistonian: Typical firms prefer courts to make interpretations on a narrow evidentiary base whose most significant This proposed rule would both reverse component is the written contract.53 the UCC's interpretivestyle and make the new interpretivestyle a default.
51. Most autonomy-based theories are premised on either a notion of "consent"or the exerciseof will, suchas the makingof a promise.See sourcescitedsupranote 25. 52. A recentcommentator referred to the UCC's interpretation rules as "quasi-mandatory," the ideabeingthatthe rulesareand shouldbe very difficultfor partiesto avoid.DavidV. Snyder, and Formalitiesin Commercial A Defenseof Custom 54 SMU and Conduct, Language Contracts. L. REV.617, 648 (2001). Because contracting has positive costs, a quasi-mandatory rule will be in practicefor many parties,who will be unwillingto bear the additionalcosts of mandatory specifying alternativeregimes. Other commentators rules are believe that the interpretation See, e.g., Omri Ben-Shahar, The TentativeCase AgainstFlexibilityin Commercial mandatory. L. REV. Law,66 U. CHI. 781, 792 (1999). 53. See 4 SAMUEL CONTRACTS WILLISTON, ? 631, at 948-49 (WalterH.E. Jaegered., Baker, Voorhis& Co., Inc.3d ed. 1961)(1920) (explainingthatthe parolevidencerule "requires, in the absenceof fraud,duress, mutualmistake,or somethingof the kind, the exclusion of extrinsic evidence, oral or written, where the parties have reduced their agreementto an integrated
570
[Vol. 113: 541
B. TwoInterpretiveIssues.-Problems of Meaningand of Language We begin by clarifying two interpretive issues that are commonly commingled: What does the language of the contract mean? And in what language was the contract written? To see why these issues are distinct, suppose that there are two sets of linguistic communities. The first set, called M, consists of a single linguistic community. This M community is composed of judges, lawyers, business persons, and potentialjurors. The members of M read and write a language thatwe call "majoritytalk," since it is the language that people typically use when communicatingwith each other. The second set is called P, and it has many linguistic communities.A community in the set P may be as small as the parties to a particular contract,or as large as the entiretradein which the parties function. The set P has many communities because there are many party dyads and many trades. The members of each community in P may write contracts in their own "partytalk"or write them in majoritytalk.54 The existence of multiple linguistic communities raises the two interpretiveissues just noted. Imagine that parties are engaged in a dispute regardingthe meaning of their written agreement.Words can be vague or ambiguous.5 If the parties agree on the language in which their contract was written, the court's interpretivetask is limited to finding what the parties intendedthat languageto say. If the partiesdivide on the question of what language they used, the court's interpretivetask expands: The court
writing").Willistonianformalismrests on two basic claims: (1) that contractterms can be interpreted accordingto their plain meanings, and (2) that written terms have priorityover unwritten See DennisM. Patterson, GoodFaith, LenderLiability,and expressionsof agreement. and the UniformCommercialCode, DiscretionaryAcceleration:Of Llewellyn, Wittgenstein, 68 TEX. L. REV.169, 187-88(1989). 54. Ouruse of the word"language" is a little loose. Linguistswould say thatwhen everyone uses Englishwords,majority talk and the varioustypes of partytalk are differentdialectsof the same language. We use a multilanguage becauseit seems to us to be a moreconvenient descriptor systemof classification. 55. Courtsseldom distinguishbetween "vague"and "ambiguous" terms.A typicaljudicial definitionof ambiguity, for example,includesany termor wordthat"hasno definitesignificance or... is capableof morethanone sensibleandreasonable Ross Bros. Constr.Co. interpretation." v. State,650 P.2d 1080, 1082 (Or. Ct. App. 1982). Morenarrowly, however,a word is vague to the extentthat it can apply to a wide spectrum of referents, or to referentsthatclusterarounda modal "best instance," or to somewhat different referents in different people. See, e.g., FrigalimentImportingCo. v. B.N.S. Int'l Sales Corp., 190 F. Supp. 116 (S.D.N.Y. 1960) includesall types of chickenor only a subset);Highleyv. Phillips, (analyzingwhether"chicken" 5 A.2d 824, 826 (Md. 1939) (determining whetherthe sale of "allthe dirt"froma tractrefersalso to subsurfacesand). In contrast,"ambiguity" requiresat least two distinct,usually inconsistent 241 F.2d 312, 314-16 (5th Cir. 1957) meanings.See, e.g., PetroleumFin. Corp. v. Cockburn, in a telegramthat supported two differentreadings);Raffles v. (analyzingmissing punctuation the seller's obligationwhen two Wichelhaus,159 Eng. Rep. 375, 375 (Ex. 1864) (determining to one of them and the ships namedPeerless sailed from Bombaybut the seller was referring it as the other).Language buyerunderstood commonlyis vaguein the sense thatthe set of objects to whicha wordappliesis rarelydelineated occurs withabsoluteprecision.Truelexicalambiguity infrequently.
2003]
571
now must ask initially whether the parties wrote their contract in majority talk or in a particularprivate language. This question raises a separate interpretiveissue because a contract's language could plainly mean mMin communityMbut also plainly mean mpin a communitywithin P, where mM and mp differ. For example, the word "wife" in a sentence in John's will reciting "I leave my money to my wife" would mean in the M community that John left his money to the woman to whom he was legally married when he died. In a well-known case, however, John and anotherpartywrote a contractin orderto dispose of part of John's estate upon his death.56 This contractused the word "wife" to refer to the woman with whom John was living when he died but whom he had never legally married;John had desertedhis legal wife years before without divorcing her. The word "wife" in this example would be clear to members of the single communitywithin set M and would also be clear to the particularcommunity within P to which the contracting parties belonged. The two meanings differed, however." The same word or phrase can thus have different meanings in different linguistic communities, which requires courts to make a choice: Should courts permit parties to write contracts in the language of the parties' choosing, or should courts create an incentive for parties to use majority talk by interpretingtheir agreements as if the agreements were writtenin that language?58 There are two traditionalapproachesto finding the correct answer to questions of contractualmeaning and contractuallanguage. They differ in
56. In re Soper'sEstate,264 N.W. 427 (Minn.1935). 57. Fora modemexample,see Columbia NitrogenCorp.v. RoysterCo., 451 F.2d 3 (4th Cir. 1971). In that case, the contractspecifieda "Minimum TonnagePer Year"of 31,000 tons. The the buyerto introduce evidencethat,in buyertook less thanthatamount,but the courtpermitted the parties'trade,statedminimumtonnageswere "mereprojections to be adjustedaccordingto market forces."Id. at 7. 58. The distinction betweenpartyandmajority talkcan blurat the edges, butremainseasy to drawin most cases. Whenpartytalkclosely correlates with majority talk-e.g., when a contract's to the same terms termsareused to convey a subsetof the generalmeaningsthatcan be attributed in majority talk-then the line betweenmajority andpartytalkmay be difficultto draw.Consider a contractthat requiresthe seller to deliver "redsweaters."Given the breadthof the category "red" in majority talk, evidenceregarding usage of trademay be helpfulin delimitingthe shades of red that the partiesconsideredpermissible.It remainsunclear,however,whetherthe parties intended to convey a more limitedsubsetof meaningfor the term"red," specific to the parties,or sought instead to convey the broadermeaning associated with the majoritydefinition:The distinction betweenmajority andpartytalkcan be difficultto drawin such an instance. termsthat obviouslycontradict By contrast,contractual providea majorityunderstandings clear line between majoritytalk and party talk. For example, in Hunt Foods & Industriesv. Doliner, DolinerprovidedHuntFoods with an "unconditional" writtenoption to buy all of the stockin his companyat a statedpriceby a stateddate.270 N.Y.S.2d937, aff'd, 272 N.Y.S.2d686 (App. Div. 1966). The parties,however,intended"unconditional" to mean conditional.When a contract'stermsso patentlycontradict majoritymeaning,the contractis clearly writtenin party talk. The New York court's decision to permit Doliner to testify as to the contract'sactual meaningcan be seen as a permission for the partiesto write in theirown privatelanguage. Id. We return to the two-language distinctionin a moment.It is also discussedfurtherin Section IV.D below.
572
[Vol. 113: 541
To the scope of the evidentiarybase each requiresto make interpretations. be sure, a minimum evidentiary base is required for any coherent This minimumbase is denoted here by Bminand is composed interpretation. of the parties' contract, a narrative concerning whether the parties performed the obligations that the contract appears to require, a standard English language dictionary, and the interpreter's experience and understanding of the world. A Willistonian, or "textualist," theory of assumes that contractsoften have "plain meanings" that are interpretation Put formally, this view asserts that a court apparentto judicial interpreters. answer on the correct find the can evidentiarybase Bmin. Courtshave added five evidentiary categories to this minimum base: (1) the parties' practice under prior agreements; (2) the parties' practice under the current agreement; (3) testimony as to what was said during the negotiations; documents(memoranda, priordrafts,letters);and (4) writtenprecontractual the agreement's words what custom relevant to determining (5) industry admits evidence in all meant to the contractingparties. When an adjudicator five additional evidentiary categories, we denote the evidentiary base as Bmax. A "contextualist"theory of interpretationholds that a court is more likely to find the correct answer if the evidentiary base expands
toward Bmax.59
A Willistonian theory of interpretationhas the obvious effect of and truncatingthe evidentiarybase that a courtuses to make interpretations, it also has the effect of creatingan incentive for partiesto write contractsin majoritytalk. Because the same word can have plain but differentmeanings in the single communitywithin set M and in a particularcommunitywithin set P, a party seldom could establish that its contractwas written in party talk unless it could introduceextrinsic evidence.60 The Willistonian theory bars this evidence.61 In addition,on this theorythe courtprimarilyconsiders the text of the contract; hence, the parties likely can reduce interpretive errorsby writing the contractin the court's language.Thus, the Willistonian theory actually resolves two distinct issues: what language the parties should speak and what evidence is admissible to show what the parties meant in the permitted language. Contextualistsresolve the same issues, though differently.Their theory lets courts consider all materialevidence to resolve interpretive issues; the practical effect of so widening the
scholarsdo not necessarilyinsiston Bmax. It is recognized,for example,that 59. Contextualist evidencein category(3) is less reliablethanevidencein category(4). 60. Extrinsicevidence refers here to evidence in the five additionalevidentiarycategories listed in the text. 61. A court applyingthe Willistoniantheorywill admit extrinsicevidence only when the contract'slanguageis vagueor ambiguous on its face. See Pysell v. Keck,559 S.E.2d677, 678-79 (Va. 2002).
2003]
573
evidentiary base has been to permit parties to write in whatever language they choose.62 Any analysis of contractual interpretationthus should answer two questions. The first is the question of meaning: Should courts use broad or narrow evidentiary bases in determining the meaning of the contract's language? The second is the question of language: Should the "linguistic default" assume that parties wrote in majority talk unless their contract recites otherwise, or should courts always admit the possibility that the parties wrote in a private language? The practical effect of admitting this possibility is to permita partyto introduceevidence in all of the evidentiary categories to show what language the parties actually used. Section C next argues that when the issue of the contract's language is settled but the meaning of that language is arguably unclear, the majoritariandefault should require courts to make interpretationsusing the evidentiary base Bmin.Section D then arguesthat the linguistic default should suppose parties to have used majoritytalk. C. TheParties' PreferencesRegardingInterpretiveStyles An interpretivestyle can be assessed along two dimensions: (1) the likelihood that the style will generatethe correctanswer (as defined above); and (2) the costs that the style imposes on courts and parties. We model the performanceof judicial interpretivestyles on these dimensions in two ways that, in combination, capture most of the relevant cases. Both models suppose that the contractis complete in the sense that the writing expresses the parties' solution to the contracting problem at issue. The parties are to a later aware,however, that their meaning may not always be transparent thus that a face the court's interpreter. They knowingly possibility interpretationmay deviate from the correct answer. Our initial model considers the set of cases in which the parties' payoffs under their contract are monotonic and continuous in the space of possible judicial A party's payoff, that is, increases as interpretations of the interpretations. contract become more favorable to it and decreases as interpretations become less favorable. In this set of cases, firms prefer courts to make interpretationson the minimum evidentiary base Bmin except in unusual circumstances. We then consider a set of cases in which payoffs are
62. In a well-knowncase, ChiefJusticeTraynor stated: The fact thatthe termsof an instrument appearclear to a judge does not precludethe to express different possibilitythat the partieschose the languageof the instrument terms.Thatpossibility... exists wheneverthe parties'understanding of the wordsused may have differedfromthejudge's understanding. rationalinterpretation Accordingly, requiresat least a preliminary consideration of all credibleevidence offeredto prove the intention of the parties. Pac.Gas& Elec. Co. v. G.W.ThomasDrayage& RiggingCo., 442 P.2d 641, 645 (Cal. 1968).
574
[Vol. 113: 541
invariant to the extent of judicial error. Here, every unfavorable that deviates from the correct answer has the same negative interpretation is close to consequences for a firm, regardlessof whetherthe interpretation the correct answer or far from it. In this set of cases, firms have reasons to prefer a broader evidentiary base than Bmin, but, we argue, litigation cost considerationswill still incline a majorityof these firms to prefer courts to use a narrowevidentiarybase. Taken together,the two models suggest that when the question of contractuallanguage has been settled but the issue regardingwhat that language means remains, the default interpretivestyle for courts should be textualist. In both models, parties negotiate a contract whose object is to maximize the surplus the deal could create, and they divide that surplus throughthe price term. The parties want a court, should a dispute arise, to give the correct answer to an interpretivequestion. Anticipating that the answer will be correct permits parties to take surplus-maximizingactions. Parties sign the contractwhen it is optimally clear, in a sense to be made more precise below. 1. The Continuous-Payoff Case In this case, we focus for convenience on a buyer. We denote the surplus under a deal as s*; the buyer's share is the difference between its valuation v* and the price: Sb* = v* - p. The judicial interpretationthat gives the correct answer is denoted i*. Thus, if the court does make the interpretationi*, the contractual surplus will be s* and the buyer will realize sb*. The illustrative contract we consider requires the seller to preparemachines prior to delivery so as to minimize the buyer's costs of adjusting the machines for their intended use. Subsequently, the parties disagree over whether the seller fully complied with its duty to preparethe machines. If the court imposes a lesser preparation obligation on the seller than the obligation that the contract, correctly interpreted,would require, the deal will be less profitable to the buyer, and its gain will fall to on the seller than i*, the sb(-) < Sb*. If the court imposes a greaterobligation > will in the space of rise to Each buyer's gain interpretation sb(+) Sb*. thus generates a particularpayoff for the possible judicial interpretations buyer. In this productpreparation example, the buyer's possible payoffs are monotonicand continuousin the space of possible interpretations; that is, as become more favorable to the buyer, its payoffs increase, interpretations and vice versa. The seller's possible payoffs also are monotonic and continuous,but they decline as the buyer's increase. When analyzing the parties' preferences regarding interpretivestyles, we begin at the litigation stage and make three assumptions:that (1) a court
2003]
575
that relies solely on the minimum evidentiary base Bmin63 will find the correct answer i* with positive probability, (2) that the likelihood that a court will find i* sometimes can be increased if the court considers evidence in categories additionalto Bmin, and (3) that courts are unbiased. We initially support the first assumption with the following reasoning. Since the written contract trumps evidence in the other evidentiary categories,64and disputes are expensive, parties benefit from producing a writing that makes clear to a later court what was promised. Therefore,it is plausible to suppose that courts sometimes can discern the parties' exact intentionsfrom the evidentiarybase Bmin alone. For the second assumption, assume that the parties specified the seller's preparationobligations in considerable detail in a preagreement memorandum that both parties signed. Admitting the memorandum into evidence will increase the likelihood that a court would give the correct answer to interpretation questions about productpreparation.Regardingthe third assumption,there is no reason to believe that courts will systematically deviate from the correct answer i* in ways that are more or less favorable to particular partiesor classes of firms. These three assumptions permit us to state formally a party's expectationat contractingtime regardingthe possible payoffs that a judicial could induce. If courts are unbiased and can find i* on the interpretation base Bmin, but can also err, then the buyer's expected payoff in our illustration,given what a court laterwill do, can be expressedas:
E[sb(i) Bmin] = Sb(i*) + -.
The left-handside of this expression is the buyer's expected payoff given a i made on the evidentiarybase Bmin. The first term on judicial interpretation the right-handside is the buyer's payoff given the correct interpretation i*; the second term is an errorterm with mean zero and positive variance. That F has mean zero means that the court is unbiased. In other words, the court is as likely to make an interpretation that is more favorable to the buyer (less favorableto the seller) than the correctanswer as the court is likely to make a less favorable interpretation.Judicial errors therefore cancel, in expectation. The errorterm has positive variance, however, because in an actual case a court's interpretation can deviate from the correct answer.65
63. RecallthatBmin is composedof the writtencontract, a dictionary, a performance narrative, andthe interpreter's of the world. experienceandunderstanding 64. Section 1-303(e)(1)(formerlysection 1-205(4)) of the UCC providesthat when thereis conflict,"express termsprevailover courseof performance, courseof dealingandusage of trade." 65. For readersunversedin statistics,varianceis a measureof how far an outcome can deviatefromthe meanof a distribution. Thus,if a courtis unbiasedbut the varianceis large,the thatthe courtmakesin an actualcase may well be very far fromthe correctanswer; interpretation when the varianceis small,the courtis likely to be close to the correctanswer.If the conversely,
U.C.C. ? 1-303(e)(1) (2003).
576
[Vol. 113: 541
The expressionthus says (in English) thatwhen a court restrictsitself to the evidentiarybase Bmin, the buyer's expected payoff equals the payoff thatthe buyer expected to get when it agreedto the contract. When the variance in the error term is large, the parties know that a court's answer is likely to deviate widely from the correct answer. Our example of the preagreement memorandum regarding the seller's performanceobligation shows that adding evidence to Bmin can shrink the variance. Simply put, if the parties here know that the evidentiarybase in a later lawsuit will include preagreementmemoranda,they will expect later to be close to the correctanswer. interpretations This specification of the relation between judicial error and party expectations implies that firms commonly prefer courts to be restrictedto the minimum evidentiary base Bmin when payoffs are continuous in the To see why, recall our second assumption: space of possible interpretations. As the evidentiary base approaches Bma,,x, the variance in the error term approaches zero. A risk-neutral party cares about the mean of the distributionbut not the variance.This is because the variance interpretation term measures risk while risk-neutral parties are indifferent to risk. Therefore, it is enough for a risk-neutral firm that the expected i*. Put another way, a E(i) equals the correct interpretation interpretation on firm's preferenceat contracttime is to have courts make interpretations the minimumevidentiarybase unless it would be costless to widen the base. But it is not costless. As the permissible evidentiarybase widens, each party has incentives to introducemore evidence and, in turn,will need to contest more evidence. Since trials are expensive, risk-neutral firms are Willistonians.66 This view should not be overstated. Some firms will have a different preference. Recall the volatile markets example in Section III.C above. There, we argued that firms functioning in volatile markets would make spot purchasesunless a particulartransactionwas importantrelative to the size of the firm. In that event, a firm often would preferto incurthe costs of making a contractin orderto avoid the costs of disruptionthat a bad market realizationcould cause. A similarpreferencecould obtain here. Thus, when performance of a particularcontract is importantto the survival of the firm-say, a contractwith a major supplier--or when the contract is new and is expected to be widely used, the firm may be unwilling to risk a seriously adverse interpretation.If contextualists are correct that larger
contextualist claimis correctthatbroader moreaccurate basesgenerate interpretations evidentiary (i.e., the varianceshrinksmore),it follows that litigationwhere the courtsees only Bminis more evidence;when the courtsees only riskyfor partiesthanlitigationwherethe courtsees additional its interpretation is morelikely to deviatefurther fromthe correctanswer. Bmin, 66. There is considerableevidence that firms prefer a formalist adjudicatory style. See Merchant Law, supra note 31, at Bernstein, Law, supra note 31; Bernstein, Private Commercial 1735-37.
2003]
577
evidentiarybases do shrink variance, then parties concerned with variance will likely prefer that courts use a contextualist adjudicatorystyle. In the example above, if the machines were crucial to a start-upventure and the buyer had little cash, the buyer ex ante would want the evidentiarybase in a later suit to be sufficiently broad so that the memorandum could be admittedinto evidence. On the other hand, only unusualcontractshave this "bet the ranch"quality. In the typical case, it is good enough that courts get things right on average. We now return to the contract-creationstage to consider a deeper justification for our first assumption. It is optimal for risk-neutralfirms to invest resources in drafting until the writing is sufficiently clear, in an objective sense, so that the mean of the distributionof possible judicial is the correctinterpretation i*. Contractssketchedout in less interpretations detail than this would generate interpretationdistributions whose mean could be anywhere. As a consequence, parties could not expect courts to protect their expectation interest in case of a dispute. As this would be inefficient, firms will attempt to write contracts with sufficient clarity to permitcourts to find correctanswers, though with error. The currentdebatebetween textualistsand contextualistsis irrelevantto for cases that resemble the performance-preparation issues of interpretation illustration.In that illustration,there was consensus regardingthe language in which the contractwas written, but that language was arguablyunclear. The contextualist in effect asserts that a larger evidentiary base shrinks variance. Indeed, if the base is large enough, the varianceapproacheszero, so parties will believe when they contractthat a court will find the correct answerwith certainty.As applied to our illustration,taking into account the memorandum,perhaps together with an industry custom as to a seller's preparationobligations, would leave little doubt regardingwhat the seller was supposed to do. The textualist, in contrast, claims that variance does not shrink materially with a broader evidentiary base because contracts often have plain meanings. Hence, permittingpartiesto introduceadditional evidence as to intent would generate costs in excess of gains. The proponentsin this debate thus disagree on the relationbetween the width of the evidentiarybase and the accuracy of a court's interpretation. But it is for courts to decide which side of the textualist-contextualist unnecessary debate is correct. Greateraccuracy is lower variance, and business parties commonly are indifferentto variance. Thus, courts that interpretcontracts as typical parties prefer would be indifferent to variance as well, and sensitive only to the costs of administeringtheir evidentiarystandard. These courts would thus make interpretations on the evidentiarybase Bmin unless partiesinstructedthem to the contrary.
578
TheYaleLawJournal
[Vol. 113:541
2. TheDiscontinuous-PayoffCase
We illustrate this case by recallingthe specialized-product examplein SectionIII.B.We denotethe qualitythatthe contract the required sellerto as The seller produce qk. producesqk but the buyer,possibly sincerely, claimsthatthe contract a higherqualitylevel. Theproduct's actual required is not in the issue is If a what the contract said. court quality dispute; decides that the contractrequireda lower quality level than what was as q(-) < qk. If the court actuallydelivered,we denotethe interpretation decidesthatthe contract a required higherqualitylevel thanwhatthe seller as q?) > qk. Just as before, we delivered,we denote the interpretation assumethatthe courtis unbiased. the meanof this three-point Therefore, distribution is qk: The seller's expectationregardinga later judicial of thecontract canbe written as E(i) = qk. interpretation The seller'spayoffshere,however,arenot continuous in the spaceof the court'spossibleinterpretations, becauseof theperfecttender rule.67 If a courtfinds thatthe contract a that than level was less or required quality to the the seller the can recover the seller equal delivered, quality price;if the courtfinds thatthe contract a higherqualitylevel, the buyer required can rejectand crediblythreaten a suit for damages.The seller then will realizea renegotiation payoffthatlikely is less thanthe pricebecausethe damage threat reduces its bargainingpower. The fact that judicial are unbiaseddoes not matterfor this seller, becausethe interpretations seller does not gain fromjudicialerrorson the low side of q (it cannot recoveran amount thanthe priceif it has delivered a qualitylevel greater thatis higher thanthe level the courtrequires), whilethe selleris harmed by on the high side of q, howeversmall(thebuyercan rejectif judicialerrors the courtinterprets thecontract to require evena slightlyhigherquality than the contract The actuallyrequired). sellerthuswantsthe courtto find the correct answer withcertainty, eventhough the selleris risk-neutral. The seller's preference has an efficiency implication. In the case, the possibilityof judicialerrordoes not createan continuous-payoff ex ante inefficiency: There,when the courtis righton average,a party's under the contractequalsits expectation interest.In the expectedpayoff case, we will see, when a court is only right on discontinuous-payoff average,the seller's expectedpayoff underthe contractis less than its
expectationinterest.An efficiency-minded court thus could prefer to widen the evidentiarybase in order to shrink the possibility of error,and thereby betterprotecta party's expectation.
67. See U.C.C. ? 2-601 ("[I]f the goods or the tenderof delivery fail in any respect to conformto the contract, the buyermay ... rejectthe whole .... .").
2003]
579
To see how the possibility of error in the discontinuous-payoffcase can create an inefficiency, and to see why we nevertheless resist the view that the default for this case should be a wider evidentiary base, we generalize the formalizationused above. We now suppose that the parties face two possibilities regardinga court's interpretation of their agreement. First, there is a positive probabilitythat a court may find the correctanswer with certainty on the evidentiary base Bmin. For example, the interpreting judge may be commercially sophisticated or the breach narrativemay be quite revealing. Second, there is a positive probability a court may err. When error of some kind is possible, the parties must specify an "error distribution": how errors in particularcases may be manifested. As in the initial model, we let partiesassume that courts thatcan err will be unbiased; in expectation, that is, courts are assumed to be correct on average, but parties know that any actual interpretationmay be off the mark. We formalize this scenario by letting P be the probabilitythat an interpreting court will find the correct answer with certainty,and letting (1 - 0) be the of the probabilitythat the court will make only an unbiased interpretation contractterms regulatingquality. To simplify this presentation,we continue with our assumptionthat an unbiased court will interpretthe contracteither correctly, or to require a higher quality level than the parties actually intended, or to require a lower quality level than was intended. Each of these possibilities is assumed to occur with the same probability, that is, one-third. We complete this story by denoting the contract price--the seller's expectation-as k and the seller's payoff in a possible renegotiation as r < k. This permits us to write the seller's expected payoff under the contractas of the time the partiesmake the deal:
E(ss) = pk+ (1 - P)(%k +
V3 r).
With probability0, the seller receives the payoff that the contractdirects, which is the price k, because the court will interpretthe contractcorrectly. With probability(1 - P), the court will make an unbiased interpretation of the quality level the contract required.Hence, when the seller is found to have complied or overcomplied, which in total occurs two-thirds of the time, the seller receives the contractualpayoff of k, and when the seller is found to have undercomplied,which occurs one-thirdof the time, the seller receives the renegotiationpayoff of r.68
68. To see why the new formalization we can rewritethe generalizesthe originaltreatment, initial model so that courtswill sometimesfind the correctanswerwith certainty.The buyer's expectedpayoffin the continuous-payoff case thenis expressedas
= + (1 - +)[sh(i*) + = Sh(i*) E(sb IBmin) shb(i*) becauseEhas mean zero: The buyer'sexpected payoff equals in expectationthe correctpayoff. When this more general formalizationis used, the second term in brackets will equal the expectation interestin the continuous-payoff in case, butthis termcan be less thanthe expectation
580
TheYale LawJournal
[Vol. 113:541
The seller'sexpectedpayoffunderthe contract-E(s)--ordinarilywill be lowerthanthe seller'sexpectation whichis the contract interest, priceof k. This is because P commonlyis less than one (the court is expected andr commonly is sometimes not to findthe correct answer withcertainty) in less thank (the seller expectsto receiveless thanthe contract price a Since protectingthe expectationinterestis efficient, the renegotiation). can increase surplusby raisingthe seller's expectedpayoff. Our parties therelationship betweenthewidthof the evidentiary regarding assumptions of a court'sinterpretation baseandthe accuracy P implythattheprobability answer base expands of a court'sfindingthecorrect risesas the evidentiary beyond Bmin. Because the seller's expected payoff under the contract as p increases-as the courtbecomesmoreaccurate-theparties, increases have reasonto prefera broader base than thoughrisk-neutral, evidentiary the evidentiary baseis not costless,however.Theparties, Bmin.69 Expanding face a tradeoff between and the efficiencyof increased therefore, accuracy the inefficiency of increased costs. contract-enforcement Two factorsinfluencehow partieswill make this tradeoff.The first the factoris the bargaining powerthe seller will have in a renegotiation; greaterthe seller's bargaining power,the more closely the renegotiation the contractual payoffr will approach payoffk, andthe closerthe seller's return will be the to contract expected price.Hence,whenthe sellerexpects to have bargaining power,the partiesex anteare less likely to preferthe of extensive possibility discoveryand trials conducted just in orderto increase13-the likelihood thatthe courtwill makea correct interpretation. Recallthatbargaining poweris a functionof the parties'relativediscount arethe same,andthe parties'disagreement rates,whichcommonly points. Theseller'sdisagreement its of howredeployable pointis largelya function investment is. The seller'sbargaining when,as in the poweris minimized in SectionIII.B,its investment illustration at all because is not redeployable thenits disagreement payoffwill be negative. In makingthe tradeoffbetweenaccuracyand cost, partieswill also considera secondfactor: how mucha given piece or categoryof evidence will increasethe likelihoodthata courtwill makea correctinterpretation. In the exampleabove,the introduction of a single piece of evidence-the memorandum-was assumed to increase materiallythe preagreement thatthe courtwouldfind the correct answer.The memorandum probability
thus was very productive, generating a large increase in accuracy at a low
the discontinuous-payoff above. This is becausewhen courts case, as it is in the textualtreatment err in the lattercase, the seller is exposed to the possible downsideof receiving less than its in a renegotiation, withoutany concomitant expectation upside. 69. In additionto expandingthe evidentiary base, the partiesalso could increasesurplusby See infra text accompanying notes 72-77. Since the changingthe contract'spayoff structure. buyerwould benefitthroughthe price termfromactionsthatincreasedsurplus,the buyerwould agreeto eithermethodwhenthe methodwouldbe cost-justified.
2003]
581
cost. When evidence is expected to be less productive,parties will be less inclined to have courtsmake interpretations on a broadevidentiarybase. This analysis of the relevant tradeoff suggests that parties in the discontinuous-payoff case would more commonly prefer a narrow evidentiary base for interpretations.Many more deals are for relatively standardgoods that can be improved by a party's specialized investment than for goods that would be worthless in uses other than those the parties contemplate.Thus, parties such as the seller here ordinarilycould redeploy a significant fraction of their investment, and so would not be seriously disadvantaged in a renegotiation. Further,the typical choice that parties expect later to face in a contract action is not whether a particularpiece of evidence will be admitted or not; the choice is whether a court will make interpretations on a motion for summaryjudgment-i.e., on the base after a often before a jury.70 Trials can be very costly. Thus, trial, Bmin--or would broad prefer parties evidentiarybases either when their performance would be highly specialized or when an evidentiarycategory in additionto Bminwould be very productive-a clear custom, for example.7' Since these illustrationsappearto captureonly a minority of cases, we suggest that the majoritarian preferencein the discontinuous-payoffcase also is for courtsto use narrowevidentiarybases when making interpretations. Turningagain to the contract-creation stage, typical parties would invest sufficient resources in draftingjust to create the evidentiarybase that would permit a court Brin to make an unbiasedinterpretation. In addition,parties have the ability to protectthe seller's expectation in the discontinuous-payoff case without expanding the evidentiary base. Their method is to change the contract's payoff structure.The structureof the discontinuous-payoffcase creates the possibility that a partywill incur a large loss because of a small judicial error.The seller in the example here thus could suffer substantiallyif the courtheld that it to had to produceonly
70. Courts frequentlyconsider extrinsic evidence when interpreting the written contract. A commentator recentlyexplained: Extrinsicevidence includes both evidence about trade customs and evidence about betweenthe parties--concerning the courseof performance interchanges of the current contract,the course of dealing in prior transactions, or the bargaining history of the current contract. Very often this extrinsicevidencewill not be solely documentary and evaluation of oraltestimonyaboutconversations will require betweenthe parties. WilliamC. Whitford, TheRole of theJury (and the Fact/LawDistinction)in the Interpretation of Written 2001 WIS.L. REV.931, 937. Over50%of contract Contracts, cases triedin federalcourts are to juries, and between 25% and 30% are tried to juries in state courts.See MarcGalanter, Contractin Court;or AlmostEverythingYouMay or May Not WantTo KnowAbout Contract 2001 Wis. L. REV.577, 591, 598 tbl.3, 602 tbl.5, 605 tbl.7, 625 tbl.15. Litigation, 71. RichardCraswellshows, however,that courtsevaluatethe of customsby probativeness evaluatingthe purposesthat the customs are meant to serve. See RichardCraswell,Do Trade
LAW118, 138-42 (Jody S. Kraus& Steven D. Walt JURISPRUDENTIAL eds., 2000) [hereinafter Thissuggeststhatcustomevidenceis costlyto admitandto contest. FOUNDATIONS].
Customs Exist?, in THE JURISPRUDENTIAL FOUNDATIONS OF CORPORATE AND COMMERCIAL
582
[Vol. 113: 541
a slightly higher level of quality than the correctly interpretedcontract would have required.It is this "knife-edge"propertyof the perfect tender rule that reduces the seller's expected payoff below the price. Parties can respond to the need to increase the seller's expected payoff, however, by contracting out of the rule. The common way to do this is to use the clause customary warrantyterm, which contains a repair-and-replacement that eliminates the buyer's right to reject but requiresthe seller to repairor replace defective tenders or parts of tenders. When the buyer cannot reject, the seller is entitled to the price, with damages deducted from it.72The customary warrantythus eliminates the knife-edge feature of the perfect tenderrule, therebyensuringthatboth parties' expectationsare protectedso long as they believe that their payoffs will be determinedby the correct of the repair-and-replacement clause. interpretation This condition will be satisfied because the repair-and-replacement clause is analytically similar to the product-preparation term considered in Subsection IV.C.1. Under that term, the buyer's expected payoff, E[sb(i)], equaled the payoff it would receive underthe correct interpretation, Sb*. To see why the repair-and-replacement clause has the same property, recall that section 2-719 of the UCC permits a seller to limit the buyer's remedies to repair and replacement,73 while section 2-608(1) permits the buyer to his "revoke acceptance" only if a "non-conformitysubstantially impairs [the good's] value to him."74Section 2-719(2) authorizesa court to refuse to enforce a repair-and-replacement clause if "circumstancescause .. . [the] limited remedy to fail of its essential purpose . . ."75 Comment 1 to section 2-719 explains that this authorizationexists because "there [must] be at least a fair quantum of remedy for breach of the obligations or duties outlined in the contract."76 In light of these sections, a court's interpretivetask under a repair-andreplacement clause is not to decide, as per the perfect tender rule, if the goods "fail in any respect to conform to the contract."77 Instead, the initial for a court is whether the of success the seller achieved in question degree the repairor replacementtask was such as to give the buyer "a fair quantum of remedy."If not, the clause becomes inoperativeand the question shifts to whether the goods are "substantially impaired" in value. The buyer's clause thus increase as the possible payoffs under a repair-and-replacement court requires the seller to exceed the correct interpretation of the clause and decrease as the seller is permittedto fall short,just as the payoffs in the
72. 73. 74. 75. 76. 77. See U.C.C. ? 2-717. See id. ? 2-719. Id. ? 2-608(1). Id. ? 2-719(2). Id. ? 2-719 cmt. 1. Id. ? 2-601 (emphasisadded).
2003]
583
case. And since the seller'spayoffsherearea function product-preparation of the buyer'spayoffs,the seller'spossiblepayoffsalso are continuous in the sameway. Bothpartiesthuswill expectthe payoffundera repair-andclauseto equalin expectation the payoffthe partiesintended. replacement This suggeststhatpartieschoosinga repair-and-replacement warranty will wantcourtsto interpret the warranty using the minimum interpretive base above,it is ordinarily enoughfor Bmin.This is because,as we have argued businesspartiesthat courtsare correctin expectation, not that they are alwayscorrect.
3. Summary
The case in whichthe parties'payoffsarecontinuous in the spaceof a coversa lot of the ground. This is because court'spossibleinterpretations the case oftenarises"naturally," as in theproduct-preparation and example, can also arise"artificially," as when partiescontract to createcontinuous to increase the efficiencyof theirpayoffstructure. Firmsin payoffsin order the continuous-payoff case ordinarily prefercourtsto follow a textualist to the style.Thecase in whichthe parties' interpretive payoffsareinvariant error of is less common. Even in this case, degree judicial probably will cause a majority of partiesto however,litigationcost considerations prefer courts to use a narrow evidentiarybase. Therefore,the best defaultfor firms is textualistwhen the issue is what their interpretive contract meant. language We conclude this Sectionwitha comment concerning judicialpractice. Courtsmakingcommonlaw adjudications commonlytake a Willistonian while the UCC strongly urges a contextualistinterpretive approach,78 rulesas mandatory. Courtsin general,however,treatinterpretation style.79 are reluctant to invoke the coercivemachinery of the stateto require Judges a party to performa contract(or to pay damages)unless the judge is satisfiedthatthe contract what the partyfailed to do. It actuallydirected seeminglyfollows that courts,not parties,should choose the rules that determinehow contracts are read. This view is understandable but The law in in make choices and firms to misguided. general permits persons For litigationthat may lead courtsto act on less than full information. example,persons and firms may waive the right to counsel, agree to
78. See PeterLinzer,TheComfort Plain Meaningand the Parol EvidenceRule, of Certainty: 71 FORDHAM L. REV.799, 805-06 (2002); RobertE. Scott, The Uniformity Normin Commercial Law and Code Methodologies, Law: A Comparative in JURISPRUDENTIAL Analysisof Common FOUNDATIONS, supranote 71, at 149, 167-69. 79. See ? 1-205 cmt. 1 ("ThisAct rejectsboth the 'lay-dictionary' and the 'conveyancer's' readingof a commercial Insteadthe meaningof the agreement of the partiesis to be agreement. determined by the languageused by them and by theiraction,readand interpreted in the light of commercial andothersurrounding practices circumstances.").
584
[Vol. 113: 541
procedureswhose stipulated findings of fact, and use summaryarbitration results courts are requiredto enforce. In sum, the law generally sacrifices accuracy in adjudication to parties' self-interested choices. Similarly, parties should be permitted to realize the cost savings from contract on minimal evidentiarybases even if, in any given case, the interpretations odds of an accurateinterpretation would be higher with a broaderbase.80 D. Private Languages, LinguisticDefaults, and the Parol Evidence Rule 1. ThePreferredLinguisticDefault We now consider the case in which the partieshave written a complete contract in some language. The issue is whether, if the contractis silent on the matter,a court should take the parties to have written in majoritytalk. The alternative judicial assumptionwould hold that, in case of a dispute,the parties prefer to have the opportunityto introduce extrinsic evidence that relevant parts of the contractwere written in the parties' private language. We next set out four reasons why the "linguistic default"should hold that the contractwas written in majoritytalk. The practical implication of this proposal is that, when a contract does not speak to the issue, the court should not go beyond the evidentiarybase Bminwhen attemptingto identify the languageof the contract.81 Many partieswould preferthe linguistic defaultwe propose because the default would (1) reduce contractingcosts, (2) minimize the opportunities for strategic behavior, (3) reduce the risk of judicial error,and (4) expand the set of efficient contractsthat parties could write. Let us begin with the issue of contracting costs. The comments to section 2-202 of the UCC
80. If partiespreferred courts to use evidentiarybases that were so broadas to foreclose judicial time for deciding other categories of cases, then courts should overridethe parties' would seem unlikely, Sucha dangerof conforming to parties'interpretive preference. preferences to be brieferthan however,given that modem-daypartiesgenerallyseem to want adjudications effects fromjudicial they now are. As a result,so long as thereare no othernegativethird-party deferenceto the partiesregarding to parties'interpretive preferences interpretive styles, deferring would be efficientin the majority of modemcontract cases. 81. The positionthat a partyalways shouldbe able to show that its contract was writtenin explainedin partytalk sometimesis justifiedon autonomygrounds.Thus,Chief JusticeTraynor Pacific Gas & Electric Co. v. G.W. ThomasDrayage & Rigging Co.: "In this state,... the intentionof the partiesas expressedin the contract is the sourceof contractual rightsandduties.A courtmustascertain andgive effect to this intention whatthe partiesmeantby the by determining words they used."442 P.2d 641, 644 (Cal. 1968). This argument is a non sequituras stated.No in a normatively one wouldclaimthatthe Statute serious of Frauds withpartyautonomy interferes in writtenform.See way, thoughthe Statutesometimesrequirespartiesto cast theiragreements U.C.C. ? 2-201. It would not be a serious interference with autonomyfor courtsalso to assume thatpartiescast theiragreements in majority talkif good reasonsexist to createthe incentivesthat this assumption thannow exists to show thatan autonomyimplies.Thus,it takesmoreargument based view of interpretation wouldjustify the rule Chief JusticeTraynor stated,even if such a view were appropriately appliedto firms.
2003]
585
linguisticdefault.Comment2 thus recites: adopt a strongcontextualist are to be read on the assumption that the course of prior "[W]ritings .and the usages of trade were taken for granted[thatis, were dealings.., meant to be aids in interpretationwhen the document was
phrased].... [u]nless carefully negated . .. ."82 But if the default were
out of a plain-meaning reversed, partiescouldcontract linguisticdefaultat the sameor lowercost. Foran example,consider: "Thisagreement is to be read in light of the customsof the widget trade."Partiesseldom would describe the actualcustomsin the agreement becausecontracting costs are incurred while disputeresolutioncosts are incurred today with certainty tomorrow andprobabilistically. will be cheaper Thus,it ordinarily just to tell courts to considercustom should an irreconcilable differencelater In addition,a minorityof contracts are writtenlargelyin private arise.83 enforcethem with languages,and the partiesthat write them frequently not fewer expert arbitrators, lay judges. Thus, parties would have to contract out of a defaultthatsupposed themto be writingin majority talk thanwouldhaveto contract out of the contrary default.84 A plain-meaning linguisticdefault-that is, a defaultthatrestricts the courtto the interpretive base Bmin-also would reducestrategic behavior. To see why, consider a contract betweenpartyA andpartyB, the relevant which of were written in partytalk.Thiscontract, turns provisions suppose, out badly for B. Therefore, B would like to raise an interpretation issue strategically, claimingthat the contractwas writtenin majoritytalk, in
82. ? 2-202 cmt. 2 (emphasis added). 83. Some scholarsarguethat if courts are Willistonian,and thus implicitlyadoptmajority talk as the linguisticdefault,partieswill incuradditionaldraftingcosts translating theirprivate languageintothe majority clearto judicialinterpreters. languagein orderto maketheirintentions See Jody S. Kraus & Steven D. Walt, In Defense of the Incorporation Strategy, in
partiescan only opt out of the plain-meaning But as the linguisticdefaultwith costly translations. discussionin the text has shown,partiescan cheaplyopt into a privatelanguageby agreeingin theircontractthat, should a disputearise, evidence shouldbe admitted that language. regarding even when contractscontaintechnicalpartytalk, most of theirwords will be writtenin Further, the majoritylanguage. For example, parties may attributea private meaning to the phrase "two-by-four" (wood supportsso describedin construction contractsare usually meantto state dimensions of 1%by 3%inches),butsuchpartiesseldomwoulduse a privatelanguageto describe the deliverydate, the place of delivery,the price, and so forth.These partieswould want words with trade-language but would not want words meaningsto be read with tradeunderstanding, in the majority written languageto be readas if theywere special.Thus,if the linguisticdefaultis partytalk, then partieswho wish to exclude party-talk of majoritytalk would be interpretations to identifyall of the majority required termsandexplicitlynegatethe use of extrinsicevidencefor them. On the otherhand,by opting into only the technical interpreting partytalk they wish to incorporate, partiescan more readilyunbundlethe two types of language---e.g.,"All measures andspecifications in this contract areto be interpreted in light of the customsof the construction industry." Hence, if courts are Willistonian when partieswant them to be, partieswill not just incurunnecessary writingcosts. 84. The current rules are mandatory interpretive For the purposeof (or "quasi-mandatory"). our discussion here, which focuses on party preferences,it is clarifyingto treat the rules as defaults.
JURISPRUDENTIAL FOUNDATIONS, supra note 71, at 193, 197-200. This argument assumes that
586
TheYale LawJournal
[Vol. 113:541
to improve B couldraisesuchan issueonly if order B's bargaining position. B couldplausiblyshow thatwordsin the parties'privatelanguagehad a B would clearbut different linguisticcommunity. meaningin the majority B also relieve would raisesuchan issue only if the different plainmeaning of a the effect of performance. Thesetwo conditions for avoiding private are there to difficult while are because private many satisfy language would it one is there only be only majoritylanguage.Hence, languages to have a clearbut for wordsin a particular coincidence privatelanguage a litigantsuch favored in the that also different majority meaning language as B. Now assume that the contractis written in majoritytalk. The B moreeasily of possibleprivate wouldpermit party languages multiplicity to claim thatstated is A move common to asserta helpfulprivate meaning. in the are only "estimates" or "projections" private prices or quantities the partiesused.85 The linguisticdefaultwe favorwouldreduce language the likelihoodthat this move would succeed,for a partywould have to decide before a disputearose to write in a privatelanguage,and then in andinterpreted to its contract be written thatthe contract partner propose Thepartner thatlanguage. wouldnot agreeunlesssucha language actually existed and had meaningsthatwere accessibleto courts.Thus,requiring would largely partiesex anteto say they are writingin a privatelanguage to rescueitself from a ameliorate the concernthat a partywould attempt bad deal by claimingthat its contract was writtenin a mythicalprivate language. in lightof Thetwo linguistic defaultsat issue cannow be reconsidered this analysis.If a courtalways permitted partiesto offer proof that they wrotein a privatelanguage, thena disappointed partywouldhavea strong to a majority-talk a fictional incentive to attribute favorable meaning private talkunless A defaultthatsupposed contract. partiesto be writingmajority Andthen theircontract recitedotherwise wouldprevent suchmanipulation. out of the andcontracted if partiesactuallydid writein a privatelanguage Willistonian default,it wouldbe difficultfor one of themlaterto behave because,as we havejust shown,the regarding interpretation strategically The are so stringent. conditionsfor avoidingparticular privatelanguages thus talk default that be to writingmajority linguistic supposesparties would reduce strategicbehaviormore effectively than the rules in the to today'sUCC. comments
85. In one case, a contract the buyerto take "'approximately 70,000 cubic yards'of required will be in this contract andalso recitedthat"'[n]oconditions concrete" whicharenot incorporated S. ConcreteServs.,Inc. v. Mableton Inc., 407 F. Supp.581, 582 (N.D. Contractors, recognized."' Id. In the litigation,his Ga. 1975). The buyertook a little over 12,500 yardsin a falling market. claim that parties in the trade understoodexplicitly specified quantitiesto be estimates was rejected.See id. at 585-86. On the other hand, in ColumbiaNitrogen Corp. v. Royster Co., 451 F.2d 3 (4th Cir. 1971), course-of-dealing and usage-of-tradeevidence was admittedto thatexpresspriceandquantity demonstrate termswere only "fairestimate[s]." Id. at 7 n.3.
2003]
587
A plain-meaninglinguistic default that assumes parties to be writing in majority talk also would reduce the risk of judicial error. There is one majority linguistic community but many possible private communities. Hence, when private languages are permissible, as they often are today, a court has two interpretivetasks: to ascertain the parties' language and to ascertainwhat the parties said in that language. The dangerthat a court will pick the wrong language is real because a private linguistic community can be as small as the particularparties to the contract.Hence, 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,"86 that agreeing to take a "minimum" quantitymeant that the buyer could take When much less,87or that an unconditionaloption was really conditional.88 such a claim is false but found to be true, the court necessarily will the contract. misinterpret To see why, recall our argumentin Section IV.C that firms are content to have courts be right on average, not right every time. The errorhere is different from the error described there. For example, assume a contract uses the word "red,"and a disappointedpartypersuadesthe court,wrongly, that the contractwas written in a private language in which the word "red" meant "green."Both red and green are vague. In this example, the space of would center around some instance of the possible judicial interpretations concept "green,"but the court here could not be right on average. It would be attemptingto find the correct shade of green while the parties, ex ante, wanted a court to find the correct shade of red. When courts are mistaken regarding the contract's language, their constructions must be inefficient because it is only efficient to protect a party's expectationinterestunderthe contract it actually wrote. Hence, parties face a heightened risk of inefficient interpretationswhen courts always entertain claims that a contractwas writtenin a privatelanguage.Typical partiesthus would prefer courtsto assume that they wrote in majoritytalk. Finally, contextualist interpretation in general, and especially contextual interpretationthat permits the use of private languages, can truncatethe set of efficient contracts that parties will write. We have just argued that contextualist interpretation can create moral hazard. It is plausibleto believe that the more complex the contract,the easier it will be to createdisputesregardingwhat the contractsays and what languageit was written in. In addition, litigation is more costly in a contextualist interpretive regime because the partiesmore frequentlywill have full trials. The fact that contextualist regimes increase the likelihood and cost of
86. See Pac. Gas & Elec., 442 P.2d 641. 87. See Columbia NitrogenCorp.,451 F.2d 3. 88. See HuntFoods & Indus.,Inc. v. Doliner,270 N.Y.S.2d 937, aff'd, 272 N.Y.S.2d 686 (App.Div. 1966).
588
[Vol. 113: 541
disputes creates an incentive for parties to use simpler, but possibly less efficient, contracts.89 An example may make this point clear. Assume thatpartiescan create a complex deal that will generatea net expected surplusof $15 at a cost of $5 to write the contract. The parties also can create a simpler deal that will cost of $1. The generate a net expected surplus of $8 at a contract-writing the write should net the contract because parties complex expected social surplus of $10 exceeds the expected social surplus of $7 that the simple contractwould create. Suppose, however, that there is a .3 likelihood of an interpretivedispute when parties use a complex contractin a contextualist interpretiveregime and no likelihood of a disputewhen partiesuse a simple contract(formally, one needs only a higher likelihood of a dispute when the contract is complex). If there is an interpretivedispute, parties, as in the continuous-payoff case described above, expect the court to interpretthe contract correctly. The correct interpretationwould permit the parties to realize (and share)the surplusthat the contractwas writtento create. In this illustration, the parties will use the simple contract if the cost of a trial would exceed $10.90Since trial costs are said often to approachthe amount at stake in disputes, it is not fanciful to suggest that a contextualist interpretiveregime will drive parties toward simple contracts.To complete the example, suppose that the cost of a trial would be $12, so that parties actually would use the simple contract.Now suppose that the state switches to a textualist interpretiveregime. For convenience, let the probabilityof a dispute be the same in both regimes, but now assume that the typical interpretivedispute will be resolved by summaryjudgment at one-half the cost of a full trial. Then parties would use the complex contract and maximize social surplus.91 When parties consider what type of contract to write-what type of deal to create-they will not only considerthe costs and gains from creating the deal initially; they will also consider the likelihood and costs of later disputes. In the example here, when the cost of disputes is added to the calculation, parties in a contextualistinterpretiveregime will use a simple but less efficient contractratherthan a complex but more efficient contract. To be sure, an example cannot show that contextualistinterpretiveregimes
89. A formaltreatment of the relationbetweencontracting and enforcement costs and the form will appearin Alan Schwartz& Joel Watson,TheLaw and parties'choice of contractual Economicsof CostlyContracting, 20 J.L.ECON. & ORG. (forthcoming Apr.2004). 90. The expected return from a complex contract given the likelihood of dispute is 7($15) + .3($15 - t) - $5, wherethe first termis the expectedgain when thereis no dispute,the second termis the expectedgain froma trial(t is the trialcost), and the thirdtermis the cost of creatingthe deal. The net expectedreturnfrom the simple contractis $7. This will exceed the return fromthe complexcontract when t exceeds$10. 91. In this example, when t = $12, V2 t equals $6, and the expectedreturnfrom using the rises to $8.20, whichexceedsthegainof $7 fromthe simplecontract. complexcontract
2003]
589
always induce partiesto use less efficient contracts.The lesson ratheris that an overlooked cost of these regimes is that they sometimes buy greater accuracyat a cost of less efficient contracting.This disadvantagereinforces our view that business parties generally prefer a plain-meaning linguistic default,which implies the use of a Willistonianinterpretivestyle. The analysis here permits us further to clarify the debate between contextualistsand textualists and also suggests a new understandingof the function of merger clauses. Contextualistsclaim that interpretations made on broadevidentiarybases are more likely to be correctthan interpretations made on narrowbases. Textualists dispute this claim. The dispute between these camps, we have shown above, is irrelevantwhen the issue is what the contract says. Business firms are content with interpretationsof their language that are correct on average, not always correct, and so prefer narrowevidentiarybases to broadones. The interpretation dispute mattersa when the issue concerns the language in great deal, however, interpretive which the partieswrote. In such a case, we have argued,a broadevidentiary base affords a disappointedpartythe opportunityto raise the language issue strategically.Broad evidentiarybases also increase the risk of judicial error and truncatethe set of efficient contracts parties can write. Our analysis thus supportsthe conclusion that courts should interpretbusiness contracts on minimal evidentiary bases whether the issue is what the contract languagemeans or what language the contractwas writtenin, unless parties explicitly instructthe court otherwise.92 Mergerclauses are understoodto restrictthe evidentiarybase available to courts when making interpretations.93 Because courts today often search for the correct clauses are difficult for parties to broadly answer, merger enforce.94 These clauses now can be seen to have an additionalfunction. A
92. This conclusionmay be thoughtto raisea regressproblem. If partiescan give interpretive thenthose instructions instructions, will themselveshave to be interpreted, as would instructions how to interpret the instructions and so forth. This problemdoes not appearto be regarding serious.The defaultwe advocatewouldrequire to tell courtsto widenthe evidentiary base. parties Forexample,the contract wouldrecite:"Use custom." Thereseems littlereasonfor partiesto say "Use custom sympathetically." Courtsseldom would need instructions as to how to interpret thatreferthe courtsto evidentiary simpledirectives now in use. categories 93. A mergerclauserecitesthatthe writtenagreement is the parties'final expressionof their A commonexamplestates:"Thiscontract intentions. containsthe finalunderstanding betweenthe partiesand representsthe final agreementon all terms. There are no verbal agreementsor in connection representations therewith. The writingis a mergerof all proposals, and negotiations, with referenceto the subject matterand provisions."For examples of similar representations clauses,see Luther Williams,Jr.,Inc. v. Johnson,229 A.2d 163, 165 (D.C. 1967);and UAW-GM Human Res. Ctr.v. KSLRecreation Corp.,579 N.W.2d411, 412 (Mich.Ct. App. 1998). 94. See Franklin v. White,493 N.E.2d 161, 166 (Ind. 1986) ("An integration clause is only some evidence of the parties'intentions.The trial court should consideran clause integration along with all otherrelevantevidenceon the questionof integration."); Suttonv. Stacey's Food Mart,Inc.,431 A.2d 1319, 1322 n.3 (Me. 1981) ("A clausedoes not controlthe question of whether a writing was intended to be a merger completely integrated agreement." (citing RESTATEMENT(SECOND) OF CONTRACTS ? 242 cmt. e (Tentative Draft Nos. 1-7, 1973))); see also RESTATEMENT (SECOND) OFCONTRACTS ? 209 cmt. b (1981) ("Written contracts... may
590
[Vol. 113: 541
merger clause, if honored, would limit the court to the evidentiary base Bmin. Parties aware that the base was so limited would have a strong incentive to write the contract in majority talk. By adopting the merger clause, therefore, the parties signal to the court that this incentive has motivated them to speak in majority talk. Thus, a court, even if not persuadedby our more general argumentthat majority talk should be the default language, should still interpretthe inclusion of a merger clause to mean thatthe partieswrote in majoritytalk.95 2. TheParol EvidenceRule A typical statementof the parol evidence rule provides that when terms are "set forth in a writing intended by the parties as a final expression of their agreement,"the terms "may not be contradictedby evidence of any prior agreement or of a contemporaneous oral agreement but may be explained or supplementedby course of performance,course of dealing, or This rule comes in a "hard"and a "soft"version. Courts usage of trade."96 that adopt the hard version of the parol evidence rule decide whether a written contract is ambiguous from the document itself; that is, the court
includean explicit declaration thatthereare no otheragreements betweenthe parties,but such a declarationmay not be conclusive.").Professor Corbin is regardedas the most persuasive advocateof this position:"[I]tcan neverbe determined of the wordsof a by mere interpretation of anything, whetherit is the final andcompleteexpressionof writingwhetherit is an integration LINTON the agreementor is a mere partialexpressionof the agreement." 3 ARTHUR CORBIN, marksomitted). CONTRACTS ? 581, at 442 (1960) (internal quotation 95. We arguethatcourtsshouldadopta Willistonian linguisticdefaultbecausepartiesprefer it and because there are "external" efficiency reasons to justify the default. First, the state subsidizesthe judicial system and thus has an independent reason to reducethe likelihoodof disputes. For the reasons given above, our default would reduce this likelihood furtherthan current law. Second, if contracts are written in majority talk, courts can create standard in which commercial vocabularies Whena phrasehas a set, easily transactions can be conducted. discoverable meaning,partieswho use it will know what the phraserequiresof them and what courts will say the phraserequires.Courtsthat insulatethe meaning of terms in the majority the same words in the same ways across languagefrom deviantinterpretations by interpreting cases thuscreatea collectivegood: a set of termswith meaningsthatare alreadyunderstood by a largemajorityof potentialcontracting parties.It follows that courtsshould encouragepartiesto use majoritytalk. See Alan Schwartz,ContractTheoryand Theoriesof ContractRegulation, 92 REVUE D'ECONOMIE INDUSTRIELLE 101, 102-03 (2000); Robert E. Scott, The Case for Formalism in RelationalContract, 94 NW.U. L. REV. 847, 853-56 (2000); Scott,supranote 78, at 157. Anotherway to put this second point is that similarcontracting problemsrecurin varying contexts. Efficiencyis enhancedif partiescan adoptpriorsolutions,and adoptionis facilitated when partiesknow that if they use the languagein which a priorsolutionwas cast, they will be takenby courtsand contract to have adoptedthatsolution.HenrySmithhas shown that partners courts are more hospitable to acontextualinterpretation when they recognize that contract for thirdparties,as withdeedsor third-party contracts. See Henry languageis intended beneficiary E. Smith, The Languageof Property.Form, Context,and Audience,55 STAN.L. REV. 1105, 1177-90 (2003). Courtsshould recognizethat the set of thirdpartieswho would benefit from acontextual is widerthanis conventionally believed.This set includespartieswho interpretation couldprofitably use the contracting solutionsthatpriorpartieshaddeveloped. 96. U.C.C.? 2-202 (2003).
2003]
591
makesan interpretation on the evidentiary base Bmin.Courtsthatadoptthe soft versionhearall extrinsicevidencebeforedecidingwhether thereis an The UCC is taken to have adopted the soft version, and a ambiguity.97 numberof commonlaw courtshave adoptedthis version as well.98As shouldbe apparent, this versionof the parolevidencerule is justifiableif courts should considerall evidence that may bear on what the parties meant;it is notjustifiableif courtsshouldconsideronly the evidencethat ex ante,wantcourtsto see. parties, to the conventional however,the text of the understanding, Contrary UCC's versionof the parol evidence rule actuallyadoptsthe linguistic defaultfor whichwe argue.Section2-202 provides thatextrinsic evidence can "explain or supplement" a writingbut cannot"contradict" the writing. This language, when given its obviousmeaning,createsa strongincentive forparties to writein majority talk.To see why, supposethata contractual has the in majority talkandthe equallyplainbut plainmeaning phrase mM differentmeaningmp in party talk. If a party can introduceextrinsic evidenceexplainingthat the contractactuallyhad the meaningmp, then extrinsicevidence never could contradict the writing.The meaningmp would not contradictthe meaning mM,because the writing, properly nevermeantmM. Put morevividly, if a partycould introduce understood, extrinsic evidenceto showthatthe partiesmeant"green" whenthey wrote their "red,"extrinsicevidence could "explain"but never "contradict" A contradiction contract. were assumedto could ariseonly if the contract be written in majority talkandthe extrinsic evidencewas offeredto showa meaningin that languageother than mM.Therefore,unless the word is to be readout of the statute, "contradict" section2-202 mustbe takento that contracts are writtenin majoritytalk, but to permitthe presuppose introduction of extrinsic evidenceto clarifyambiguities in this language.99
97. See EricPosner,TheParol EvidenceRule,the Plain MeaningRule,and thePrinciplesof Contractual 146 U. PA.L. REV.533, 534 (1998);Whitford, Interpretation, supranote 70, at 939. 98. See, e.g., Ben-Shahar, supranote 52, at 790; Snyder, supranote 52, at 624-25. 99. This proposedinterpretation of section 2-202 actuallywas made in SouthernConcrete Services,Inc. v. Mableton there Contractors, Inc., 407 F. Supp.581 (N.D. Ga. 1975).The contract the defendant required to take "'approximately Id. at 582. The 70,000 cubic yards'of concrete." defendant purchased12,542 cubic yards. It defendedthe subsequent lawsuitwith the claim that tradecustomand supplementary between the partieswould show "thatthe quantity agreements in the contract stipulated was not mandatory... andthatbothquantity andpricewere understood to be subjectto renegotiation." Id. The court excludedthe evidencebecause it believed that an explicitquantity wouldbe contradicted was thatthe requirement requirement by an understanding not mandatory. It explained:"To admit evidence of an agreementwhich would contradictthe expresstermsof the contract wouldclearlyevisceratethe purposeof ? 2-202."Id. at 585.
592
The Yale Law Journal 3. Course-of-Performance Evidence
[Vol. 113: 541
The UCC explicitly invites courts to consider the parties' course of performance under a contract.'l0 Such evidence is said to be "always relevant" to the contract's meaning.'0' If the state is to provide the interpretive theory that the parties want, however, then supplementing contracts with course-of-performanceevidence would frequently be a mistake. The parties' course of performanceunder a contract differs from evidence in the other evidentiary categories because it can be offered not only to show what the parties originally meant, but also to prove that the parties' meaning has changed. Thus, evidence that the buyer accepted shipmentsat quality levels below those specified in the contractmay show that the parties modified the contract's quality requirement. Admitting evidence to prove a change in meaning is consistent course-of-performance with the traditionalcontract law rule that the parties' agreement may be inferredfrom acts or silence. 02 Nevertheless, courts should be reluctantto admit act-or-acquiescenceevidence to show a change in the meaning of a written contract.The existence of a writing indicates that the parties once believed that the gains from writing things down exceeded the costs. In the absence of evidence that this calculus has changed, the best inference to be drawn is that parties want modifications to be written as well. If this inference is not compelling, it should become so when the contractcontains a term requiringmodificationsto be in writing. Courts,however, accepting the UCC's invitation,often hold that conduct can effect a waiver of the "no oral modification"clause.103 The Code and the courts' use of course-of-performanceevidence to establish a change in meaning reflects a misunderstandingof the parties' likely intentions. To see why, consider the quality-level illustrationin the preceding paragraph.Suppose that the contract price was $2000 per lot
100. ? 2-202(a). 101. U.C.C. ? 2-208 cmt. 2 (1989) (providingthat "a course of performanceis always relevantto determinethe meaning of the agreement"). Revised Article 1, approvedin 2001, in new section substantially preservesthe section 2-208 definitionof "courseof performance" 1-303(a)but deletes the commentsto formersection 2-208. See U.C.C. ?? 1-303, 2-208 (2003). To date,revisedArticle 1 has been adoptedonly in a few states.Section2-208 and its comments thusremainas partof the enactedUCCin the largemajority of jurisdictions. 102. See, e.g., U.C.C. ? 2-208 cmt. 1 (1989) ("Thepartiesthemselvesknow best whatthey have meant by their words of agreementand their action under that agreementis the best indicationof what that meaningwas.").On the currentlegal statusof section 2-208, see supra note 101. or rescissiondoes 103. See U.C.C. ? 2-209(4) (2003) ("Although an attemptat modification not satisfy the requirements of subsection (2) [excludingmodificationsexcept by a signed (4) is intended, writing].. . it can operateas a waiver.").Comment4 explainsthat"[s]ubsection despite the provisions of subsection[] (2). . ., to prevent contractualprovisions excluding modificationexcept by a signed writing from limitingin other respectsthe legal effect of the Id. ? 2-209 cmt.4. parties'actuallaterconduct."
2003]
593
delivered and the contract contained a no-oral-modification clause in additionto the quality specification. The parties expect that deviations from the specified contract quality will sometimes occur. For the buyer, the expected loss from the average deviation (the probability of a deviation times the cost) is $100. Thus, sometimes a deviation will create a cost that approacheszero (perhapsthe buyer has forgiving customers or the quality shortfallcan be quickly corrected).At other times, a deviation can impose a loss whose expected value equals or exceeds $500 (say the buyer has a new, potentially large customer for whom quality is important,or the particular deviation would be slow to correct in a high-demandperiod). The parties also know that it would not be cost-justified to litigate against the average quality shortfall(the litigationcost would be, say, $150 per deviation). 04 The contract in this example has three salient features: (1) The price does not fall with declines in the level of quality supplied, (2) there is a writtenspecification of the quality the seller is to deliver, and (3) there is a no-oral-modificationclause. These contractual features together with the cost of deviations supporttwo conclusions with respect to the interpretive relevance of course-of-performanceevidence. First, features (1) and (2) together with our cost assumptions imply that the parties do not expect to litigate the average quality deviation. Rather, the buyer will accept nonconformingdeliveries that cause average losses, with the cost of these deviations reflected in a lower fixed price. Feature (3)-the no-oralmodification clause-implies that the seller should not infer from a series of acceptancesof goods whose defects cause the buyerto incur losses in the neighborhood of the average (or less) that the buyer also will accept a nonconforming delivery that would cause it to incur a large loss. The second conclusion that the example supports, therefore, is that high-cost deviationsin productquality are prohibited. This example captures an important feature of contracting behavior. When business parties incur costs to cast obligations in written form, they do so partly to permit a party to stand on its rights under the written contract when standing on its rights matters. Course-of-performance evidence thereforecommonly will be irrelevantto show what the contract originallymeant or what it currentlymeans. The parties' amicable behavior after the contract likely evidences only their view regarding how the average case should be treated.Courts, however, see the unusual case that
104. This examplewas stimulated assumesthat model, but he apparently by Ben-Shahar's deviations imposethe same loss in each periodrather thanthe same loss on average.Ben-Shahar, supra note 52, at 796-800. On his assumption,partiesare indifferentto whethercourts admit evidence to show a change in meaning. Ben-Shaharlater relaxes the course-of-performance same-lossassumption, however,and then concludes,as we do, thatthe soft versionof the parol evidencerulewill disadvantage typicalparties.
594
[Vol. 113: 541
the contract was written to govern.105 Courts thus should recognize that partiessometimes have multiple intentions.Their actions underthe contract will evidence their intentions for typical cases but seldom will evidence their intentions for the atypical case. Thus, a court is likely to make a category mistake when it relies on parties' behavior in nonlitigatedcases to infer how partieswant a litigatedcase to be treated.106
V. THE LEGAL DEFAULT PROJECT
In Part IV, we argued that sophisticated parties commonly prefer a default theory of interpretation that instructs courts to use narrow evidentiarybases and to presume that business contractsare written in the majority language. Courts can only interpretwhat is said, so our analysis assumed that the parties' writing was complete for the subjects at issue. Contracts,however, are often incomplete in relevant respects. Therefore, we now consider whether firms would prefer the state to complete these contracts with default legal terms. The somewhat surprising answer we derive from contract theory is that most state-created defaults will be useless or inefficient. Firms would prefer the state not to create inefficient defaults because firms will contractout of them; thus, the only effect these defaults will have is to increasetransactioncosts. Commercial parties commonly want to condition their obligations on the nature of their contractpartnersor on the state of the world that will materializeafter the contractis written. For example, a seller would like to condition its warrantyobligation on the buyer's intensity of use: The seller would charge more or warrantless for more intense users because these users are more likely to make claims. A buyer also would like to condition price or quantityon the natureof demandex post. If demandturnsout to be high, the buyer would be willing to pay more or to take more product,but the buyer would want a lower price or less productif demandturnedout to be low. These considerations imply that contracts will inevitably be
106. Arbitrators but courtsdo not. Our analysis obey the parties' interpretive instructions, thus identifiesa reasonforpartiesto use arbitration, but it cannotsupport the inferencethatparties use courtsbecausethey preferthe interpretive use. Thereare otherreasonsfor styles in current using courts,such as the ability to get discovery,to have appeals,to have the substantivelaw & and the like. See KennethS. Abraham appliedby expertsin it, to createeffective precedents, J.W. Montgomery, INS.L.J. 355 (2003); CharlesL. III, TheLawlessnessofArbitration,9 CONN. Private: TheQuietRevolution in Contract L. REV. Knapp,TakingContracts Law, 71 FORDHAM 761 (2002). The widespread use of mergerand no-oral-modification termsin contractsintended for courtsalso suggests,in line with our views, thatpartieswho preferjudicialenforcement also undera differentinterpretive preferenforcement style.
JURISPRUDENTIAL FOUNDATIONS, supra note 71, at 12, 16.
105. Our argumenthere is similar to Karl Llewellyn's view of custom. He believed that customsgovernordinary cases but seldom are relevantto the unusualcases that causepartiesto litigate. See Alan Schwartz, Karl Llewellyn and the Origins of Contract Theory, in
2003]
595
incomplete. There is an infinite numberof possible future states and a very large set of possible partnertypes. When the sum of possible states and partnertypes is infinite and contracting is costly, contracts must contain gaps. Partiescannotwrite contractsabout everything. Incomplete contracts sometimes produce lawsuits because parties will not always agree ex post regarding the treatment of omitted contingencies.l07Courts in such cases no longer can simply engage in because, by definition, the contracts lack words to interpret. interpretation The courts' task thus shifts to the developmentof rules to resolve gap cases. Hence, some default rules are judicially created.'"8Restatement and statutorydraftersalso create defaults when, in their view, certain gaps are likely to recur.Though our analysis has relevance for courts, we focus here principallyon the Restatementand the UCC, askingjust when the state can create good defaults for business parties. 09In particular,we derive criteria for efficient defaults, and then argue that these criteria are difficult for draftersto satisfy. A. The Casefor Defaults The conventional view is that, but for the cost, parties would write complete contracts. The state may increase efficiency in cases where contractingcosts preventparties from solving contractingproblems. To see how, consider the problem of developing a damages rule-a contractterm specifying the transfer,or the method of computingthe transfer,that a party in breach must pay to its contract partner. Any particularset of parties would bear the full cost of solving this problembut likely could not capture the full gain because it can be difficult to prevent laterparties from copying successful solutions that have evolved into contract terms."1 In this
107. Partiestend to omit low-probability states.For example,if very low demandis thought cost to specify theirobligationsin the very quite unlikely,partieswill not incurthe contracting low-demand state.A lawsuitcould ariseif this statematerializes andthe partiescannotagreeon a new price. 108. Since these rules are facilitative,courts should permit future parties to vary them. Judicial creation of such gap-filling defaults is not inevitable. A court instead can refuse enforcement when gaps cause a contract to be obligationally incomplete-that is, when the terms the contract does containprovidean insufficient basison whichto grounda remedy.The common law ruleis thatcontracts termsincomplete or indefinitearenot legallybinding.It leavingmaterial thatthe indefiniteness may be paradoxical rule is itself a default.Partiescan opt out of it by more in the contract. completelyspecifyingtheirobligations 109. The Restatement defaults differ from statutorydefaultsbecause a restatement is not A restatement self-executing. provisioncannotbecomea legal defaultuntilit is bothpromulgated Law Institute by the American (ALI)andthenadopted by a courtwhen resolvinga dispute.Thus, ouranalysishereappliesbothto the privatelegislatures thatpromulgate defaultsandto the courts thatadoptthem. innovations 110. Foranalysesof how contractual spread,see GeraldF. Davis,AgentsWithout Principles?TheSpreadof the Poison Pill Through the Intercorporate Network,36 ADMIN. SCI.
596
[Vol. 113: 541
circumstance, the total social gain from having a rule-a solution to a contractingproblem-may exceed the social cost, but parties themselves will not create the rule. There is a case, therefore, for a publicly supplied contract law that contains efficient solutions to common contracting problems.1" This conclusion implies that contractlaw rules should be defaults. This is because, in a large economy, parties are heterogeneous so that not everyone will want the same thing. The justification for a default rule is that it does for parties what they would have done for themselves had their contracting costs been lower. Hence, parties who can do for themselves-that is, who can create their own solutions-should be free to do so. While this conclusion may seem obvious, courts sometimes create barriersto contractingout of Restatementor UCC rules, on the groundthat these rules reflect either better or fairer solutions than those that parties develop.112 To the contrary,we argue that commercial parties themselves are the best judges of what constitutes a good contractingsolution, and that defaultrules should not be chosen on fairnessgrounds. Perhaps a more precise way to state the ground on which drafters should choose defaults is that a good default minimizes contractingcosts. Parties, if left free, will supplant or modify state-suppliedterms that they dislike. In a world of free contracting,unpopulardefaults thus will raise contracting costs but not otherwise affect parties' behavior. As a consequence, the state's power is limited to reducing contracting costs, which it does best by enacting popular defaults. This leaves open the question of what makes a default popular. This is a difficult question to answer in general but, as we have seen, firms prefer contract terms that maximize joint surplus. Defaults thus will be popular with firms if they maximizejoint surplusand unpopularotherwise.113
Q. 583 (1991); and Gerald F. Davis & Henrich R. Greve, CorporateElite Networksand Governance Changesin the 1980s, 103 AM.J. SOC.1 (1997). 111. See CharlesJ. Goetz& RobertE. Scott, TheLimitsof Expanded Choice:An Analysisof the InteractionsBetweenExpressand Implied ContractTerms,73 CAL.L. REV.261, 291-93 (1985). 112. See id. at 284-85. The courts' behavior explains the intensive lobbying by firms the proposedrevisionsto Article2, especiallythe warranty terms.See RobertE. Scott, regarding TheRise and Fall of Article 2, 62 LA. L. REV.1009, 1049-53 (2002). Firmsknow thatjudicial actioncan makea UCCdefaultstickyin practice. 113. RusselKorobkin has arguedthatthe "endowment effect"makesdefaultshardto change. Individual manifestthe endowment effect when they demandmuchmore subjectsin experiments to sell an item-for example, a coffee mug-than they would bid to purchasethe item. Put another are in parta functionof the legal assignment of property way, persons'preferences rights; of that assignment,as the Coase Theoremsupposes.In the realm of they are not independent contractlaw, a partysubjectto the endowmenteffect would ask much moreto give up a default thatis favorable to it thanthe partywould bid to obtainfrom its contract a clause thatis partner identicalto the default.As a consequence, for suchpartiesdefaultsaremorelike mandatory rules; the endowmenteffect would preventpartiesfrom convenientlyalteringthe allocationsthat the defaultscreate. Korobkinconductedexperimentscasting first-yearlaw studentsin the role of
2003]
597
This reasoning shows the error of choosing defaults just because they are fair. Partieshave the incentive (and often the ability) to contractout of even fair defaults that do not maximize surplus."14As a good example of this response, commentatorsand courts once thought (and may still think) that it is fair for merchant sellers to make implied warrantiesof quality. Merchant sellers, however, routinely attempt to disclaim the implied warranty of merchantability in section 2-314 of the UCC, and their merchant buyers routinely consent."5 Thus, section 2-314 in business contexts often does nothing more than increase transactioncosts. Drafters and courts thereforeshould ask what parties would want, not what parties
andfoundthatthey advisedclients in a manner thatmanifested endowment effects.See attorneys, Russell Korobkin,The Status Quo Bias and ContractDefaultRules, 83 CORNELL L. REV.608 is RussellKorobkin,TheEndowment Effectand Legal Analysis, (1998). A morerecenttreatment 97 Nw. U. L. REV.1227(2003). are not relevantto the types of transactions we consider. ProfessorKorobkin's experiments the endowmenteffect to the existenceof a real optionon the subjectof Recenttheoryattributes sale. A party uncertainabout the value of a good to her can delay a purchaseor sale while To buy or to sell thusexercisesthe optionbecauseinformation is no moreinformation. gathering As a consequence, the seller'sprice is the sum of the good's longerrelevantafterthe transaction. cost andthe (positive)value of the optionto delay sale; the buyer'sbid is the sum of the value of the good to him andthe (negative)valueof the optionto delay purchase. Addingoptionvalues to them froma bid priceproducesdifferencesbetweenthemthatareof an ask priceand subtracting as are manifestedin the psychologicalexperiments. See Massimo the same orderof magnitude Paradiso & AntonellaTrisorio,TheEffectof Knowledgeon the DisparityBetweenHypothetical and Real WillingnessTo Pay, 33 APPLIED ECON.1359 (2001); Eric van Dijk & Daan van Buyingand Selling Exchange Goods: Loss Aversionand the Endowment Knippenberg, Effect, 17 J. ECON. PSYCHOL. 517 (1996); JinhuaZhao& Catherine L. King,A New Explanation for the LETTERS 293 (2001). An implication of these articlesis that the WTP/WTA Disparity,73 ECON. endowment effect shouldbe small or absentwhen sophisticated partieswho know whatthey are thusremarked: "Lossaversionplays littlerole in doing tradegoods for money.DanielKahneman routineeconomictransactions, in which a seller anda buyerexchangea good andmoney,both of which were held for that purpose."Daniel Kahneman, ReferencePoints, Anchors,Norms, and Consistentwith this view, sophisticated actors tradingin real marketshave been shown not to manifestendowment are independent of the initial locationof property effects; theirpreferences Market Eliminate rightsto traded objects.See JohnA. List, Does Market Anomalies?, Experience 118 Q.J. ECON. 41, 42-43 (2003) ("[M]arketexperiencematters:across all consumertypes, marketlike experienceand the endowmenteffect are inversely related.In addition,within the I group of subjectswho have intense tradingexperience(dealersand experiencednondealers), find that the endowmenteffect becomes negligible.").Our analysis assumes away endowment effects becausewe studytransactions who "trade betweenfirmswith "intense experience" trading goods for money." 114. If the statewants a particular rule actuallyto controlbehavior,it shouldmake the rule to createtheirown contracts, mandatory. but sometimesonly at Currently, partiesare permitted the cost of contracting out of state-created defaults.The costs of contracting out will exceed the gains for some firmsbut not others.As a consequence, a fairbut inefficientdefaultdisadvantages firms with relativelyhigh contracting costs, but cannotfully realizethe state's fairnessconcern becauseotherfirmswill createtheirown deals. 115. SectionV.C below arguesthatthe UCC's impliedwarranty termis inefficientbecauseit providesa seller with too little guidanceregarding the performance obligationand also creates moralhazard.
Mixed Feelings, 51 ORGANIZATIONAL BEHAV.& HUM. DECISION PROCESSES 296, 301 (1992).
598
[Vol. 113: 541
16 We should want." argue in the next Section that this is a difficult question to answer.
B. The Cost Concernand Default Rules There are three criteria for a good default rule: It must be conditioned on only a few possible states of the world, be relatively simple in form, and be efficient for a wide variety of contract parties. The first criterion is necessary because there can be an infinite numberof possible future states of the world to which a rule could apply, while the state's draftingcosts are finite. Thus, a publicly supplied rule that had to address many possible future states seldom would be cost-justifiedto create. The second criterion, of simplicity, is a function of institutionalcompetence. Contractlaw rules are createdby courts and drafters.Courtscannot conduct investigationsinto the efficiency propertiesof possible rules and rule combinations.Drafters also have limited resources. The American Law Institute (ALI), which creates restatements, and the National Conference of Commissioners on Uniform State Laws (NCCUSL), which together with the ALI created the UCC, do not hold legislative-style hearings on proposed rules, cannot hire neutraleconomic or industryexperts to help in rule creation,and generally The rely on the part-timelabor of law professors and private attorneys."'7 draftersthus cannot,and do not, write rules for business contexts thatmatch the complexity of, say, the tax code; rather, contract law necessarily is restricted to the consideration of problems whose solutions can be embodied in simple rules. The thirdcriterion,of efficiency, is perhapsthe most difficult to satisfy, because parties in large economies are heterogeneous.Default rules would be too expensive to create if efficient solutions were party-specific. Then there would need to be as many legal rules as there are sets of contracting
116. Scholarshave identifieda set of default rules that are termed"information-forcing" because the rules createan incentivefor a partyto disclose relevantinformation to its contract See Ayres& Gertner, rulesincludethoseproviding that partner. supranote 1. Information-forcing unless they are foreseeableand reasonably the rulescreate damagescannotbe recovered certain; an incentivefor a promiseewho fearsbreachto disclose in advanceto the promisor the loss the promisee would suffer. The promisorthen will be able to price more accuratelyand to take rulesfall withinthe analysisherebecause optimalprecautions againstbreach.Information-forcing parties can create them on their own. For example, sellers routinelypropose contractsthat disclaimliabilityfor consequential damages.This createsan incentivefor buyerswho may suffer these damagesto disclose them to the seller and then to purchaseinsurance by bargaining away the disclaimer.An information-forcing default thus will be popularif partieswould agree to disclose in the circumstances in which the rule applies.It has been shown thatthe foreseeability defaultmay not always be popularin this sense. See Johnston, supranote 10, at 626-38 (stating that parties facing a monopolistmay not want to disclose the profits they would earn from successfultransactions becausethe monopolist thenmay pricediscriminate againstthem). 117. See Alan Schwartz& RobertE. Scott, ThePolitical Economyof PrivateLegislatures, 143 U. PA.L. REV.595, 607-37 (1995).
2003]
599
parties. The task, then, is to find rules that would be efficient-surplusmaximizing-in a wide variety of contexts."' The difficulty of satisfying the three criteriaexplains why there are very few default rules"9 in contract and commercial law. We illustrate the stringency of the criteria by considering two variants of the investment example analyzed in Section III.B above. Assume first that, after the parties contract for the seller to produce the specialized product, but before the goods are produced,the seller's factory is destroyed in a fire. The seller no longer can produce the goods, but it can pay money damages.'20 The common law default rule nevertheless excuses the seller when an exogenous event makes the contemplated, specialized performance impossible to render.12 This excuse rule seems to satisfy the criteriafor an efficient default. First, the costs of creatingthe rule are low. There are only two possible futurestates of the world-either an exogenous event prevents performanceor it does not-and there is need for a rule in only one of them (when the seller cannot perform).The rule also is simple in form;the seller is excused or not, dependingon whetherthere was a fire or not. Finally, the rule likely is efficient for a wide variety of parties.This is because buyers in general are betterinsurersagainst lost valuationsof specialized investments than are sellers; buyers usually are better informed than sellers about the consequences of sellers' breach. Excusing the seller requires the buyer eitherto insureon the marketor to reveal its valuationto the seller.
118. This note puts the text's point formally;the text goes on to give examples.A market type is a two-elementset composedof a seller and a buyer.Denote a particular type as ti. There are ti ..., t,,types in the market, andthereare m sets of partiesof each type. Thus,m(ti)denotes the totalnumber of partiesof a particular type. Let c,, denotethe cost for each t, set of partiesto createa contract term,and let cs,be the cost to the stateof creatinga defaultrule thatis identical to thatterm.We supposecs5> c,pbecausestate-created rules requirelegislativeor administrative action.Denotethe benefitto a set of t, partiesfromhavinga particular contract term(or default rule)as b(t,).Partiesto a particular contract will not createthe termif ci > b(t). The stateshould createa defaultif the totalbenefitsto all t; partieswould exceed the state'scost of rule creation. Thisefficiencycondition can be writtenas
Csi< m(ti) b(ti).
(1996). 119. Recall that a rule specifies the relevantconduct in advance--e.g., no driving above fifty-fivemilesperhour. 120. The buyer'sexpectation interestwouldbe its valuationless the price,which was $15 in the example.
121. See, e.g., Taylor v. Caldwell, 122 Eng. Rep. 309 (Q.B. 1863); RESTATEMENT (SECOND) OFCONTRACTS ? 263 (1981).
Implications of Transaction Cost Economics, 152 J. INSTITUTIONAL & THEORETICAL ECON.287
This conditioncannotbe satisfiedwhen m(t) = 1 because c5j > and we have assumedthat cp,, becomesmoreefficientas rulecreation wouldbe inefficient > b(t,)).Staterulecreation private (cpi m(t,)increases(the numberof partieswho would benefit from a rule goes up), but the text next arguesthat for contractlaw rules, m(t,) is more likely to be close to one than close to a large number. A moreextensiveanalysisof the heterogeneity TheNormative pointis in Alan Schwartz,
600
[Vol. 113: 541
This example is deceptively simple, however. The set of possible causes of a seller's incapacityto performcan be large, ranging from fire or flood, to a temporary or permanent supply shortage, to a government regulation,and so on. A seller's ability to anticipateand to take precautions against the causes of incapacity thus will differ across sellers and among causes. It would be efficient to excuse the seller only when its inability to performresulted from causes that were difficult to anticipateand prevent; otherwise, sellers would use the excuse doctrinejust to escape deals that turned out badly. Thus, an excuse rule must resolve difficult causation issues. The seller also may be able to performin part or to performin full but only for some contractbuyers. There is a question how a seller should make the resulting allocation decisions. Moreover, some buyers may be more able than others to take precautionsor to insure efficiently. Drafting rules to cover all of these possible causes, effects, and parties would be quite costly. As a consequence, the excuse case today is regulated by a standard,not by a set of rules. Section 2-615(a) of the UCC provides that a seller's failure to perform "is not a breach.. . if performance as agreed has been made impracticable by the occurrence of a contingency the non-occurrenceof which was a basic assumptionon which the contractwas made."122Courts decide after the fact whether a performancewould have been "impracticable" and whether the parties had "basically assumed"that the allegedly excusing cause would not occur or would not materiallyaffect performanceif it did occur. Now consider a second illustration.Recall that the investmentexample implicitly assumed that the parties' contract specified what a specialized product would do, so that if the seller delivered something else, breach would be clear. A default rule might also specify the seller's quality obligation, but recall that the productwas specialized to the buyer's use. A legal rule that set out this particularseller's quality obligation thus would apply only to one case. It is inefficient to draft for one case. A particular seller's quality obligation, however, sometimes can be generalized to an industry.It may be, for example, that book publisherswould agree on what requirementsa good printing press must meet. Even so, if an Article 2 drafterwere to describe, for every industry, the quality that sellers must deliver to buyers, the drafting cost would likely exceed the social gain. Unsurprisingly, the law regulates the seller's quality obligation with standards.Under section 2-314(2), goods "to be merchantablemust be at least such as (a) pass without objectionin the trade . .. and (c) are fit for the ordinarypurposes for which such goods are used."'23 Courts after the fact
122. U.C.C. ? 2-615(a) (2003). Section 261 of the Restatement containsthe same rule in almostthe same language. RESTATEMENT OFCONTRACTS (SECOND) ? 261 (1981). 123. U.C.C.? 2-314(2).
2003]
601
decide whetherit was enough for a seller to satisfy subsection (a) or (c) ("at least such as"); whether the goods would pass without objection; what "ordinary"seller purposes are; and whether the goods "are fit" for such purposes. Contract and commercial law thus contain very few default rules, because the criteriafor good default rules are so difficult to satisfy. Rather, the Restatement(Second) of Contractsand Article 2 of the UCC primarily containstandards; these texts are replete with provisionsrequiringpartiesto behave "reasonably,""conscionably," "fairly," "in good faith," and the like.124 These codifications also fail to address important contracting problems. For example, parties to long-term contractsface the problem of keeping each party's gain under the contractabove the opportunitycost of performancein every period. This problem is best addressed with index clauses, which tie the current period price to current cost and demand conditions.Efficient indices are party-or industry-specific,however, and so are too costly for contract law to create. 25 Instead, the law ignores this long-term contracting problem in favor of letting parties solve it for themselves. In sum, the cost concern forces contract law commonly to ratherthanrules, when it regulatesat all. regulatewith standards C. TheMoralHazard Concernand Default Standards The project of creating publicly supplied default standardshas been unsuccessful for two reasons. First, firms often need specific guidance regarding the performance obligation. Second, standards create unacceptable moral hazard. Regarding the need for guidance, a seller commonly must know what quality level to produce. Telling the seller that its product should "at least" satisfy a buyer's "ordinary purposes" is generally unhelpful. In practice, sellers of complex products thus disclaim the implied warrantyof merchantability in favor of an express warrantythat describes important aspects of the seller's performance obligation. As anotherillustration,when an exogenous event induces the seller to consider breaching, both parties need to know whether breach would be legally permissible.An erroneousdecision could expose one or the other partyto a
124. See id. ?? 2-205 to -209, 2-302, 2-305 to -307, 2-309, 2-311 to -312, 2-314 to -315, 2-317, 2-503 to -504, 2-508, 2-511, 2-513, 2-602 to -610, 2-612, 2-614 to -615, 2-704 to -706, 2709 to -710, 2-712, 2-714 to -716, 2-718 to -719, 2-723. The UCC also adoptsstandards in order to avoidresolvingcontroversial issues.See Schwartz & Scott,supranote 117. Ouranalysishereis meantto show thateven if the ALI and NCCUSLwere bolder,the cost concernwould generate manystandards. 125. An index clause links the transaction price between the partiesto verifiableexternal indices, such as the ConsumerPrice Index (CPI), that correlatewith the economic conditions facingthe parties.The degreeof correlation is a functionof the type of party(some firms' costs will move more closely with the CPI than other firms' costs will, for example),so that index clausesvarywithparticular partiesor partytypes.
602
TheYaleLawJournal
[Vol. 113:541
to section damages judgment. Tellingthe sellerat thispoint,withreference is also is that breach is if 2-615, permissible performance "impractical" Andtellingthebuyerthatits return (i.e., a payment unhelpful. performance undersection 2-609 if of an installment of the price) can be suspended of littleuse. In is similarly wouldbe "commercially reasonable" suspension standards the in to this difficulty, parties commonly ignore response andinsecurity sections2-615 and2-609 in favorof writingforcemajeure clausesthat specify preciselyboth the exogenouscausesthatwill excuse the seller'sperformance andthe permissible responsesthatthe obligation take when insecure can about seller's the that Parties performance.126 buyer The statethus writedetailed rulesin theircontracts. needspecificguidance when it createsa standard thatpartiesroutinely resources wastes drafting reject. thai We introduce withan example. themoralhazard Suppose difficulty in the sel a set of contracts is obligationally but contracts some complete, lack a solutionto a contracting For example,the contracts state problem. thus contracts and The but not to do inflation. index quantities prices prices in are insufficiently will be The suboptimal some state-contingent. prices statesof theworld.A courtcouldenforcesucha contract b3 possiblefuture anc the contract a between the difference awarding disappointed promisee marketprices. The issue is whetherthe state should fill the gap with 4 in ligh thatcouldrecite:"A contract standard be unreasonable pricecannot should of the conditions The state when is due." be obtaining performance to default reluctant adoptthis standard becausethe case for a public ij thi fromdrafting absent.Contracting costscouldnothaveprevented parties standardfor themselves.It would have cost the parties little to havy the sellerto delivera "reasonable" required qualitywithina "reasonable' becam timebutto haveexcusedthe sellerfromdelivery if its performance standard default Drafters should reluctant to enact be "impractical." contracts. without firstaskingwhy the standards weremissingfromprivate of more Standards the likelihood areunpopular becausethey increase that are hazard. Partiessometimes haveincentives to takeactions privatel
126. Force majeureclauses apparently have a long pedigree.Consider,as an example,tt following clause in a contractbetween Michelangeloand the heirs of Pope Pius III requirir to producefifteenmarblestatuesfor the cathedral in Sienawithina specifiedtime: Michelangelo Item, whereas the said Michelangelo,by virtue of the said agreement,has bound himselfto causemarbleto be brought for the of Carrara to Florencefromthe mountains makingof the said statues;andwhereas,by reasonof the besiegingof the Pisanswithin the territoryof Pisa, war has once more broken out; and whereas the Florentine to divertthe courseof the River Arno in such wise that the Republicmay endeavour of the said marblefromthe mountains to the city of Florencemay of Carrara transport therefore be hindered; andwhereasalso the said Michelangelo may fall ill, which God period]shall be suspendedduringthe time avert.., .then... the said [new contractual of the saidhindrance. IL GIGANTE: AND ANTON 1492-1504,at 215GILL, DAVID, FLORENCE, MICHELANGELO, THE (2002).
2003]
603
optimal but publicly undesirable.For example, a buyer has an incentive to claim that a product is unsuitable for one of its "ordinary" purposes when the market price drops.'27 A seller has an incentive to claim that its when its costs rise;28 or to attempt performancehas become "impractical" to get a price increase by using our illustrative standard to claim that inflation has made the contract price "unreasonable."When a standard governs, the partywho wants to behave strategicallymust ask what a court will laterdo if the party is sued. The vaguer the legal standardand the more that is at stake, the more likely the party is to resolve doubts in its own favor. A party that resolves doubts in this way will attempt to maximize private gains at the expense of joint welfare maximization. Therefore, a standardis efficient only when the party on whom it confers discretion has the incentive to maximize joint returns in the course of maximizing its private gain.29 Few UCC standardspass this test.'" Consequently,parties contractaway from them. For example, the customaryproductwarrantyfor machines and other equipmentcontractsout of the quality standard(section 2-314), the cure standard(section 2-508), and the revocation-of-acceptance standard (section 2-608).13' The common force majeureclause contractsout of the excuse standard(section 2-615), and many contracts contractout of the consequential-damages standard(section 2-715), the warranty-damages standard (section 2-714), and the standards governing insecurity and repudiation (sections 2-609 to 2-610).132 This evidence strongly anticipatory that suggests typical partiesdislike contractlaw's standards.'33
127. See, e.g., T.W.Oil, Inc.v. Consol.EdisonCo., 443 N.E.2d932 (N.Y. 1982). 128. See, e.g., Aluminum Co. of Am. v. Essex Group,Inc.,499 F. Supp.53 (W.D. Pa. 1980). 129. Thistest sometimescan be met. For example,the standard containsa mergeragreement the buyerto exit withoutpenaltyif a "material adversechange"occurs in the permitting tennrm interimbetweensigningthe contractand concludingthe merger.This termdoes not specify the events that could constitutea materialadversechange because there can be many such events whose effects will vary with the ex post state of the world. A standard neverthelessis efficient herebecausethe buyerwill exit only if the seller's value has fallen materially. The threatof exit createsan incentivefor the sellerto takesynergy-increasing actionsthatlower the probability that its value to the buyerwill fall materially.See RonaldGilson & Alan Schwartz, Understanding MACs and MAEs (2003) (unpublishedmanuscript,on file with authors), available at This exampleis offeredto suggestthatsome http:7/islandia.law.yale.edu/ccl/papers/mergerl5.pdf. standards statutory the inferencethatpartieswill writea may be efficient,but it may also support standard in theircontracts whena standard wouldmaximizesurplus. Reiningin GoodFaith. 35 U.C. DAVIS L. REV.319 (2002) (analyzingin detailthe ways in which courts have used the good faith standard that regulatesdiscretionin outputand requirements contracts undersection 2-306(1) to undermine the controlson discretionthat the partieshad set forthemselves). 132. Before 1978, partiesroutinelycontracted out of section 2-609 with ipso facto clauses, which statedpreciselywhen a partycould cancel the in consequenceof its partner's agreement financial difficulty.The current Codeoutlawsthese clauseson the (erroneous) Bankruptcy ground thatthey permit solventpartiesto depletethe bankruptcy estate.See 11 U.S.C. ? 365(e)(1) (2000). For an
analysis, see Yeon-Koo Che & Alan Schwartz, Section 365, Mandatory Bankruptcy Rules 131. See ALANSCHWARTZ & ROBERT E. SCOTT, COMMERCIAL PRINCIPLES TRANSACTIONS: ANDPOLICIF-S 204-24 (2d ed. 1991). 130. Sec. e.g., Victor P. Goldberg, Discretion in Long-Term Open Quantity Contracts:
604
TheYale LawJournal
[Vol. 113:541
couldfill, thebest whencontracts havea gapthata standard Therefore, is "missing" for a decisionmaker inference to drawis not thatthe standard This is fromthese contracts, thatthe standard butrather has been rejected. couldhavedrafted thebest inference becauseof the ease withwhichparties Parties theirown standard andbecausemanystandards createmoralhazard. a standard-in favor of probablyrejectedan ex ante solution--drafting the written terms with when circumstances renegotiating changed, whethera new the disagreement pointsthatwould determine constituting with standards dealwas efficient.Andto summarize, ordinarily fillinggaps The statewastesresources in drafting will be inefficientfor threereasons: in contracting out; and when courtsare them;the partieswaste resources to use standards to actively policebargains, maycreatesets parties expected havepreferred to omit.134 of rulestheywouldotherwise on the costs of In sum, the projectof creatingdefaultrulesfounders forheterogeneous in that function rulecreation complexcommercial parties the creation Thesecosts essentially of all buta few environments. preclude founderson the defaultrules.35The projectof creatingdefaultstandards to do, and for specificguidance needof parties as to whattheyaresupposed on the tendency of standards to create moral hazard.These gloomy would hold even when the information conclusions thata rule or standard courts. the is to to be to accessible assumed known the and require parties
and Inefficient Continuance,15 J.L. ECON.& ORG. 441 (1999). A concrete example of Forest Products,Inc., contractingout occurs in NorthwestLumberSales, Inc. v. Continental 495 P.2d 744, 748-50 (Or. 1972). 133. Ian Ayres has suggestedthat standards may be good defaultsbecause precedentcan crystallize aroundthem, therebyprovidingparties with guidance.See Ian Ayres, Making a Contributions and Fischel, 59 U. CHI.L. REV.1391, Difference:TheContractual of Easterbrook 1415-16(1992). This argument will becomeoperative assumesthatlegal standards impliedterms in parties'contracts, to applythem.The premisegenerally therebygiving courtsthe opportunity fails becausepartiescontract out of the standards. 134. The last dangeris said to have occurred in Englandwhen the courtsadoptedan excuse standard. As English lawyer AndrewRogers, Q.C., wrote in his introduction to a book on the excuse doctrine: It is only in relativelyrecenttimes thatthe Englishcourtsrelented in theirdemand that the strict words of the contract be adhered to, notwithstanding changes in The circumstance. The resultingdoctrineof frustration has not workedsatisfactorily. courtsfoundit difficultto determine ... the limitsfor its application.... underthe generallaw bringsin its trainautomatic ... [F]rustration discharge.To avoid such results,the partiesto the contract to drafttheirown particular are required code. Thatin turnmeansa substantial costs. increasein transaction the costs involved, in an effort to meet the difficulty many Notwithstanding contractors haveundertaken riskallocation. fairlydetailedcontractual Andrew Rogers, Forewordto FORCE OFCONTRACT, at v, v-vi MAJEURE ANDFRUSTRATION ed., 2d ed. 1995). (EwanMcKendrick 135. A rareexampleof a successfuldefaultrule is the requirement that a breacher pay the otherpartythe differencebetweenthe contractand marketprices. The rule applies in only one the stateof the world:whenthereis breach.It is simpleto applybecausethe courtonly compares contractand marketprices, and it is efficient for many partiesbecause the rule protectsthe Othergood defaultrulesarehardto find. interest. expectation
2003]
605
For example, the parties and a court will know whether a fire destroyedthe seller's factory in whole or in part. Information often is asymmetric, however. Partiesmay not know relevantthings about each other, and courts may not know relevant things about parties. The existence of asymmetric information exacerbates the moral hazard concern and thus makes the creationof efficient defaults even more difficult. D. TheAsymmetricInformationConcern The existence of asymmetric informationtruncatesthe set of contracts that parties can write. As a consequence, it also truncatesthe set of defaults that drafters can write. To see why, it is helpful to begin with an information taxonomy from the contract theory literature. A datum of information is "unobservable" if a party cannot observe it. Buyers ordinarily cannot observe a seller's production cost. A datum of informationis "observablebut not verifiable"if a party can observe it, but cannot verify the information'sexistence to a thirdparty such as a court at an acceptable cost. For example, an employer usually can know which employees sometimes shirk, but it would be expensive relative to the gains to prove to a court that a particularemployee shirkedtwenty percent of the time. A datum of informationis "verifiable"if a party can both observe it and establish its existence to a thirdparty. Informationis asymmetricwhen it is eitherunobservableor unverifiable. To understand the constraints on contracting behavior caused by asymmetric information, we change the relation-specific investment example of Section III.B in two ways. First, we let the seller's costs be stochastic. In some states of the world, the seller's cost to produce the specialized product will be lower than the buyer's valuation, as in the original example; in other states of the world, the seller's costs will exceed the buyer's valuation, so that tradewould be inefficient. Second, we let the buyer invest to increase the value to the buyer of the seller's performance. These changes generate the well-known overinvestment problem.136 The buyer's damages if the seller breaches will be the difference between its realized valuation and the price. The buyer thus will invest until the marginalcost of furtherinvestment equals the marginal increase in value. This is too much investment,however. In those states of the world in which it is inefficient for the partiesto trade,the buyer's investmenthas no social value; it serves only to increasethe damages the seller will pay.
136. The problem was identified in Steven Shavell, Damage Measuresfor Breach of J. ECON. 466 (1980), and elaboratedon in William P. Rogerson,Efficient Contract,11 BELL RelianceandDamageMeasures 39 (1984). for Breachof Contract,15RANDJ. ECON.
606
[Vol. 113: 541
This problem would vanish if the seller could observe the buyer's production function-the functional relation between the possible investment levels the buyer could choose and the values associated with those levels. The parties then could write a liquidateddamages clause that would award the buyer the difference between (1) the value the goods would have had were the buyer to invest optimally and (2) the price. This value would be lower than the unconstrained value that would be produced if the buyer never took into account that in some states of the world its investment would generate no value at all (because the parties would not trade in those states). Such a contract can seldom be written, however, because sellers can seldom observe buyers' production functions. As a consequence, a buyer would specify as liquidated damages the unconstrainedvalue to the buyer of the seller's performance, and then invest to realize that value.137 To be sure, this example does not show that parties could never induce efficient investment when information is asymmetric. When parties can observe, though not verify, relevantvariablessuch as the buyer's valuation, theorists have developed a numberof contractsthat could induce sellers to invest efficiently to reduce costs and buyers to invest efficiently to increase These contracts, however, are "parameterspecific"-that is, the value.138 the contractsset and the actions they requiredepend on the particular prices thatpartiesface.139 values, costs, and probabilitydistributions
137. Courtscannotprevent this practice becausea courtcan observeonly whatthe partiescan observe.If the courtdoes not know the value that optimalinvestment would have generated,it cannotawardthat value to the buyer.It must insteadawardthe value the buyeractuallylost. A of writingfirst-best when efficientcontracts nice, thoughtechnical,discussionof the impossibility both the seller's costs and the buyer'svaluationare unobservable is PatrickW. Schmitt, On the Interplayof Hidden Action and Hidden Informationin Simple Bilateral TradingProblems, 103 J. ECON. THEORY 444 (2002). 138. Concisebut moderately technicaldiscussionsof these contracts are Patrick W. Schmitz, TheHold-UpProblemand Incomplete Contracts: A Surveyof RecentTopicsin ContractTheory, 53 BULL.ECON. RES. 1 (2001); and Alan Schwartz,IncompleteContracts,in 2 THENEW PALGRAVE DICTIONARY OF ECONOMICS AND THELAW,supra note 30, at 277. A nice-though also technical-example is BenjaminE. Hermalin& MichaelL. Katz,Judicial Modification of ContractsBetweenSophisticated Parties: A More CompleteViewof IncompleteContractsand TheirBreach,9 J.L.ECON. & ORG. 230 (1993). 139. Asymmetricinformation also can create"ambiguity aversion,"and when it exists an attempt by the stateto fill contractual gapswouldusuallybe futile.To see why, assumethatn > 1 distributions the occurrence of a possibleevent or stateof affairs.A probability may characterize fashionwhen it calculatesits expectedpayoffsby using the partybehavesin an ambiguity-averse lowest probability in any of the n possible distributions. Suppose,then, that partyF would be betteroff if state6) materializes thanpartyG wouldbe, while partyG may be betteroff if state li materializes than partyF would be. When F is calculatingits expectedpayoffs, it will use the distribution thatputs a relativelyhigherweight on state 1* obtaining,in which it is probability worse off, thanit wouldput on state )o obtaining, in which it is betteroff. For similarreasons,G will put a relativelygreater Whenthe partiesact in this fashion,they weighton state o obtaining. calculatetheirexpectedpayoffs differently even thoughthey agreeon the monetary payoffs;the distributions. partiesestimatedifferentlybecause they are estimatingwith differentprobability This behaviorcan createan inefficiency.The sum of the payoffs that ambiguity-averse parties
2003]
607
The lesson is thatthe statecannothelp when asymmetric information from efficient the contract. Partiesthus would writing preventsparties thatpermitted the sellerto obtain"a commercially rejectdefaultstandards reasonable to its cost of production," or thatpermitted pricein proportion the buyerto recover"damages in proportion to commercially reasonable reliance investmentsmade in good faith."Under the formerproposed the selleralwayswouldclaimto havehigh costs, while underthe standard, latter, the buyer always would claim that all its investmentswere reasonable andweremadein good faith.As another commercially example, the UCCprovidesin section2-715 thata sellermusteitherperform or pay the buyerconsequential betweenthe damagesmeasured by the difference contract Thisrequirement priceandthe valueof the goods to the buyer.140 efficient because it induces the seller to performwhen is superficially would increasevalue and to breachotherwise.Valuations, performance however,often are difficultto verify. As a consequence, buyershave an to overstate theirvaluations, incentive therebyinducingsellersto perform even when it would be less costly to breachand pay true damages.141 Commercial to thisproblem out of parties respond by routinely contracting section 2-715. In place of the law, parties create complex repair-andprovisionsthat strive for efficiency in other ways.142An replacement of thatpartiesconfront appreciation the information problems compelsthis A good defaultdoes for partieswhatpartieswouldhave done conclusion: for themselveshad theircontracting costs been lower. Whenasymmetric information from (even prevents parties writingcertaintypes of contracts
with the expectto occurwill be less thanthe trueexpectedsurplus(since eachpartyis calculating most unfavorable probability).As a consequence,the parties may perceive an efficient deal, which in fact would create gains in excess of costs, as creatingcosts in excess of gains. A to determine contract andto rely on renegotiation responseto this problemis to use an incomplete the termsof tradein some possible futurestates.This can be more efficientthanwritinga more the partieswill know completeex ante contractbecausewhen a state of the world materializes, what their actualgains from tradingin that state would be. They will then tradewhen tradeis efficient and otherwisewill not. Ambiguityaversionis more likely to occur when partiesface is the correct distribution high uncertainty (so that it can be hardto knowjust which probability one to use). And when uncertainty is great,the parties'preference for contractual incompleteness rules or standards would not advancethe parties' suggests that filling gaps with state-created objectives.A recentdiscussionthus stated:"Theargument... suggestswhy incompletecontracts may not be a paradoxin a world with ambiguityaversionsince their inefficiencywill not be Aversion do not helpvery much."SujoyMukerji, 'readilyfixable'-complete contracts Ambiguity andIncompleteness REV.1207, 1218 (1998). Form,88 AM.ECON. of Contractual 140. Undersection 2-715(2)(a),the buyercan recover(in additionto directdamagesunder sections2-712, 2-713, or 2-714) "anyloss resultingfrom generalor particular and requirements needs of which the seller at the time of contracting had reasonto know."U.C.C. ? 2-715(2)(a) 141. The prospect of inefficient performancesalso would induce excessive precautions againstbreach. 142. These clauses disclaimthe seller's liabilityfor consequential damagesand restrictthe See SCHWARTZ buyer'srightto reject,butobligatethe sellerto repairor replacedefectivetenders. & SCOTT, supranote 131, at 204-09.
(2003).
608
TheYale LawJournal
[Vol. 113:541
when theircontracting defaultserves no costs are zero), a state-supplied

purpose.'43
E. Summary Muchof whatis commonlycalledcontract law consistsof the sets of in the contained defaultrulesand,muchmorefrequently, defaultstandards and UCC Article 2. We arguehere that the efforts of the Restatement writersandUCC drafters to createthis law have been largely Restatement wasted. Default rules have proved too expensive to write for large Defaultstandards economies. founder overthe parties'need heterogeneous andthe property of standards to createmoralhazard. for ex anteguidance The moral hazarddifficultyis exacerbated information. by asymmetric on behaviorthat partiesor courts cannot Defaults that are conditioned forprivate ends.Thus,parties contract observewill be exploited awayfrom them. As a consequence, inefficientdefaultsonly raise transaction costs
unnecessarily.14'"
What, then, is the properrole of courtsin resolvingdisputesover contracts? It is appropriate forcourtsto applya defaultstandard incomplete
143. KarenEggleston,EricPosner,andRichard Zeckhauser recognizethatpartiesmay write but they urge courtsto interpret information, simple contractsas a consequenceof asymmetric these contractsliberally.That is, courts should "freely"insert terms in a way that "(roughly KarenEgglestonet al., TheDesign speaking)maximizesthe ex antejoint surplusof the contract." and Interpretation Matters,95 Nw. U. L. REV.91, 127 (2000). of Contracts:WhyComplexity These authors do not show how courtscouldcreateefficienttermswhen sophisticated partieswith money at stake could not, nor do they considerthe possibilitythat sophisticated partieswould contract out of unhelpful judicialefforts. 144. This claimmay appear vulnerable to the objectionthatinefficientdefaultsare "harmless bromides" becausepartiesare alreadywritingthe contract; cost of avoidinga bad theirmarginal legal rule, thus, is slight. This objectionis mistakenon two levels. First,the total social cost of avoidingan inefficientlegal rule is not slight.The small cost of changingthe rule for a particular contract must be summedover all of the transactions thatthe rule affects in the decadesthatthe rulewill exist. Second,it is actuallycostly to changea badrulewhenwritinga particular contract because courts tend to regard state-created defaults as presumptivelyfair or efficient. This institutional bias raisesthe cost of contracting v. Postma,188 N.W.2d 31, out. See, e.g., Hayward 33 (Mich. Ct. App. 1971) (holdingthatpartiesmustuse clearand unequivocal languageto shift liabilityfor the risk of loss from seller to buyer);Davis v. Small Bus. Inv. Co. of Houston,535 S.W.2d 740, 744 (Tex. Civ. App. 1976) (holding that a contractual to provision purporting allocate to the debtor the burden of "all" expenses incurredin preservingcollateral was insufficientto triggerthe "unlessotherwiseagreed"provisionof Texas's equivalentof former section 9-207(2)(a)of the UCC, which required to the that"reasonable expenses"be chargeable MotorCo., 525 S.W.2d 238, 240 (Tex. Civ. App. 1975) (takinga debtor);Caudlev. Sherrard similarapproachwith respectto Texas's equivalentof section 2-509 of the UCC). Moreover, to give the expresslanguageof the contracta meaningthat judicial interpreters may be reluctant conflicts with the relevantdefault.See, e.g., NanakuliPaving& Rock Co. v. Shell Oil Co., 664 F.2d 772, 794-805 (9th Cir. 1981) (holdingthat a mergerclause that excludesevidenceof prior dealingsdoes not barevidenceof usageof tradeto alterthe pricetermin the contract); Legnosv. United States,535 F.2d 857, 858 (5th Cir. 1976) (holdingthat despitethe expressterm "F.O.B. vessel" in a contract,international of the contracting contextrequires that"theintention nations, rather thandefinitional niceties,mustbe given controlling weight").
2003]
609
courts as long as that standard does not createmoralhazard.Otherwise, have a choice: to dismiss a case on the groundthat a contractis too result. to reacha reasonable indefiniteto enforce,or to readthe contract and Courtsexerciseboth choicestoday,sometimesrefusingto enforce,145 for partly performedcontractswhose dismissal sometimes,particularly the contractto reach a reasonable would create hardship,interpreting from solution.We argueherethatdrafters shouldnot inferdefaultstandards cases.Rather, effortsto dojusticein particular somecourts'understandable drafters our view is that the UCC or Restatement commonlyshould do courts should be to that and hospitable attempts by laterpartiesto nothing, rulesof thecase. alteror avoidearlier in PartV thatfirmswouldrejectthe "default We have argued project" if they could becausethe projectgeneratesmany inefficientterms that contracting partiesmustincurcosts to avoid.Thereis, however,a welfareWhile role for the statein creatingcertaindefaultstructures. maximizing they almostnever partiesoften will incurthe costs of creatingcontracts, to createa structure, such as a close corporation will find it cost-justified forparties thatcreatedefault act. Structures law or a bankruptcy procedures differentformsof enterprise, to follow when conducting (or terminating) have of followingtheseprocedures, andthatspecifythe legal consequence welfare contract createdvery large law, therefore, may gains. Substantive on the wronglevel of generality: Businesspartiesarebest simplyfunction whereas the stateis bestsuitedto create suitedto createtheirown contracts, fit. Thesestructures withinwhichtheparties'contracts the broadstructures havebeenlargelyunexamined fromtheperspective of contract theory.
RULES VI. MANDATORY
contractlaw rules ban terms that partieschoose; hence, Mandatory that theserulesare inconsistent with the commitment to partysovereignty arejustifiableon two grounds. we have defended.The rulesnevertheless The first is to preventexternalities, the classic exampleof which is price
145. It is widely believed that contemporary courts ignore the indefinitenessdoctrine. Conventionalwisdom holds that courts should (and do) fill contractualgaps with general standards of reasonableness andgood faith.See, e.g., 1 CORBIN, supranote 94, ? 95, at 400. This conventionalwisdom is misleading.In a recent article,one of us shows that the indefiniteness doctrinehas survivedthe influenceof the UCC and the widespread academicsupportfor filling courtsdismissclaims See Scott,supranote 33. In dozens of cases, American gaps with standards. for breachof contracton the groundsof indefiniteness, often withoutgrantingany relief to the disappointed promisee. Scott argues that parties write incompletecontractsin the face of a doctrinewhen they can createefficientextralegalmechanismsfor coping rigorousindefiniteness with problemsof hiddeninformation and hiddenaction.Using the languageof this Article,his claimis thatpartiessometimescan solve moralhazard whencourtscannot. problems
610
[Vol. 113: 541
fixing.146 The second ground is to ameliorate a market failure that disclosure cannot cure. As an illustration, consumers are thought to be poorly informedabout the odds of productdefects, but it is very difficult to communicateprobabilityinformationin the format of a product label. The law's response is to ban disclaimersunder the doctrine of strict liability in tort. But neither of these traditional grounds can justify the mandatory interpretationrules that we discussed in Part IV. Terms such as merger clauses and no-oral-modificationprovisions affect only the parties to the contract. There is no market failure to which the rules barring their enforcementrespond. This suggests that there is a third, largely unremarked,ground for mandatory contract law rules such as the rules that govern contract interpretation. This third ground for mandatory rules is a form of context, runs like this: paternalismwhose logic, in the interpretation
(A) There are good reasons for courts to enforce the parties' intentions. (B) Parties sometimes use contractterms, such as merger clauses or no-oral-modification clauses, that restrict the ability of courtsto find the parties' intentions. (C) There is no good reason for the existence of such "intentionblocking"terms. (D) The terms thus are inconsistent with the parties' true or deep intention,which is to have the court enforce their actual deal; hence, courts should enforce the actualdeal. The interpretation rules are paternalisticbecause courts overridethe parties' expressed preferences out of concern for the parties' welfare. Courts, that is, do not conceive of themselves as imposing an agreement that parties would reject had they considered the matter under ideal deliberative conditions. Rather,courtsbelieve thatthey are supplying the agreementthat those conditionswould have produced.147 The difficulty lies with the premise specified in step (C), that parties have no good reason to write contracts with the suspect terms. Our discussion in Part IV shows that rational, well-informed and uncoerced
146. The rules prohibiting fraud and duress also function to prevent a party from externalizingcosts. Fraud and duress do not create social wealth but ratherredistributeit (unfairly) between the parties. Were these practices permitted,the costs of enforcing the redistributions wouldbe externalized to society.Thus,fraudandduressare groundsfor declining to enforcea contract, becausethe resulting dealdoes not maximize joint welfare. 147. Traditional accountsof paternalism requirethe paternalist-here a judge-to aim at the agent'swelfare.Fora discussion,see Richard J. Arneson,Mill Versus Paternalism, improving 90 ETHICS 470 (1981); and SeanaValentineShiffrin,Paternalism,Unconscionability Doctrine, 29 PHIL. & PUB. AFF.205 (2000). andAccommodation,
2003]
611
termsthat contract partieshave good reasonsto use the intention-blocking andthatmanycourtsrefuseto enforce.If thepremisein the Codedisfavors (C) is rejected,however, then the conclusion in (D)-that courts are implementingthe deep intentions of the parties-must fall as well. rules thus is misplaced with respect to the interpretation Paternalism ratherthan advancesthe parties' welfare: Parties because it frustrates In this idealconditions wouldwantwhatthe stateprohibits. choosingunder In Part,we discussthreeothercontextsin which the rulesare mandatory. haveno thatparties eachcase, therulescan onlybe justifiedon the premise but in each case good reasonto use the termsthatthe legal rulesprohibit, fails.Theserules,too, reflecta misplaced thatpremise paternalism.
A. Parties CannotBan Modifications
butthey to be written, areformally modifications Parties freeto require in theircontract frommodifying freeto prevent themselves arenot formally a the of existence from In the courts'view, the best inference the future.148 to preventmodifications is thatthe parties'originalintention modification has become outmoded.Expressedmore fully, partieswould not change withoutgood reasons,so parties the deal's substance theirmindsregarding theirminds can haveno good reasonto preventthemselvesfromchanging do the deal's substance. however, havegood Contracting parties, regarding reasonsto freezetheiroriginaldeal.We next give an exampleto show that modification bans can enable partiesto induceefficient relation-specific is to encourage contracts investment.149 Since a majorreasonfor enforcing ban the refusalto enforcea modification investment,150 relation-specific law itself. violatesa basicjustification forthe existenceof contract The example involves the productionand sale of a product.The finished productcan have two values for the buyer-high or lowis made on the ex post stateof the worldandhow the product depending thatwill investment anddelivered. The sellercan makea relation-specific increasethe likelihoodthat the productwill have the high value. We
148. See RESTATEMENT OFCONTRACTS ? 311 cmt. a (1979); see also Zumwinkel (SECOND) a no-oral-modification v. Leggett, 345 S.W.2d 89 (Mo. 1961) (holding unenforceable clause); Co., 122 N.E. 378, 381 (N.Y. 1919) (Cardozo,J.) ("Those Beatty v. GuggenheimExploration who makea contract may unmakeit. The clausewhich forbidsa changemay be changedlike any other.... Whenevertwo men contract,no limitationself-imposedcan destroy their power to contract again."); Jolls, supra note 1 (discussing reasons why parties might want to ban these codes thusare Civil law codes, on the otherhand,discourage renegotiation; modifications). Did moreconsistent with the argument in the text thanwith the commonlaw. See EricBrousseau, AND the Common ... and MayIt Change?,in 6 LAW LawBiased [sic] the Economicsof Contract Kirateds., 2001). LAW INCIVIL COUNTRIES ECONOMICS 79, 83-85 (BrunoDeffains& Thierry modifications 149. A more extensiveformaldiscussionof the parties'preferences regarding is in Schwartz & Watson,supranote 89. 150. See supraSectionIII.B.
612
[Vol. 113: 541
assume that the parties can observe the value the productturns out to have, but that values are unverifiable.Also, the buyer cannot observe whetherthe seller made the investment or not. These informationassumptionsprevent the parties from inducing efficient investment with a simple fixed-price contract.Such a contractwould either pay the seller to invest efficiently or condition the price on the product's realized value. But since neither the seller's behavior nor the product's value could be proven in court (as both values are unverifiable), neither party could enforce this simple contract. There is, however, a more complex contractthat would induce the seller to invest efficiently. This contractwould require that: (1) the seller make an up-front payment to the buyer that divides the expected surplus from the deal; and (2) after the product's value is realized, the seller make a take-itor-leave-it offer to the buyer that requiresthe buyer to pay a price equal to the value (i.e., a high or a low price, dependingon whetherthe value is high or low). To see how this contract would work, we assume that each party's payoff should the parties not deal-its disagreement payoff-is zero. Further,if the seller invests, at a cost of c > 0, the productwill have a high value for the buyer with .8 probability.If the seller does not invest, its cost is zero but the productwill be certain to have a low value. After the seller invests (or not), it produces the product.We normalize the productioncost to zero, so that it is always efficient for the parties to trade. In the example, a high value is $40 and a low value is $20. It would be efficient for the seller to invest if the expected gain less the cost would exceed the low value: i.e., if .8($40) + .2($20) - c > $20, or if c is $16 or less. We let c = $12 so that investmentwould be efficient. The expected surplusunder the contractis .8($40) + .2($20) - $12 = $24. Suppose for convenience that the parties have equal bargainingpower; then they will divide the expected surplus equally. On these assumptions,the buyer will accept the contractif the seller makes an up-frontpayment of half the expected surplus, or $12. The buyer's ex post gain would be zero because the trading price would exactly equal the product's value, but the buyer's ex ante gain is the $12 payment. The seller would offer the contract and invest because its net expected gain also would be $12: the $24 expected surplusless the up-front paymentof $12. The difficulty is that following the seller's investment the buyer will pursue a strategy that will reduce the seller's returnto below the returnit could earn without a contract.To see how, suppose the productturnsout to be of low value. The buyer will pay the $20 price and earn the product's
value less the price plus the up-front payment, or $20 - $20 + $12 = $12.
Now let the productturnout to have a high value. The buyer will reject the seller's take-it-or-leave-itoffer, return the up-front payment and offer to renegotiatethe deal. At this stage, the parties know that they will earn zero
2003]
613
if they fail to tradeandwill share$40 if they do trade.Consequently, the will the contract. Just as in the in Section III.B, example parties renegotiate the seller's investment cost is sunk.Ourassumptions thatthe partieshave whichsunk powerandengagein Nashbargaining (under equalbargaining costs are ignored)thus implythat the partieswill agreeto sharethe $40 Thebuyerwill rejectthe contract value,so thatthenew pricewill be $20.~5' and renegotiate in the high value statebecauseit will then earn$40 (the contract, value)- $20 (thenew price)= $20 underthe renegotiated product whichexceedsthe $12 it wouldhaveearned under the original contract. The buyer's strategy,if anticipated, would cause the seller not to however.To see why, assumethatthe sellercontracts andinvests. contract, value turnsout to be high, the seller will earnthe renegotiated If product cost. If product valueturnsout to be low, $20 priceless the $12 investment the buyerwill enforcethe contract, so thatthe sellerwill not havethe right of the up-front to the return In the low-valuecase, the sellerthus payment. will earnthe $20 priceless the $12 investment cost andless the $12 down a net for loss of On the abovefor high assumed $4. probabilities payment, and and low values, the expectedreturnto the seller from contracting investingwill be $5.60. If the seller insteadchooses not to incurthe $12 investment will have a low value for sureandthe seller cost, the product = $8. If the will earn$20 (the contract price)- $12 (the up-front payment) then, the seller will do betterby not investing.The partiesdo contract, will do betterstillby not contracting at all. Thenit alsowill seller,however, not invest.Rather, a product the sellerwill produce with low valueat a cost of zero and offer it to the buyerin a spot purchase. The transaction price will splitthe $20 surplus. The sellerthuswill earna net $10 if it does not contract the anda net $8 if it does. Therefore, the partieswill not contract, sellerwill not invest,andproduct low. will be inefficiently quality The lesson is thatpartiessometimesare unableto writecontracts that induce efficient relation-specificinvestment because renegotiation-a 52 An enforceableban on contract modification--cannotbe banned. modifications would permitthe seller to reinstatethe contract'soriginal the renegotiated price term.A courtcould observethe originalcontract, and the each. Thus, transfers the made under contract, monetary parties
151. The seller in this example could not sue for the $40 price because we assume that valuesarenot verifiable.Thus,the sellercouldnot provethatthe valuewas high. product 152. The investmentin this example is termed"cooperative" because the seller invests to increase the buyer's value. When renegotiationis permitted,cooperativeinvestmentcan be impossible to motivate by contract.See Yeon-Koo Che & Donald B. Hausch, Cooperative Investments and the Valueof Contracting, 89 AM.ECON. REV.125 (1999); Ilya Segal, Complexity andRenegotiation: A Foundation 57 (1999). Our 66 REV. ECON. STUD. Contracts, for Incomplete earlierexamplesinvolvedself-investment (the seller investsto lower her costs; the buyerinvests to increasehis value). Efficientself-investment contracts. is easier to motivatewith appropriate Bothkindsof investment arecommon.
614
[Vol. 113: 541
when the buyer returnedthe $12 up-front payment and paid $20 under a renegotiatedcontract,the court would know that productquality was high. The court then would order the buyer to pay another$20. Expecting that a court will undo any renegotiation,the buyer would realize that its options were limited to purchasing the low-quality product on the spot market, earning a net $10, or making the contract, accepting whatever ex post take-it-or-leave-it offer the contract permitted the seller to make, and earninga net $12. The buyer thus will make and comply with the contract. Anticipating the buyer's behavior, the seller will agree to contract, and it will invest efficiently. To summarize, parties can have good reasons for banningmodifications.153 B. Parties MustAccept SubstantialPerformance Courts will generally require parties to accept substantialrather than full performanceunless, in the court's view, the deviation is material.154 default by Parties sometimes try to opt out of this substantial-performance making full performance an express condition of the promisee's duty to pay. Courts, however, frequently refuse to require exact compliance with The basic the express condition because "the law abhors a forfeiture."155 premise is that the performingparty would not have agreed to a contract that would penalize it severely for minor deviations. Since the paying party would have known this, neither party would have thought that the contract requiredthe equivalentof a forfeiturefor a slight nonconformity,in spite of
153. Modification bans also can be efficient when one of the partiesis risk-averse.To see and a risk-averse betweena risk-neutral agentwho is principal why, considera possible contract it is commonlyassumed,cannotverify the agent'sbehavior supposedto do a task.The principal, the agentto bearrisk;thatis, to a court,so the contract mustmotivatethe agent.This will require her pay must be contingenton the outcome,so she will try to producea good result.The agent bearsriskbecausethe outcomeis a functionbothof herefforts,which increasethe likelihoodof a deal good result,andchance.Afterthe agentacts,butbeforethe resultis known,a Pareto-superior between the parties becomes possible: Since the agent has acted, she no longer needs to be risk fromher butshe still bearsriskbecausenature has yet to act. The dealwill transfer motivated, to the risk-neutral principal by payingthe agenta fixed fee thatwill lie somewherebetweenthe this renegotiation, contract's however, good- andbad-state payoffsto her.If the partiesanticipate it will be the they will knowthatthe agent'spayoff will not be a functionof the outcome;rather, fixed fee the agentexpectsto get in the renegotiation. Thus,the agentcannotbe motivatedto try hard because her payoff actually is noncontingent. When trying hard is efficient, permitting renegotiationthus is inefficient. The parties, ex ante, again have good reasons to prevent themselvesfromchanging theirminds.See Jolls,supranote 1, at 209-24. is 154. See RESTATEMENT OFCONTRACTS ?? 237, 241 (1981). If the performance (SECOND) the defaultrule is that the promiseemust transferthe price less a sum that would substantial, it for the deviationfromfull performance. As we will see below, whenthe promisee's compensate valuation is not verifiable the rightto deductdamagesis hollow. 155. See id. ? 227 cmt. b ("The policy favoringfreedomof contractrequiresthat, within broadlimits.., .the agreementof the partiesshould be honoredeven though forfeitureresults. makes an event a conditionof an When, however,it is doubtfulwhetheror not the agreement see also is preferred that will reduce the risk of forfeiture."); obligor's duty, an interpretation
E. SCOTT & JODYS. KRAUS,CONTRACT 718-20 (3d ed. 2002). ROBERT LAWANDTHEORY
2003]
615
what the written words appearto say. The parties, courts believe, have no good reason to require forfeitures; thus, courts strictly construe express conditionsthatrequirefull performance. We offer an example, drawn from the famous case of Jacob & Youngs, Inc. v. Kent,'56to show that parties have good reasons to construct deals thatmake forfeiturespossible. Considera contractthat requiresan owner to make progress payments to a builder as constructionprogresses. The last payment, which is sizable, is due when construction is completed. We assume that values are unverifiable.Thus, a court cannot observe the value to the owner of a building completed in accordancewith the contract.On this assumption,there can be moral hazardon both sides. If the builderhas the burdenof proof, the owner may claim that defects in the final version reduce her value substantially,even though they do not. The unverifiability of valuationswould make it difficult for the builder to disprove this claim. Conversely, if the owner has the burden of proof, the builder may deliberatelyrender a defective performancebut claim that, in fact, it had substantiallycomplied. the potentialeffect of such moralhazard,assume that the To understand builder has the burden of proof and that the owner will cheat by withholding the entire final payment if the building is less than perfect. Since perfectionis difficult to achieve, the builderwill expect not to receive the final payment. Therefore, it will not renderthe final performance.But then the owner will know that the contract's penultimateperformancewill be the final one, and it will then cheat. The builder, anticipatingthis, will not render the penultimate performance, and so forth. In equilibrium, therefore, the builder will render no performance at all. The parties' contracting problem thus is to induce the builder to perform when the of values makes strategicbehavior likely. unverifiability The parties' solution follows from the contextualnatureof verifiability. A datumof informationmay not be verifiable to a courtbecause explaining mattersto a generalistjudge or a lay jury can be costly in relation to the gains. The same datum of informationmay be verifiable to an arbitrator, however. The arbitrator's expertise makes her cheaperto inform;she acts in an informalsetting; and she has a reputationalstake in not appearingto be biased in favor of buildersor owners. To ensure that parties cooperatewith the arbitrator, the parties will make her decision conclusive in the absence of fraud,bias, or mistake. In Jacob & Youngs,the partieshad adoptedthis common solution to the moral hazardproblem, making an architect the arbitrator.157The architect refused to certify that the builder had fully complied, though the defect
156. 129N.E. 889 (N.Y. 1921). 157. See id. at 890.
616
[Vol. 113: 541
appeared trivial.'"5 The seeming disjunction between the size of the withheld final payment and the nature of the noncompliance suggested possible fraud or mistake by the architect. The builder, however, did not attempt to impeach the architect's decision. Rather, the builder asked a court to hold that perfect compliance was not a condition to receiving the entire last payment; the court agreed.'59It believed that forfeiture of the entire last payment would have been unfair and that the parties could not have intended this result.'60 The issue the case posed, however, was not whether the parties had or lacked good reasons to permit forfeitures. Instead, the issue was whether the parties had good reasons to make the architect's findings conclusive. When the parties adopt a sophisticated governance scheme, it is mistaken paternalism for a court to require a promisee to accept substantialperformanceon the ground that the parties lack good reasonsto requirea perfect tender. C. Parties CannotAgree to Penalties Contractingparties are permittedto specify the damages the breaching promisor must pay, provided that the specified damages represent a reasonableestimate of the promisee's lost expectation.'61 Parties, however, are not permittedto specify damages that exceed a reasonableestimate of the promisee's expectation. The logic here is the same as in the above examples. The parties made the contract in order to give the promisee a particularperformance for a price. They know that breaches sometimes occur, so the partiesalso have good reasons to ensurethat the promisee will receive a monetary substitute for performancein the event of breach. But contracting parties supposedly do not have good reasons to award the promisee much more than its lost expectation when the promisor fails to perform. The parties' deep contractualgoal is advanced, therefore, by a mandatoryrule declaringpenalty clauses unenforceable. This premise fails for two reasons. First, courts will sometimes implement the penalty rule inaccurately. Courts regulate liquidated damages clauses by comparingthe differencebetween anticipateddamages underthe expectationdamages default rule to the stipulateddamages in the contract. Expectation damages, in turn, are based on the verifiable losses that the promisee anticipates from breach. Any liquidateddamages clause
158. See id. 159. See id. at 890-91. 160. See id. 161. See U.C.C.? 2-718(1) (2003) ("Damages in forbreachby eitherpartymay be liquidated the agreement but only at an amountwhich is reasonable in the light of the anticipated or actual harmcausedby the breach.... A termfixing unreasonably largeliquidated damagesis void as a penalty.").
2003]
617
that incorporatesobservable but nonverifiablevalues thus will be to a penaltyclaimeven whenthe clauseaccurately vulnerable measures the promisee's lost expectation.'62 Moreover, courts sometimes find clausesto be penaltiesin complexcases compensatory liquidated damages becausethe courtshave failedto understand just how the clauseprotected the promisee'sexpectation.163 These difficultieswith currentlaw permit to the invoke As a consequence, penaltydoctrinestrategically. promisors arediscouraged fromusingliquidated clauses parties damages sophisticated whentheseclauseswouldotherwise be optimal. The second problemwith currentlaw is directly relevant to the advanced above.Thepremise thatparties haveno goodreasonto argument for penaltiesis itself mistaken. contract canpermit Rather, parties penalties to induce efficient relation-specific investmentsin certain asymmetric information environments. Commercial partiesthus have good reasonsto to in return to the these As an illustration, circumstances. agree penalties examplein SectionVI.A that showedthe errorof preventing partiesfrom antimodification clauses in contracts. An alternative writing way to produce efficientinvestment in thatcase wouldbe for the partiesto contract with a thirdpartythatif eitherpartyshouldrequesta modification, it wouldhave to pay a largepenaltyto the thirdparty.Thispartywouldhavean incentive to enforcethe prohibition on modifications in orderto collectthe penalty. sucha suit, the buyerin our examplewouldbe deterred from Anticipating seen in These schemes are not requesting renegotiation. practice third-party becausepenaltytermsare unenforceable undercurrent law.64 Banninga liquidateddamages clause because, and only because, it requiresthe breachingparty to make a transferthat exceeds its contractpartner's thus wronglyinterferes with the parties'sovereignty and may expectation
generateinefficiency.165
To summarize, law containsa number rulesthat contract of mandatory or a marketfailure.These rules apply in the absenceof an externality override contractual termsthatappear withthe intentions to be inconsistent
162. See Goetz& Scott,Liquidated Damages,supranote 1, at 568-76. 163. See Alan Schwartz,TheMythThatPromiseesPreferSupracompensatory An Remedies: L.J.369, 383-87 (1990). Analysisof Contractingfor DamageMeasures,100 YALE 164. For a review of the variousways, includingthe one above,that enforcingpenaltiescan enhance efficiency, see Aaron S. Edlin & Alan Schwartz,OptimalPenalties in Contracts, 78 CHI.-KENT 33 (2003). L. REV. 165. See id. Whenthe sellerhas marketpower,partiesmay use penaltiesto deterthe entryof into the seller's market.See id. at 40-42. Undercurrent competitors law, however,courtsstrike whatthey perceiveto be penaltytermswhetherthosetermswereusedto increaseinvestment or to impede entry. We argue here that courts should only ban inefficient penalties;to ban other Anotherway to put this claim is that a party should penaltiesis only misplacedpaternalism. a or perpetuate always be free to arguethat a damagestermwould createa negativeexternality market thatwe maintain would still be sufficientreasonfor courtsto strikesuch failure--grounds a term. But it is a mistake to treat as a sufficient proxy for these inefficiencies a liquidated a partyin expectation. damagesclausethatwouldovercompensate
618
[Vol. 113: 541
that rational, informed, and uncoerced parties would have under the circumstances. It appears to courts and to the drafters of the UCC that actual parties have no good reasons for choosing these terms, and that a defensible paternalismthus would not enforce them. In the cases we have analyzed, however, commercial parties turn out to have good reasons for the things they do. Put anotherway, the contractterms that courts and the UCC refuse to enforce actually advance the parties' welfare. A paternalistic justification for contractlaw's mandatoryrules thereforefails. This should not be surprising.In Part II, we showed that commercial parties pursue a goal-joint welfare maximization-that the state supports, and generally can choose the means that best implement this goal. The rules regulating contractsbetween business firms thus should be mandatoryonly when the parties' contractcreatesan externalityor is the productof marketfailure. VII. CONCLUSION This Article does not ask the conventional normative question: What contractlaw should the state provide?Rather,it asks: What contractlaw do business firms want the state to provide? A contract law for firms, we answer, would be narrowerand more deferentialto contractingpartiesthan the contract law we now have. Of first-orderimportance,firms want the state to enforce the contracts that they write, not the contracts that a decisionmaker with a concern for fairness would prefer them to have written. Enforcement of written agreements presupposes a theory of The commitmentto party sovereignty that we defend in this interpretation. Article requirescourts to delegate to partiesboth the choice of a contract's substantiveterms and the choice of the interpretive theory that will be used to enforce those terms. Commercial parties, we show, commonly prefer adjudicatorsto be accurateon average in ascertainingthe meaning of their agreements ratherthan accurate in every instance; therefore, these parties want courts to make interpretations on the smallest evidentiarybases that will supportinterpretations that are accurateon average.Courtsthatdefer to party preferences regarding interpretation thus will use a textualist interpretive style, one that restricts the evidentiary base to the written agreement and not much more. In addition, most firms prefer courts to interprettheir contractswith the presumptionthat the contractis written in what we call majoritytalk, the languagethat firms and courtsusually speak. For this reason, too, a textualistinterpretive theory is the best default. Much of today's contract law is in the form of default rules and standards.These defaults cause more harm than good. An efficient default rule-one that firms will accept-is simple in form, conditioned on few states of the world, and maximizesjoint gains in a wide variety of contexts. A default standard is efficient only when parties can live with vague
2003]
619
definitions of their contractingobligations.Because standardsconfer will be unsatisfactory considerable discretion on parties,a standard if, as a underit. resultof thatdiscretion, partiesare likely to behavestrategically are finite even costs for As we show, partiesare heterogeneous, drafting rules must be sometimes complex, parties public decisionmakers, to redistribute ratherthan to maximize commonlywill exploit standards and information often is asymmetric. Whenthe statetriesto joint surplus, create write efficient default rules and standards, these circumstances business obstaclesthat the state can seldom overcome.Unsurprisingly, out of these faileddefaults.The effectivedomainof partiesoften contract contract law thusis smaller thanis widelybelieved. state-supplied The welfare-maximization that we advancejustifies courts in goal enforcement to unconscionable contractsaffectedby contracts, refusing thatcreateexternalities. fraudor duress,andcontracts Thisgoal, however, cannot supportmany of the mandatory rules that today govern much contractingbehavior between firms. These rules bar enforcementto termsthat efficientlycope with problemsof hiddeninformation contract andhiddenaction. A normativetheory of contractlaw that takes party sovereignty law over the last seriouslyshows thatmuch of the expansionof contract has been law is ill-advised.Contract today composedof a few fifty years defaultrules, many defaultstandards, rules. and a numberof mandatory to defaults, Mostof the mandatory rulesshouldbe repealed and or reduced most of the defaultsshouldvanishfromthe law. Advocating freedomof contract for firmsis uncontroversial. contract of freedom seriously, Taking current contractlaw. A law merchant however,would radicallytruncate to ourtime wouldbe a merchants' the law; andfor merchants, appropriate less publiclysupplied lawthebetter.

Schwartz-Scott - Contract Theory and The Limits of Contract Law

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Schwartz-Scott - Contract Theory and The Limits of Contract Law

Transféré par

Droits d'auteur :

Formats disponibles

The Yale Law Journal Company, Inc.

ContractTheory and the Limits of ContractLaw

CONTENTS I. II. INTRODUCTION .................................... 543

AN EFFICIENCY JUSTIFYING THEORY OFCONTRACT...................... 550

t SterlingProfessorof Law,Yale Law School;Professor,Yale Schoolof Management.

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

Limitsof Contract Law

Limitsof Contract Law

Limitsof Contract Law

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

23. U.C.C. ? 2-302(1) (2003).

Limitsof Contract Law

Limitsof Contract Law

Limitsof Contract Law

To make our examplemore concrete,supposethe buyervalues the

Limitsof Contract Law

Law Limitsof Contract

564 is as of To), (measured

(Pk+ z)] + 3 [(Pk +y) - Pk]+

[(Pk+ Y)- (Pk-

The Yale Law Journal

[Vol. 113: 541

Judicial Strategies, 21 J. LEGAL STUD.271, 308-13 (1992).

The Yale Law Journal

[Vol. 113: 541

The Yale Law Journal

[Vol. 113: 541

The Yale Law Journal

[Vol. 113: 541

U.C.C. ? 1-303(e)(1) (2003).

The Yale Law Journal

[Vol. 113: 541

Customs Exist?, in THE JURISPRUDENTIAL FOUNDATIONS OF CORPORATE AND COMMERCIAL

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

The Yale Law Journal

[Vol. 113: 541

The Yale Law Journal

[Vol. 113: 541

Limitsof Contract Law

The Yale Law Journal 3. Course-of-Performance Evidence

[Vol. 113: 541

The Yale Law Journal

[Vol. 113: 541

JURISPRUDENTIAL FOUNDATIONS, supra note 71, at 12, 16.

The Yale Law Journal

[Vol. 113: 541

The Yale Law Journal

[Vol. 113: 541

Implications of Transaction Cost Economics, 152 J. INSTITUTIONAL & THEORETICAL ECON.287

The Yale Law Journal