siena memos and papers on law and economics

Simone Maria Sepe

Good Faith and Contract Interpretation:

A Law and Economics Perspective

Simple

42/06

UNIVERSITY OF SIENA FACULTY OF ECONOMICS R.M. GOODWIN

SIMONE MARIA SEPE

Good Faith And Contract Interpretation:

A Law And Economics Perspective

University of Siena & Yale Law School

Simple 42/06

1

GOOD FAITH AND CONTRACT INTERPRETATION:

A LAW AND ECONOMICS PERSPECTIVE

INTRODUCTION

1. THE GOOD FAITH DEBATE: AN OVERVIEW 1.1. STATUTORY AND COMMON LAW CONCEPTUALIZATIONS

1.1.1 THE UNIFORM COMMERCIAL CODE CONCEPTUALIZATION 1.1.2. THE RESTATEMENT CONCEPTUALIZATION

1.2. SCHOLARLY CONCEPTUALIZATIONS 1.2.1. THE EXCLUDER APPROACH (OR CONTRACTUAL MORALITY APPROACH) 1.2.2. THE FOREGONE OPPORTUNITY APPROACH (OR ECONOMIC APPROACH) 1.2.3. THE LAW-AND-ECONOMICS APPROACH 1.2.4. THE NEW-FORMALISM APPROACH

1.3. A DIFFERENT PERSPECTIVE

2. IS THE OBLIGATION OF GOOD FAITH EFFICIENT? 2.1. GOOD FAITH AS A MEANS TO DETER OPPORTUNISM ARISING FROM

CONTRACTUAL INCOMPLETENESS AND TRANSACTIONAL INSECURITY 2.2. THE TRADE-OFF IMPLIED BY THE OBLIGATION OF GOOD FAITH

3. A TRANSACTION-COST ANALYSIS OF THE GOOD-FAITH OBLIGATION 3.1. CONTRACTUAL TRANSACTION COSTS 3.2. THE CONTINUUM BETWEEN SPECIFICATION AND DISCRETIONARY TERMS 3.3. THE EFFICIENCY CONDITION FOR THE GOOD-FAITH OBLIGATION

3.4. DETERMINANTS OF THE EFFICIENCY CONDITION: NATURE OF THE PARTIES

AND TRANSACTIONAL ENVIRONMENT 3.4.1. THE DISTINCTION BETWEEN UNSOPHISTICATED AND SOPHISTICATED PARTIES (A) UNSOPHISTICATED PARTIES (B) SOPHISTICATED PARTIES

3.4.2. THE DISTINCTION BETWEEN NON-IDIOSYNCRATIC AND IDIOSYNCRATIC

TRANSACTIONS (A) NON-IDIOSYNCRATIC TRANSACTIONS (B) IDIOSYNCRATIC TRANSACTIONS

4. CONCLUSIVE REMARKS 4.1. WHY THE GOOD-FAITH OBLIGATION SHOULD BE A DEFAULT RULE

4.2. HOW PARTIES SHOULD WRITE CONTRACTS 4.3. HOW COURTS SHOULD INTERPRET CONTRACTS

2

INTRODUCTION

The widespread acknowledgement of the implied contractual obligation of good

faith is a relatively recent phenomenon in the American legal landscape.1 Not until the

introduction of the Uniform Commercial Code, in the 1960s, was a general obligation of

good faith imposed by state legislatures.2 In a few decades, however, the requirement of

good faith in contractual relations has become one of the most important3 and, at the same

time, controversial issues of American contract law.4

1 See Robert S. Summers, The Conceptualization of Good Faith in American Contract Law: A General

Account, in REINHARD ZIMMERMANN & SIMON WHITTAKER (edited), GOOD FAITH IN EUROPEAN CONTRACT

LAW 119 (2000); E. ALLAN FARNSWORTH, FARNSWORTH ON CONTRACTS § 7.17b 393 (2004). Still, the

obligation of good faith was not totally unknown to American courts. In the judicial landscape, the obligation

of good faith emerged at the turn of the twentieth century, mostly in decisions of New York courts. See, e.g,

Wood v. Lucy, Lady Duff-Gordon, 222 N.Y. 88, 118 N.E. 214 (1917); Kirke La Shelle Co. V. Paul

Armstrong Co. 263 N.Y. 79, 87, 188, N.E. 163, 167 (1933). Until the enactment of the Code, however, such

cases remained scattered law.

2 Most jurisdictions have expressly recognized a duty of good faith in their common law of contracts

only after the publication of the Uniform Commercial Code (U.C.C.), in 1957. See Steven J. Burton, Breach

of Contract and the Common Law Duty to Perform in Good Faith, 94 HARV. L. REV. 369, 404 (1980) (setting

out, in appendix, a lengthy list of cases indicating jurisdictions that expressly recognize a common law general

obligation of good faith in every contract).

3 See E. ALLAN FARNSWORTH, FARNSWORTH ON CONTRACTS § 7.17b (2004). (“The concept of good

faith has, in a relatively few decades, become one of the peculiarity American cornerstones of our common

law of contracts.”)

4 The copious literature on the good-faith obligation seems to confirm its controversial nature. Limiting

the reference to the scholars who have discussed good faith as a general principle of law, rather than in

relation to a particular field of application, see, among the others, E. Allan Farnsworth, Good Faith

Performance and Commercial Reasonableness Under the Uniform Commercial Code, 30 U. CHI. L. REV. 666

(1963); Robert S. Summers, "Good Faith" in General Contract Law and the Sales Provisions of the Uniform

Commercial Code, 54 VA. L. REV. 195 (1968) [hereinafter, Summers, Good Faith In General Contract Law];

Robert S. Summers, The General Duty of Good Faith—Its Recognition and Conceptualization, Symposium:

The Restatement (Second) of Contracts, 810 CORN. L. REV. (1982); Steven J. Burton, Breach of Contract and

3

Three dominant views have emerged from the vivid scholarly debate developed

around the parties’ implied obligation of good faith. At one extreme, advocates of a

textualist approach to contract interpretation,5 concerned that such a general concept as

good faith might inject uncertainty in legal relationships, support a restrictive view of the

obligation. According to this view, it should be limited to a prohibition of intentional

dishonesty and courts should not invoke good faith to modify explicit contractual terms or

restrict powers attributed by contract. At the other extreme, supporters of a contextualist

approach to contract interpretation argue that the obligation should be interpreted

the Common Law Duty to Perform in Good Faith, 94 HARV. L. REV. 369 (1980) [hereinafter Burton, Common

Law Good Faith]; Steven J. Burton, Good Faith Performance of a Contract Within Article 2 of the Uniform

Commercial Code, 67 IOWA L. REV. 1 (1981) [hereinafter Burton, Good Faith Performance]; Steven J.

Burton, More on Good Faith Performance of a Contract: A Reply to Professor Summers, 69 IOWA L. REV.

497 (1984) [hereinafter Burton, More on Good Faith Performance]; STEVEN G. BURTON & ERIC G.

ANDERSEN, CONTRACTUAL GOOD FAITH (1995); Eric G. Andersen, Good Faith in the Enforcement of

Contracts, 73 IOWA L. REV. 299 (1988); B.J. Reiter, Good Faith in Contracts, 17 VALP. U. L. REV. 705

(1983); Thomas A. Diamond & Howard Foss, Proposed Standards for Evaluating When the Covenant of

Good Faith and Fair Dealing Has Been Violated: A Framework for Resolving the Mistery, 47 HASTINGS L.J.

585 (1995-1996); Michael P. Van Alstine, Of Textualism, Party Autonomy, and Good Faith, 40 WM. AND

MARY L. REV. 1223 (1998-1999); ROGER BROWNSWORD, NORMA J. HIRD, AND GERAINT HOWELLS, GOOD

FAITH IN CONTRACT: CONCEPT AND CONTEXT (1999); Marietta Auer, Good Faith: A Semiotic Approach, 2

EUROPEAN REVIEW OF PRIVATE LAW 279 (2002); Ejan Mackaay and Violette Leblanc, The Law-and-

Economics of Good Faith in the Civil Law of Contract, 2003 Conference of the European Association of Law-

and-Economics, Nancy; Emily M.S. Houh, Critical Interventions: Toward an Expansive Equality Approach to

the Implied Obligation of Good Faith in Contract Law, 88 CORNELL L. REV. 1025 (2003).

5 The textualist approach to contract interpretation is so termed because it confines the action of the

interpreter basically to the sole text of the contract. This approach is also termed “classical’ or “Willistonian”

as it was dominant in contract interpretation during the late nineteenth and early twentieth century and is often

associated with the views of Professor Samuel Williston. See 1 SAMUEL WILLISTON, A TREATISE ON THE LAW

OF CONTRACTS § 95, at 349-50 (Walter H.E. Jaeger ed., 3d ed. 1961). (“The court will give [written contract]

language its natural and appropriate meaning; and, if the words are unambiguous, will not even admit

evidence of what the parties may have thought the meaning to be.”)

4

expansively to guarantee the substantive justice of contractual relations.6 Under this view,

the requirement of good faith incorporates equitable standards of fairness and decency

whose respect might justify departures from express contractual provisions. In between

these extremes is the law-and-economics view of the duty of good faith as equivalent to a

prohibition of opportunistic behavior.7 The duty precludes a party from profiting of

contractual incompleteness to extract uncompensated benefits from her counterparty.

Thus, for law-and-economics scholars, the basic function served by good faith is to save

parties the cost of negotiating and drafting express contractual provisions that prevent

opportunistic behavior (the “specification cost”).8

Two common features bring together these otherwise disparate views. They all

conceive of the obligation of good faith as an immutable rule of law and focus on how

courts should apply such a rule. I propose a different approach. I claim that the obligation

of good faith (hereinafter, also “good faith”)9 should be a default rule that parties should

6 The contextualist approach to contract interpretation, so termed in contraposition to the classical

textualist approach, challenges the idea that express terms always represent the best evidence of the parties’

agreements, at least without an examination of “the context” of that agreement. Arthur Corbin and Karl

Llewellyn are among the most distinguished representatives of the contextualist approach. See, e.g., Arthur L.

Corbin, The Interpretation of Words and the Parole Evidence Rule, 50 CORNELL L.Q. 161, 161-70 (1965)

(observing that written words are intrinsically ambiguous); Karl N. Llewellyn, The Rule of Law in Our Case-

Law of Contract, 47 YALE L.J. 1243 (1938).

7 The law-and-economics view of the good-faith obligation and the notion of opportunistic behavior

are discussed in detail in the following of the work. See infra Par.s 1.2.3, 2.1.

8 A more detailed description of specification cost is discussed infra. See text following note 85.

9 In this work, the expression “good faith” indicates, alternatively, (i) the “good-faith obligation”, or

the “good-faith requirement”, intended as both the duty of and the entitlement to good faith of contracting

parties, and (ii) the “good-faith rule”, intended as the legal imposition of the good-faith obligation or

requirement.

5

include in their agreement only when it maximizes the ex-ante value of their contractual

relationship. In this work, I discuss then under what conditions the requirement of good

faith proves efficient and propose a basic framework of reference for the parties’ decision

to include or exclude good faith in their contract.

In accordance with the law-and-economics view, I conceive of the obligation of

good faith as the rule of law that prohibits each contracting party from taking advantage

of the contract’s incompleteness to expropriate her counterparty’s expected contractual

benefits. I challenge, however, the law-and-economics argument supporting the efficiency

of good faith. For law-and-economics scholars, good faith reduces contractual

specification cost by supplementing the parties’ agreement with an implied general term

that prevents opportunistic behavior. This view, however, overlooks the higher risk of

judicial error that parties face when courts interpret their agreement pursuant to the

obligation of good faith (i.e., under a good-faith interpretative regime).

Due to the intrinsic vagueness of the concept at hand, the definition of the parties’

entitlements is less clear when they decide to address a particular contingency “through

good faith”,10 rather than through an explicit contractual provision. In addition, under a

good-faith interpretative regime, the court is not obliged to abide by the express terms of

the contract, but can determine in its own discretion what good faith means in relation to

both specified and unspecified contingencies. Thus, when a contract includes good faith,

the court might make errors even if the parties’ entitlements are clearly specified by

express provisions. It follows that the inclusion of good faith in the parties’ agreement

proves efficient only when the specification cost that it permits the parties to save exceeds

10 See infra Par. 2.2. at text between notes 79-80.

6

the expected cost they bear due to judicial error. Put differently, the obligation of good

faith implies a trade-off between specification-cost saving and expected cost deriving

from judicial error (the “good-faith trade-off”).

My claim is that contracting parties are in the best position to evaluate the good-

faith trade-off. Because of their informational advantage over courts, they know better

when good faith serves efficiency. Hence, parties themselves should decide whether to

include or exclude good faith in their agreements. From a practical viewpoint, this means

that the interpretative regime should be determined by private autonomy, rather than be a

judicial decision based on a-priori assumptions. Parties know better than courts which

interpretative stile is more likely to yield the “correct answer”11 and, hence, maximize

the ex-ante value of their contractual relationship. As a result, courts should commit to

party sovereignty in contract interpretation.12

In the good-faith regime I propose, parties are free to choose whether (i) to exclude

good faith from their contracts and, thereby, opt for a literal interpretative regime in

which the contract is the only evidentiary base courts should use in enforcing their

agreement; or (ii) to include good faith and opt for a good-faith interpretative regime,

giving courts indications on the evidentiary base that should be used to interpret the

content of the good-faith obligation.

Finally, I conclude that the obligation of good faith should be a default rule of law.

It should not be a mandatory rule, as traditionally argued, because it does not always

11 See infra note 64.

12 Few exceptions should be admitted to this principle. See infra Par. 4.3.

7

serve efficiency. In particular, I show that in the case of idiosyncratic transactions, 13

imposing good faith and, therefore, a good-faith interpretative regime, would increase

contractual transaction costs and, ultimately, compromise economic efficiency. However,

the obligation of good faith should still be provided by the law because is the rule that the

majority would choose. I explain that unsophisticated parties always benefit from the

inclusion of good faith in their contracts.14 The proposed default regime, in turn, provides

that when parties fail to indicate an explicit choice, they should be considered

unsophisticated, and a good-faith interpretative regime should apply. In this case,

however, courts remain free to choose the evidentiary base they prefer to interpret the

contract.

The proposed analysis is organized as follows. In Part I, I outline and discuss the

most significant statutory, judicial, and academic conceptualizations of the obligation of

good faith, focusing attention, in particular, on the law-and-economics literature. In Part

II, I illustrate the content of the obligation of good faith and then challenge the traditional

law-and-economics view of the obligation, by introducing the problem of the good-faith

trade-off. In Part III, I develop a transaction-cost analysis of good faith, explain the

interaction between the good-faith obligation and express contract terms, and discuss the

efficiency condition for good faith. Then, I elaborate on the circumstances in which

parties should include or exclude good faith to maximize the ex-ante value of their

contractual relationship. Finally, in Part IV, I draw some conclusive remarks on why good

13 See infra Par. 3.4.2 (A).

14 See infra Par. 3.4.1. (A).

8

faith should be a default rule, how parties should write their contracts, and how courts

should interpret such contracts.

1. THE IMPLIED OBLIGATION OF GOOD FAITH: AN OVERVIEW

As stated by an eminent law-and-economics scholar “notwithstanding the extensive

literature on the subject, no consensus exists on precisely what the duty of good faith

means.”15 The claim can be extended to the decisions of American courts. No universal

(or even predominant) paradigm of good faith has emerged from the large number of

judicial opinions that have focused on the matter.16 In addition, the Uniform Commercial

Code and the Restatement (Second) of Contracts have adopted divergent definitions of

the obligation. As a result, a vivid debate has developed around the meaning and function

of good faith. To clarify the current status of the debate, I briefly describe the most

significant conceptualizations of good faith elaborated by the American legislator, courts,

and scholars.

1.1. STATUTORY AND COMMON LAW CONCEPTUALIZATIONS

1.1.1. The Uniform Commercial Code Definition

The Uniform Commercial Code (U.C.C.) adopts a definitional approach to good

faith. Section 1-203 of the Code’s General Provisions states that “[e]very contract or duty

within this Act imposes an obligation of good faith in its performance and enforcement.”

15 Daniel R. Fischel, The Economics of Lender Liability, 131 YALE L. J. 140 (1989).

16 Indeed, legal suits based on good faith range over a wide variety of legal areas, including

commercial law (in particular as regards franchise, dealership, and distributorship agreements), banking law

(in particular, with reference to the liability of lenders), labor law, and so on.

9

Section 1-201(19), then, defines the obligation of good faith as “honesty in fact in the

conduct or transaction concerned.” In the original draft of the Code, however, the general

formulation of the good-faith duty also included a reference to “reasonable commercial

standards.”17 Only in the final version, it was decided to limit the application of the

reasonableness standard to specific transactions.18 Over the years, however, the number of

these specific transactions has been progressively expanded,19 with the result that it is

currently disputed which good-faith standard the Code actually endorses.

Scholars have fiercely debated when the general honesty-in-fact standard (also

referred to as the subjective test of good faith) should be superseded by the

reasonableness standard (also referred to as the objective test of good faith). In particular,

two dominants views have emerged. The first is that of the so-called communitarian

17 See Section 1-201(18) of the May 1949 draft of the U.C.C. (“’Good faith’ means honesty in fact in

the conduct or transaction concerned. Good faith includes good faith toward all prior parties and observance

by a person of the reasonable commercial standards of any business or trade in which he is engaged.”)

18 The ABA Section on Corporation, Banking and Business Law was among the most ardent opponents

to the adoption of the reasonableness standard in the final version of the U.C.C.. The participants at the

Section recommended, indeed, the elimination of such a standard. Finally, however, compromise prevailed

and the reasonableness standard was relegated to specific transaction, mostly transactions “among merchants”.

See Section 2-103(b) of the U.C.C. (good faith “in the case of a merchant” means “honesty in fact and the

observance of reasonable commercial standards of fair dealing in the trade.”) For some insights on the

controversy surrounding the enactment of the original version of the U.C.C. (1957), see JAMES J. WHITE AND

ROBERT S. SUMMERS, HANDBOOK OF THE LAW UNDER THE UNIFORM COMMERCIAL CODE § 14-6 at 563

(West, 2d. ed 1980).

19 The commercial reasonableness standard, originally applied only to Section 2-103 (b) of the code,

has been progressively extended to the following additional Sections of the U.C.C.: § 2A-103(3) (lease

transactions); § 3-103(a)(4) (negotiable instruments); § 4-104(c) (action of banks when dealing in negotiable

instruments); § 4A-105(a)(6) (fund transfer by banks); § 8-1029(a)(10) (investment securities), and § 9-

102(43) (secured transactions).

10

scholars,20 who have strongly opposed the honesty-in-fact standard for the reason that it

fails to address several forms of non-dishonest bad faith (such as conduct involving

recklessness or carelessness.)21 The second is that of the so-called abolitionists,22 who

have argued, instead, that the reasonableness standard is too vague and evolutionary a

standard to be of any practical guidance.23

Courts, however, seem not to have been much influenced by this academic dispute,

but rather have privileged one or the other of the Code’s standards depending on the

particular circumstances of the case. Some courts have, indeed, adhered to a subjective

test of good faith, which turns on the party’s actual belief that she was acting in bad faith

at the time of the alleged breach.24 Others, instead, have followed an objective test that

20 Communitarian scholars conceive of good faith as a super-eminent principle of law derived from

objective community standards of fairness and decency. The communitarian group includes, among others,

Professor Robert S. Summers, Professor Allan Farnsworth; Professor Steven Burton. See supra notes 1, 4.

21 See, e.g., Summers, Good Faith In General Contract Law, supra note 4, at 211-215; Robert

Braucher, The Legislative History of the Uniform Commercial Code, 58 COLUM. L. REV. 798, 812 (1958).

22 Abolitionists predicate the elimination of any reasonableness standard of good faith in favor of the

honesty-in-fact standard. Historically, the term was born out of the recommendation to the drafters of the 1949

version of the Code by the ABA Section on Corporation, Banking and Business Law. See supra note 18. The

most known abolitionist scholar is probably Clayton Gillette. See Clayton P. Gillette, Limitations on the

Obligations of Good Faith, 1981 DUKE L.J. 169.

23 Abolitionists not only claim that the reasonableness standard is a vague concept, which would only

increase litigation, but also that commercial standards may, in fact, evolve more quickly than those to which

courts adhere. The risk, thus, is that business evolution might be hindered by courts in the name of good faith.

See Walter D. Malcom, The Proposed Commercial Code, 6 BUS. LAW. 113, 128 (1951) (reporting the

recommendation to the Code drafters by the ABA Section on Corporation, Banking and Business Law).

24 See, e.g., Jackson v. State Bank, 488 N.W.2d 151, 156 (Iowa 1992); Schluter v. United Farmers

Elevator, 479 N.W. 2d. 82, 85 (Minn. Ct. App. 1992); Daniels v. Army Nat'l Bank, 822 P.2d 39, 43 (Kan.

1991); Brill v. Catfish Shaks of America, Inc., 727 F. Supp. 1035, 1040-41 (E.D. La. 1989); Adams v. First

State Bank, 778 S.W.2d 611, 614 (Ark. 1989); Karner v. Willis, 710 P.2d 21, 23 (Kan. 1985); Burton v.

11

examines whether a reasonable person would consider the conduct in question as honest

and justified.25 In some cases, courts have adopted both tests adding the objective

standard of reasonableness to the subjective standard of honesty.26 Ultimately, then, the

Code’s provisions–rather than fostering uniformity, as its drafters intended–27seem to

have left parties unable to predict the scope of their good-faith obligation.28

1.1.2. The Restatement Conceptualization

The Restatement (Second) of Contracts adopts a different approach to good faith,

although this may not be apparent at first sight. Section 205 states that “[e]very contract

imposes upon each party a duty of good faith and fair dealing in its performance and

Security Pac. Natl. Bank, 197 Cal. App. 3d. 972, 243 Cal. Rptr 277 (1988); Quest v. Barnett Bank, 397 So. 2d

1020, 1022 (Fla. Dist. Ct. App. 1981); Wohlrabe v. Pownell, 307 N.W.2d 478, 483 (Minn. 1981); Farmers

Coop. Elevator, Inc. v. State Bank, 236 N.W.2d 674, 678 (Iowa 1975); Fort Knox Nat'l Bank v. Gustafson,

385 S.W.2d 196, 200 (Ky. 1964).

25 See, e.g., Wilder v. Cody Country Chamber of Commerce, 868 P.2d 211, 220 (Wyo. 1994); Cenac v.

Murry, 609 So. 2d. 1257, 1272 (Miss. 1992); Richland Nat'l Bank & Trust v. Swenson, 816 P.2d 1045, 1051

(Mont. 1991); Warner v. Konover, 553 A. 2d 187, 191 (N.H. 1989); Hubbard Chevrolet Co. v. General

Motors Corp., 873, 876 (5th Cir.) cert. denied., 493 U.S. 978 (1989); Richards Eng'rs, Inc. v. Spanel, 745 P.2d

1031, 1033 (Colo. Ct. App. 1987); Mitchell v. Ford Motor Credit Co., 688 P.2d 42, 45 (Okla. 1984); Smith v.

Union State Bank, 452 N.E.2d 1059, 1064 (Ind. Ct. App. 1983); Universal C.I.T. Credit Corp. v. Shepler, 329

N.E.2d 620, 623-24 (Ind. Ct. App. 1975); Black v. Peoples Bank & Trust Co., 437 So. 2d 26, 29-30 (Miss.

1983); Sheppard Fed. Credit Union v. Palmer, 408 F.2d 1369, 1371 & n.2 (5th Cir. 1969).

26 See, e.g., Son’s of Thunder, Inc. v. Borden, Inc., 690 A.2d. 575, 587 (N.J. 1997); Petra Int'l Banking

Corp. v. First Am. Bank of Va., 758 F. Supp. 1120, 1142 (E.D. Va. 1991); Watseka First Nat'l Bank v. Ruda,

552 N.E.2d 775, 781 (Ill. 1990); Martin Specialty Vehicles, Inc. v. Bank of Boston (In re Martin Specialty

Vehicles, Inc.), 87 B.R. 752, 766 (Bankr. D. Mass. 1988) (applying Massachusetts law), rev'd on other

grounds, 97 B.R. 721 (D. Mass. 1989); Reid v. Key Bank, Inc., 821 F.2d 9, 15 (1st Cir. 1987); K.M.C. Co. v.

Irving Trust Co., 757 F.2d 752, 761 (6th Cir. 1985).

27 See U.C.C. § 1-102 (2) (c).

28 In similar terms, see Summers, supra note 1, at 119; BURTON & ANDERSEN, supra note 4, passim.

12

enforcement.” The provision is substantially equivalent to that of U.C.C. Section 1-203.

The official comment accompanying Section 205, however, makes clear that the

conceptualization of good faith endorsed by the Restatement is very different from that of

the Code. The comment specifies that good faith may have different meanings depending

on the particular context in which it applies.29 Thus, rather than mandate a specific

standard of conduct, the implied obligation of good faith “excludes a variety of types of

conduct characterized as involving ‘bad faith’ because they violate community standards

of decency, fairness, or reasonableness.”30

As in the case of the Code, scholars have been divided on the approach adopted by

the Restatement. Communitarians have supported it, claiming that the Restatement’s

open-ended approach prevents a larger number of bad faith behaviors. Abolitionists,

instead, have strongly criticized the Restatement’s view, claiming that it introduces

uncertainty. This risks endangering party autonomy allowing courts to override express

contract terms on the basis of their own conceptions of decency, fairness, or

reasonableness.

Thus, neither the Code nor the Restatement conceptualization seems to have been

of much guidance in providing parties, on one side, and courts, on the other, with a clear

good-faith parameter. On the contrary, the discrepancy between the two

29 Restatement (Second) of Contracts § 205 cmt. (a) (“The phrase good faith is used in a variety of

contexts, and its meaning varies somewhat with the context.”)

30 Id. The list of examples of bad faith reported in the comment to the Restatement includes, among

others, the evasion of the spirit of the deal, the lack of diligence and slacking off, willfully rendering only

substantial performance, the abuse of power to determine compliance, and interference with or failing to

cooperate in the other party’s performance. It must be emphasized that the Restatement’s list of example

substantially reproduces that elaborated by Professor Summers, in Summers, Good Faith In General Contract

Law, supra note 4, at 195. See also infra note 34.

13

conceptualizations has promoted inconsistent judicial responses and yielded widely

divergent evaluations of a party’s conduct depending on the different good-faith

approaches adopted by courts.31

1.2. SCHOLARLY CONCEPTUALIZATIONS

1.2.1. The Excluder Approach (or Contractual Morality Approach)

The good-faith analysis elaborated by Professor Summers is one of the earliest and

most influential.32 His excluder approach has been widely adopted by courts33 and

substantially influenced the conceptualization of good faith endorsed by the

Restatement.34 Summers was, in fact, the first to introduce the idea that no unifying

31 See Diamond & Foss, supra note 4, at 585-586.

32 The other influential early work on good faith is that of Professor Farsnworth. His main suggestion

is that the obligation of good faith in contractual relationships should be understood as a means to further the

parties’ agreement. Precisely, for Farnsworth, good faith requires “cooperation on the part of one party to the

contract so that another party will not be deprived of his reasonable expectations.” See Farsnworth supra note

4, at 669. The work of Professor Farnsworth, as that of Professor Summers, has largely influenced the drafting

of the Restatement.

33 See, e.g., Occusafe, Inc. v. EG&G Rocky Flats, Inc., 54 F.3d 618, 624 (10th Cir. 1995); Bank of

China v. Chan, 937 F.2d 780, 789 (2d Cir. 1991); Kedra v. Nazareth Hosp., 868 F. Supp. 733, 737 (E.D. Pa.

1994); Coca-Cola Bottling Co. v. Coca-Cola Co., 769 F. Supp. 599, 652 (D. Del. 1991); Kleiner v. First Nat'l

Bank, 581 F. Supp. 955, 960 n.5 (N.D. Ga. 1984); Larson v. Larson, 636 N.E.2d 1365, 1368 (Mass. App. Ct.

1994); Bourgeous v. Horizon Healthcare Corp., 872 P.2d 852, 856 (N.M. 1994); Somers v. Somers, 613 A.2d

1211, 1213 (Pa. Super. Ct. 1992); Garrett v. Bankwest, Inc., 459 N.W.2d 833, 845 (S.D. 1990); Carmichael v.

Adirondack Bottled Gas Corp., 635 A.2d 1211, 1216-17 (Vt. 1993).

34 Although Summers did not participate to the drafting of the Restatement, his work has been

acknowledged by Professor Braucher, who coordinated the Restatement’s enactment, to have largely

influenced the Restatement’s elaboration. See Summers, supra note 1, at 125. (quoting Professor Braucher: “I

am indebted … to Professor Summers …. He made considerable effort and collected this very large number of

cases in which judicial opinion has insisted on some obligation of good faith and fair dealing in the

14

meaning of good faith can be devised. For him, good faith “is best understood as an

‘excluder’ … which serves to [rule out] a wide range of heterogeneous forms of bad

faith,”35 including conduct that is neither fraudulent nor negligent. Hence, to identify

which conduct good faith imposes in a given context, one should focus on the kind of

bad-faith conduct that is excluded in that particular context and then derive the content of

good faith as the opposite of that bad-faith conduct.36

There is, however, a second reason for Summers’s argument in favor of an open-

ended conceptualization of good faith: to guarantee the substantive justice of contractual

relations. Good faith imposes on parties separate moral standards of conduct, which may

override the explicit terms of the contract if these do not satisfy requirements of decency,

fairness, or reasonableness.37 In this context, any attempt to provide a single, unifying

definition of good faith becomes reductive. Parties’ conduct, instead, must be evaluated

case by case by courts. Since they are situated “in the midst of the relevant circumstantial

performance and enforcement of contracts. And then he tried to categorize them, and I have borrowed heavily

from his classification scheme.”)

35 Summers, Good Faith In General Contract Law, supra note 1, at 196.

36 See Summers, supra note 1, at 126 (offering the following example to explain how the excluder

approach should be applied in practice:

[A] judge may say: “A public authority must act in good faith in letting bids.” And from the

fact or the language of the opinion it may appear that the judge is, in effect, saying: “The defendant

acted in bad faith because he let bids only as a pretense to conceal his purpose to award the contract

to a favored bidder.” It can be said that “acting in good faith” here simply means: letting bids without

a preconceived design to award the contract to a favored bidder.)

37 Summers, Good Faith In General Contract Law, supra note 4, at 197-200.

15

details,”38 courts are indeed in the best position to grant respect to the requirements of

contractual morality. 39

Summer’s moral approach to good faith is also reflected in the Restatement, which

makes express reference to the community standards of fairness, decency, and

reasonableness. Consequently, much of the criticism directed against the indeterminacy of

the Restatement applies to Summers’s analysis as well. Abolitionists have thus argued

that the excluder approach encourages judicial arbitrariness and jeopardizes freedom of

contract.40 More generally, it can be observed that the moralistic label attached to good

faith by Summers has long constituted one of the major arguments adduced to repudiate

the good-faith obligation and argue that it compromises economic efficiency.

1.2.2. The Foregone Opportunity Approach (or Economic Approach)

A second leading conceptualization of the duty of good faith is that elaborated by

Professor Burton,41 whose work has often been described as an economic analysis of

38 Id., at 264.

39 In Summer’s view, the duty of good faith is, thus, "a piece with explicit requirements of contractual

morality” Summers, Good Faith In General Contract Law, supra note 4, at 196. For a case endorsing

Summer’s contractual morality approach, see Koehrer v. Superior Court, 226 Cal. Rptr. 820, 828 (Ct. App.

1986) (stating that "the obligations stemming from the implied covenant of good faith and fair dealing are

imposed by law as normative values of society.")

40 See, e.g., Clayton P. Gillette, Commercial Relationships and the Selection of Default Rules for

Remote Risks, 19 J. LEGAL STUD. 535, at 650 ("Summers' advocacy of good faith as a means for excluding

certain conduct from acceptable commercial behavior appears to be predicated on the ability of and

desirability for judges to make ad hoc determinations of whether permitting specific commercial conduct

would be 'just."').

41 See, e.g., Burton, Common Law Good Faith, supra note 4; Burton, Good Faith Performance, supra

note 4; Burton, More on Good Faith Performance, supra note 4; BURTON &. ANDERSEN, supra note 4.

16

good faith, in contrast to Summers’s moral approach.42 For Burton, good faith is a means

to further the parties’ economic interests, rather than some indefinite moral values. It is

the instrument the law provides to ensure that a party does not attempt to recapture

opportunities foregone upon signing. In Burton’s view, at the time of contracting, each

party agrees to commit her resources to a particular purpose, foregoing the benefits which

could come from other uses of those resources.43 Good faith prevents that a party may

appropriate ex post these other benefits to the detriment of her counterparty. Whether an

opportunity has been foregone depends on the reasonable expectations of the parties, at

the time of contract formation, on both the benefits to be received under the contract and

the costs to be paid for those benefits.44

From Burton’s perspective, the duty of good faith serves its core function when one

of the parties has a discretionary power to set the terms of performance (such as the

quantity, price, or time).45 When the contract attributes discretion, there is in fact a higher

risk that one party may seek to recapture foregone opportunities. Thus, the obligation of

good faith prevents that a party may abuse her discretionary power to “refuse[] to pay the

expected cost of performance.”46 (For example, in requirements contract with a fixed

42 See, e.g., Emily M.S. Houh, The Doctrine of Good Faith in Contract Law: A (Nearly) Empty Vessel?

9, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=622982#PaperDownload, last visited

March, 7, 2005; Van Alstine, supra note 4, at 1254.

43 For instance, when a buyer enters into a requirements contract with a fixed price, she foregoes the

opportunity to fill her requirements by trading on the market, even if the market price falls below the

contract price. See BURTON & ANDERSEN, supra note 4, at 100-102

44 Burton, Common Law Good Faith, supra note 4, at 387.

45 Id., at 373. See also BURTON & ANDERSEN, supra note 4, at 45-50

46 Burton, Common Law Good Faith, supra note 4, at 373.

17

price, the obligation prevents that the buyer may abuse her power to terminate at will for

profiting of a sudden fall in the market price of the good.)

Although Burton shifted the good-faith analysis from a moralistic perspective to an

economic one, in his elaboration the echo of the traditional view of good faith is still

strong. For instance, rather than providing operational criteria for establishing the parties’

reasonable expectations, he seems to repose an almost unconditioned trust in the courts’

ability to identify such expectations, with the result that much of the economic rigor of his

elaboration is frustrated. 47 A more substantial criticism concerns the concept itself of

foregone opportunity. For Burton, by concluding a contract, the parties decide to pursue

some opportunities and forego others. To be chosen or foregone, however, an opportunity

must first be known. This contradicts behavioral studies on agents’ bounded rationality.48

Empirical evidence has long showed that parties, due to both intrinsic limits of cognition

and limited availability of information, do not know, nor can know, all the feasible

alternative actions open to them.49 Thus, it may well be that an opportunity, rather than

being foregone in favor of another, is simply unknown.

1.2.3. The Law-and-Economics Approach

47 In this terms, see Victor P. Goldberg, Discretion in Long-Term Open Quantity Contracts: Reining in

Good Faith, in 35 U.C. DAVIS L. REV. 319, 323 (2002) (stating that Burton fails to provide a framework to

“define the contours of good faith” and that, in fact, “once the analytical framework is understood, it is clear

that “good faith” does no work.”)

48 For a thorough discussion of the matter from a law-and-economics viewpoint, see Thomas S. Ulen,

Cognitive Imperfections and the Economic Analysis of Law, 12 HAMLINE L. REV. 385, 385-86 (1989). See

also, Melvin A. Eisemberg, The Limits of Cognition and the Limits of Contract, 47 Stan. L. Rev. 211, 214

(1995).

49 Among the pivotal works, see HERBERT A. SIMON, THEORIES OF BOUNDED RATIONALITY, IN

DECISION AND ORGANIZATION 161 (2d ed. 1986).

18

A rather different conceptualization of good faith is that developed by law-and-

economics scholars.50 Under this view, the obligation of good faith is considered an

implied contractual term that prohibits opportunistic behavior. The latter occurs when a

party “behaves contrary to the … [counterparty’s] understanding of their contract, but not

necessarily contrary to the agreement’s explicit terms, leading to a transfer of wealth” 51

from the counterparty to the opportunistic one. Rational actors, however, anticipate such a

risk and draft contracts that prevent parties from acting opportunistically. Still,

contractual specification is costly and time-consuming. Furthermore, contracts are

ontologically incomplete so that no matter the level of specification of their agreement,

parties are not able to foresee all possible opportunistic behaviors.52 Under this approach,

the implied obligation of good faith serves efficiency by establishing a general prohibition

50 On the concept of opportunistic behavior in the law and economics analysis, see, among the others,

RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 81 (3ed. 1986) (defining opportunism as taking

advantage of the other party’s vulnerability); VICTOR P. GOLDBERG, READING IN THE ECONOMICS OF

CONTRACT LAW (1989) (proposing a view similar to that of R. Posner); OLIVER WILLIAMSON, MARKET AND

HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATION (1975); Oliver Williamson, Transaction Costs: The

Governance of Contractual Relations, 22 J.L. & ECON. 233 (1979); OLIVER WILLIAMSON, THE ECONOMIC

INSTITUTION OF CAPITALISM 47 (1985) [hereinafter WILLIAMSON, ECONOMIC INSTITUTIONS] (defining

opportunism as “self interest seeking with guile”); George M. Cohen, The Negligence-Opportunism Tradeoff

in Contract Law, 20 HOFSTRA L. REV. 941 (1992) [hereinafter Negligence-Opportunism] (proposing a broad

definition of opportunism as “any contractual conduct by one party contrary to the other party’s reasonable

expectations based on the parties’ agreement, contractual norms, or conventional morality); George M.

Cohen, Interpretation and Implied Terms in Contract Law, BOUDEWIJN BOUCKAERT & GERRIT DE GEEST,

EDS., ENCYCLOPEDIA OF LAW AND ECONOMICS (Edward Elgar, 2000); Timothy J. Muris, Opportunistic

Behavior and the Law of Contract, 65 MINN. L. REV. 521 (1981) (considering opportunism as the transfer of

wealth from one party to the other contrary to the parties’ understanding).

51 Muris, supra note 50, at 521.

52 The theory of incomplete contract, then, may serve as another argument against Burton’s

conceptualization of foregone opportunities. See supra text accompanying notes 48-49.

19

against acting opportunistically that is automatically incorporated in every contract. It

saves parties the cost (and the difficulty) of expressly bargaining for specific contractual

provisions against the risk of opportunistic behavior.53

As compared to Burton’s economic approach, the law-and-economics

conceptualization of good faith shifts the focus of attention from the welfare of

individuals to that of society as a whole. Law-and-economics scholars are not concerned

with the expropriation of wealth of the individual party in itself. Rather differently,

opportunism is viewed as bad because it impairs the value maximization of contractual

exchanges, by increasing specification cost.54 Hence, any residual moralistic idea of good

faith is dismissed under this view.

The law-and-economics approach to good faith raises, however, several difficulties.

From a practical viewpoint, determining what must be deemed opportunistic behavior can

be very complicated, in particular when a party is contractually given some discretionary

53 The opinion of Justice Posner in Market Street Assocs. Ltd. Partnership v. Frey, 941 F.2d 588, 595

(7th Cir. 1991) represents probably one of the clearest description of the law-and-economics (opportunistic-

behavior) approach to good faith.

The concept of the duty of good faith like the concept of fiduciary duty is a stab at

approximating the terms the parties would have negotiated had they foreseen the circumstances that

have given rise to their dispute. The parties want to minimize the costs of performance. To the extent

that a doctrine of good faith designed to do this by reducing defensive expenditures is a reasonable

measure to this end, interpolating it into the contract advances the parties' joint goal.

… Good faith is a compact reference to an implied undertaking not to take opportunistic advantage in

a way that could not have been contemplated at the time of drafting, and which therefore was not

resolved explicitly by the parties."

54 Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 VA. L. REV. 1089, 1090

(1981); Fischel, supra note 15, at 152; Muris, supra note 50, at 524.

20

power. Not only can opportunism take subtle forms,55 but when a party is granted

discretion, drawing a line between opportunistic and non-opportunistic behavior can

become extremely difficult.56 Moreover, discretionary powers are often granted to curb

the risk that a party may behave opportunistically.57 Thus, invoking good faith to restrict

such powers might well jeopardize their deterrence effect, with the result that economic

efficiency might be impaired, not enhanced.58

1.2.4. The New-Formalism Approach

The most recent elaboration of good faith is due to that school of thought that some

have labeled the “new formalism”,59 being it predicated on a return to the classical

55 Muris, supra note 50, at 525. (arguing that opportunism can take subtle forms; i.e., it can be difficult

to detect, easy to mask as legitimate conduct, and discoverable only at a high cost.)

56 Fischel, supra note 15, at 141 (discussing the interpretative problem of the duty of good faith and

arguing that “distinguishing opportunistic from non-opportunistic behavior can be very complicated if not

impossible.”) For an attempt of distinguishing opportunistic from non opportunistic behavior, see Varouj A.

Aivazian, Michael J. Trebilcock & Michael Penny, The Law of Contract Modifications: The Uncertain Quest

for a Benchmark of Enforceability, 22 OSGOODE HALL L.J. 173 (1984).

57 Take the example of lender/borrower relationships and, in particular, the case of the lender’s right to

discretionally decide whether to advance further funds. The function of this power is clear. The lender, by

having the right to decide whether disbursing additional funds, will have control on the borrower’s

performance. The borrower, in turn, will be induced to perform well in order to obtain further advances.

58 Fischel, supra note 15, at 142 (stating that expansive interpretation of the duty of good faith to

control opportunistic behavior may impair the deterrence effect of discretionary powers); see also Goldberg,

supra note 47, at 322-323.

59 See David Charny, The New Formalism in Contract, 66 U. CHI. L. REV. 842, 842-886 (1999)

(coining the appellative and affirming: ”[w]e are now in the midst of a third phase, a phase of ‘anti-formalism’

that seek to discredit and displace Llewellyin’s claim to found commercial law in immanent commercial

practice.”) Among the most prominent neo-formalist scholars, see Goetz & Scott, supra note 54; Alan

Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. PA. L. REV. 595 (1995)

[hereinafter, Schwartz & Scott, The Political Economy]; Alan Schwartz & Robert E. Scott, Contract Theory

21

formalistic approach of the late nineteenth and early twentieth century.60 Very much as in

the era of classical formalism, but through the modern language of economic analysis of

law, the stalwarts of the new formalism emphasize the importance of legal certainty and

predictability. This leads them to support a textualist approach to contract interpretation

and to consider every contractual power as a presumptively absolute and unrestricted

power. On this view, there is not much room left for good faith, but for cases that closely

resemble fraud. Neo-formalists urge that courts adopt a passive approach to contract

interpretation and refer to a minimum evidentiary base, basically solely the contract.61

Judicial intervention grounded in good faith is radically dismissed as intrinsically

broadening the evidentiary base of interpretation and, therefore, compromising legal

certainty.

Several arguments justify this restrictive approach. Among these, one seems

particularly relevant with respect to the application of the good-faith obligation: the

asymmetry of information between parties and courts. The argument is essentially based

on freedom of contract and economic efficiency. Parties are in the best position to assess

the relative costs and benefits of their relationship and, thereby, distribute contractual

entitlements to maximize the value of their exchange. Because courts do not have the

and the Limits of Contract Law, 113 YALE L.J. 541 (2003) [hereinafter Schwartz & Scott, Contract Theory];

Robert E. Scott, The Case for Formalism in Relational Contract, 94 NW. U. L. REV. 847, 864-65 (2000)

[hereinafter Scott Case for Formalism]; Robert E. Scott, A Theory of Self-Enforcing Indefinite Agreements,

103 COLUM. L. REV. 1641 (2003) [hereinafter Scott, A Theory].

60 See supra note 5. See also Van Alstine, supra note 4, at 1224 (claiming that a renewed emphasis on

formalism in the interpretation of legal texts, statutes, and treaties characterized the 1990s legal scholarship.)

61 See, e.g., Schwartz & Scott, Contract Theory, supra note 59, at 572 (individuating the evidentiary

base to be allowed in a strict textualist approach in (i) the contract, (ii) an English language dictionary, and

(iii) the interpreter's experience and understanding of the world.

22

same information that parties do,62 any judicial intervention that does not conform to the

letter of the contract risks altering the parties’ distribution of rights and, ultimately,

reducing the expected value of their exchange.63 Where court intervention is based on

such a vague concept as good faith, the argument goes, such a risk becomes almost a

certainty. An active role of courts, based on the implied obligation of good faith, simply

produces errors. This leads parties to misallocate resources in an attempt to reduce the

likelihood of court’s mistake and thereby results in a reduction of the overall wealth.

1.3. A DIFFERENT PERSPECTIVE

One common feature brings together the otherwise disparate conceptualizations of

good faith outlined above: they all focus on courts as the primary recipients of their

guidelines on the application of good faith. Due to its intrinsic vagueness, good faith

raises, indeed, an interpretative problem. When courts must evaluate parties’ conduct

pursuant to such a requirement, rather than to an express contract term, it might become

more difficult for them to find the “correct answer”.64 The latter is the solution to “a

particular contracting problem that the parties intended to enact” 65 when they concluded

the contract. Hence, each of the above conceptualizations of good faith attempts to

62 See, e.g., Alan Schwartz, Relational Contracts in the Courts: An Analysis of Incomplete Agreements

and Judicial Strategies, 21 J. LEGAL STUD. 271, 277, 280 (1992) (affirming that courts, due to the non-

verifiability of parties’ information, are not able to enforce value-maximizing terms and suggesting that a

literal approach to contract interpretation would be more efficient). In similar terms, see Scott, Case for

Formalism, supra note 59, at 864-65.

63 See, e.g., Schwartz, supra note 62, 271; Schwartz & Scott, Contract Theory, supra note 59, at 549;

Schwartz & Scott, Political Economy, supra note 59, passim.

64 The expression is borrowed by Schwartz & Scott, Contract Theory, supra note 59, at 568.

65 Id.

23

identify the judicial interpretative stile of the good-faith obligation that is more likely to

yield the “correct answer”.

In this work, I propose a different approach. I claim that the interpretative stile that

courts should adopt in enforcing contracts should be determined by parties themselves.

Parties have more and better information on the substantive terms of their exchange than

courts do. Indeed, the parties’ information tends to be “observable but not verifiable” by

third parties. This means that courts may observe information, but cannot verify its

existence at reasonable costs and with reasonable accuracy.66 In turn, they are unable to

enforce the parties’ obligations on the basis of such information.67

For neo-formalists, the above considerations on court error mandate the exclusion

of the good-faith requirement and, therefore, of the good-faith interpretative regime, since

the risk of judicial error increases when a contract includes good faith and courts adopt a

contextual interpretation. I argue, instead, that the information asymmetry between parties

and courts simply implies that parties are in a better position to devise the interpretative

stile which is more likely to yield the “correct answer” (as above defined). In practice,

66 Id., at 605 (making the example of the employer who knows that his employees sometimes shirk.

The information of the employer, however, will not be directly verifiable by the court. Thus, for the employer,

it might be expensive to prove that a particular employee shirked twenty percent of the time. Due to the cost

of making the information verifiable, the employer might well not be able “to use it” in front of the court.)

67 See Karen Eggleston, Eric A. Posner, & Richard Zeckhauser, The Design and Interpretation of

Contracts: Why Complexity Matters, 95 NW. U. L. REV. 91, 119-20 (2000) (considering the case of a long-

term contract for the delivery of some perishable goods, in which the parties agree that the goods to be

delivered must be of some standard quality. At a certain point the buyer refuses the goods, claiming that

their quality is substandard. By the time the court settles the dispute, the court will most likely be unable to

verify the quality of the goods because of their perishable nature. And, even if the goods had not yet

perished, a court might not be able to distinguish a good of standard quality from one of substandard quality

due to lack of specific expertise.)

24

this means that parties themselves should decide whether to include or exclude good faith

in their contract. They are, indeed, in the best position to evaluate if the inclusion of such

a requirement is likely to be beneficial or detrimental.

2. IS THE OBLIGATION OF GOOD FAITH EFFICIENT?

2.1. GOOD FAITH AS A MEANS TO DETER OPPORTUNISM ARISING FROM

CONTRACTUAL INCOMPLETENESS AND TRANSACTIONAL INSECURITY

A commonly acknowledged rationale for the obligation of good faith lies in the

impossibility of writing complete contracts.68 Indeed, if parties were able to foresee all

present and future contingencies and to negotiate contractual provisions at no costs, there

would be no justification for the existence of the duty of good faith. Any aspect of the

contractual relationship would be governed by express contractual provisions, and any

potential breach of the parties’ agreement would be explicitly regulated therein.

Yet, complete contracts do not actually exist. Parties are unable to specify all

contingencies, and specification is not costless. Consequently, contracts are ontologically

incomplete. In addition, most contractual relationships are affected by transactional

insecurity.69 This condition describes the risk parties face when, as is commonly the case,

their contractual performances takes place sequentially. Because one of the parties carries

out her side of the exchange before the other, there is an intrinsic risk that the

68 See Fischel, supra note 15, at 141. (“the rationale for imposing a duty of good faith on lenders

relates to the impossibility of drafting a contract covering every possible contingency.”); Coehn, Negligence-

Opportunism, supra note 50, at 954; WILLIAMSON, ECONOMIC INSTITUTIONS 45.

69 The expression transactional insecurity is borrowed by Professor Kronman. Anthony T. Kronman,

Contract Law and the State of Nature, 1 J.L. ECON. & ORG. 5, 6 (1985).

25

counterparty may default on her own contractual obligations and thereby deny the

performing party the bargained-for benefit.70 Consider, for instance, lending agreements.

The performance of the bank (i.e., the disbursement of the loan) is entirely fulfilled before

that of the borrower (i.e., the payment of interests and principal). Hence, the bank will be

subject to an intrinsic risk that the borrower, after receiving the loan, will fail to comply

with the obligations established by the contract and thus deprive the bank of its

bargained-for benefit.

From an economic viewpoint, the concept of transactional insecurity is a

consequence of the character of parties’ investments in contractual relationships. Many of

these investments are situation-specific and tend to become more and more specialized

over time. As a result, after a specific investment is made (sunk), parties bear relatively

high transaction costs to redeploy it to alternative uses (where this is even possible).71 The

difference between the investment’s value in the intended use and its next best available

use gives rise to a stream of so-called quasi-rents, which may be expropriated by the other

70 Id. See also POSNER, supra note 50, at 81; Richard A. Posner, The Law & Economics of Contract

Interpretation, AMERICAN LAW & ECONOMICS ASSOCIATION ANNUAL MEETING, PAPER 56, at 3 (2005)

[hereinafter Posner, Contract Interpretation] (stating that “if exchange were simultaneous … there would be

little need either for contracts or for legal remedies for breach of contract”); Wisconsin Knife Works v.

National metal Crafters, 781 F.2d 1280, 1285 (7th. Cir. 1986) (Posner, J.) (holding that because the

performance of the parties to a contract is not typically simultaneous, one party may find herself at the mercy

of the other unless the law of contracts protects her. Preventing that one party may take advantage of the

vulnerabilities to which sequential performance (i.e., transactional insecurity) gives rise is the most important

thing that law does to facilitate exchanges.)

71 See WILLIAMSON, ECONOMIC INSTITUTIONS, supra note 68, at 55, 115. (Williamson calls this

condition “asset specificity”. With this expression, he identifies, in particular, “durable investments that are

undertaken in support of particular transactions, the opportunity cost of which investments is much lower in

best alternative uses or by alternative users should the original transaction be prematurely terminated.”)

26

party; this is the so-called hold-up problem.72 Consider, for instance, a contract for the

supply of automobile bodies to a particular car company. In such a case, the supplier will

need to make a specific investment to fulfill her contractual obligations; for example, she

will need to buy tailored stamping machines and dies.73 As a result, the supplier will be

subject to the risk that the car manufacturer, by threatening to terminate the contract

unless she agrees to a price reduction, may expropriate her specific investment’s quasi-

rents.

Because contracts are incomplete and affected by transactional insecurity,

unforeseen events may arise enabling one party to act opportunistically to deprive the

other of her expected contractual benefits, that is, the (expected) economic surplus

without which a rational actor would not have concluded the contract (at least, not with

the same economic conditions).74 The implied obligation of good faith, thus, can be

72 The seminal work on the matter is the 1978 article of Benjamin Klein, Robert G. Crawford &

Armen A. Alchian. See Benjamin Klein, Robert G. Crawford & Armen A. Alchian, Vertical Integration,

Appropriable Rents, and the competitive Contracting Process, 21 J.L. & ECON. 97 (1978); see also Benjamin

Klein, Transaction Cost Determinants of “Unfair” Contractual Arrangements, 70 AM. ECON. REV. 356, 358-

359 [hereinafter Klein, “Unfair” Contractual Arrangements]; Benjamin Klein & Keith B. Leffler, The Role of

Market Forces in Assuring Contractual Performance, 89 J. POL. ECON. 615 (1981). In general terms, a hold-

up occurs “when a transactor … decides it is wealth-maximizing to take advantage of contractual

incompleteness to expropriate the rents on the specific investments made by its transacting partner”. Benjamin

Klein, Hold-Up Problem, in THE NEW PALGRAVE DICTIONARY OF LAW AND ECONOMICS, VOL. II, 241 (2001).

73 The example in the text is based on facts of the well known case of Fischer Body-General Motors.

74 It must be emphasized, however, that the expected contractual benefits of the parties depend on the

order of preferences that they have agreed. For example, consider an agreement for the construction of a

country residence, in which one of the contractual provisions specifies the determined brand of pipes to be

used in the construction of the house. The orally-agreed purpose of such a provision is to ensure that only

pipes of a certain qualities are used in the construction. In this case, the homeowner cannot claim to have

been expropriated his expected contractual benefits if the builder installs pipes of the same functional

27

defined as the rule of law that prevents each contracting party from taking advantage of

the contract’s incompleteness and transactional insecurity to expropriate the

counterparty’s expected contractual benefits.75

In accordance with most law-and-economics scholars,76 I see the expropriation of

the parties’ expected contractual benefits more as a problem of efficiency than one of

justice. I assume, indeed, that the proper normative goal is to maximize the ex ante value

of contractual relationships. Opportunistic behavior, thus, is inefficient because parties, in

their attempt to deter such behavior, write longer, more complete, contracts and bear

higher specification cost. In turn, the parties’ net gains from the contractual relationship

are reduced,77 and so is the overall welfare of society. Unlike law-and-economics

scholars, however, I argue that whether good faith is efficient depends on the trade-off

between the specification cost that it permits the parties to save and the expected cost that

they bear under the good-faith interpretative regime due to judicial error.

2.2. THE TRADE-OFF IMPLIED BY THE OBLIGATION OF GOOD FAITH

quality of the one indicated in the contract, but of a different brand. Things, however, would be different if

the homeowner was a pipe fanatic and clearly manifested her fanaticism to the builder.

75 Cf. Market Street Assocs. Ltd. Partnership v. Frey (J. Posner), supra note 53, at 588, 594-5. See also

Original Great Am. Chocolate Chip Cookie Co. v. River Valley Cookies, Ltd. 970 F.2d 273, 280 (7th Cir.

1992) (Posner, J.)

76 See Schwartz & Scott, Contract Theory, supra note 59, at 546 (arguing that “efficiency is the only

institutionally feasible and normatively attractive goal”); Goetz & Scott, supra note 54, at 114 (proposing the

obligation to produce at the joint maximization volume as the meaning of the best-effort clause); Muris, supra

note 50, at 555, fn. 91 (arguing that the problem with opportunistic behavior is not the loss of wealth of the

expropriated party, but rather the depreciation of the parties’ exchange); Eggleston, Posner, & Zeckhauser,

supra note 67, at 93 (affirming that the normative goal of both law-and-economics and the economics of

contract is the maximization of the contractual value.)

77 Cf. Goetz & Scott, supra note 54, at 1116-1117.

28

Law-and-economics scholars claim that the obligation of good faith is efficient

because it saves parties the cost of drafting lengthy anti-opportunism provisions.78 This

view, however, overlooks the different deterrent effect in controlling the risk of

opportunism that express contractual terms and implied obligation of good faith have.

When a particular contingency is foreseeable,79 but costly to negotiate for and specify in

writing, parties have two choices at their disposal. They can either include in the contract

an express term that provides for their respective rights and duties at the occurrence of

that particular contingency (“specify the contingency”) or, to this end, rely instead on

their implied obligation of good faith (“address the contingency through good faith”).

For law-and-economics scholars, the decision between these two options depends

on the specification cost implied by a particular contingency80 and on the likelihood that

such a contingency will materialize.81 Under this view, when specification cost is high

78 See POSNER, supra note 50, at 89-94, 97-98. See also Market Street Assocs. Ltd. Partnership v. Frey

(J. Posner), supra note 53, at 596.

79 For foreseeable contingency, I intend an “elementary event” which the parties are able to foresee at

reasonable costs. See Pierpaolo Battigalli & Giovanni Magli, Rigidity, Discretion, and the Cost of Writing

Contracts, 92 AM. ECON. REV. 798, 802 (2002). Unforeseeable contingencies will be discussed infra, Par.

3.4.1. (A)

80 See, e.g., RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 104-05 (5th ed. 1998); Posner,

Contract Interpretation, supra note 70, at 4. See also, in general, Steven Shavell, Contracts, in 1 THE NEW

PALGRAVE DICTIONARY OF ECONOMICS AND THE LAW 436, 438 (Peter Newman ed., 1998); Ronald A. Dye,

Costly Contract Contingencies, 26 INT'L ECON. REV. 233, 234 (1985).

81 See Eggleston, Posner, & Zeckhauser, supra note 67 at 108. (Observing that “negotiation costs can

be an important determinant of contract form … . [ and that this] is illustrated by cases involving low-

probability contingencies.”) As an example of low-probability contingency, the authors report the

international incidents occurred in the closing of the Suez Canal. None of the contracts for shipping goods

from East Asia to West contained a clause allocating the parties’ obligations in the event of closure of the

Suez Canal because nobody anticipated that the Suez Canal might be closed.) Moreover, it must be

29

and/or the likelihood a contingency will materialize is low, parties should address such a

contingency through good faith. I argue, however, that parties bear a different risk when

they decide to address a contingency through good faith rather than through an express

contractual provision.

When parties specify a contingency, they do not bear a risk of opportunism as to

that contingency, but rather a risk of contractual breach. In other words, being the

contract complete in relation to that particular contingency, the condition necessary for

opportunism to arise is missing.82 In contrast, when parties decide to address a

contingency through good faith, they still bear a risk of opportunistic behavior, because

of:

(i) the less precise definition of their respective entitlements that derives from

addressing a contingency through good faith (as above defined), rather than specifying it

through an express contractual provision; and,

(ii) the increased risk that the enforcing court may fail to find the “correct answer”,

due to both (a) the contract’s indeterminacy on the allocation of the parties’ entitlements,

and (b) the exercise of judicial discretion resulting in re-distributions of the parties’

entitlement.

When parties decide to address a contingency through good faith, they might face

uncertainty as to the precise content of their rights and duties at the occurrence of that

contingency. Consider, for instance, a sales contract, in which the parties agree that the

emphasized that whether parties will decide to invest resources for specifying a low-probability contingency

will depend much by the expected cost they are likely to bear if such a contingency materializes.

82 See supra text accompanying note 75.

30

good must be delivered on a certain date, but do not specify the delivery time. They

agree, instead, that the delivery must take place “in good faith”.

Then, it seems reasonable to say that both the buyer and the seller would agree that

a 3 a.m. delivery does not comply with the good-faith requirement they have agreed

upon.83 The situation, however, would not be so clear in the event of a 5.45 a.m. delivery.

Would the buyer be entitled, in such a case, to claim bad faith on the seller’s side and

refuse the good? Or would it be the seller, instead, who can claim the bad faith of the

buyer if the latter refuses the good? Thus, the parties’ obligation of good faith might not

effectively prevent the risk of opportunism because it might be not clear what constitutes

opportunism. Indeed, the parties might honestly disagree on the allocation of their

respective obligations. The seller could believe that it is within her right to deliver the

good at 5.45 a.m. On the other side, the buyer could believe that she has the right to

refuse the good because of what she considers an absurd time of delivery. Whether it

would be the buyer or the seller to be right is not clear. Hence, it would be difficult to say

if the buyer or the seller is behaving opportunistically.

This simple example shows why addressing a contingency through good faith is not

equivalent to specifying that contingency in writing. Due to the less precise definition of

the parties’ entitlements, under a good faith regime the risk that the court may fail to find

the “correct answer” increases. This, in turn, might induce parties to neglect their

obligation of good faith and behave opportunistically when an unspecified contingency

(i.e., a contingency that is not addressed by an express contract term) materializes. In

83 Indeed, the delivery of the good at 3 a.m. would more likely constitute a case of material breach of

the contract than one of opportunistic behavior. For a thorough discussion of the, often subtle, difference

between opportunism and material breach, see BURTON & ANDERSEN, supra note 4, at 197-267.

31

practice, a party will bet on the judicial outcome whenever the payoff she expects from

behaving opportunistically is higher than the expected cost of the judicial outcome.84

In addition, when a contract includes good faith, courts enjoy discretion in its

interpretation. The literal approach mandates the enforcement of contracts as written. In

the good-faith interpretative regime, instead, the court is free to decide what an express

contract term means. This discretion may lead to a reallocation of the contractual

entitlements agreed upon by they parties.

The foregoing discussion shows that the specification-cost test proposed under the

traditional law-and-economics approach is not sufficient to prove the efficiency of good

faith. I claim, indeed, that the efficiency question depends on the trade-off between (i) the

specification costs that the rule permits the parties to save; and (ii) the expected cost they

bear in a good-faith interpretative regime due to court error (the “good-faith trade-off”).

From this perspective, whether parties are better off by including or excluding good faith

is a matter of transaction-cost analysis.

3. THE TRANSACTION-COST ANALYSIS OF GOOD FAITH

3.1. CONTRACTUAL TRANSACTION COSTS

To solve the good-faith trade-off means to understand whether the inclusion of a

good-faith requirement in the parties’ agreement ultimately increases or reduces C, the

84 See A. Mitchell Polinsky & Steven Shavell, Legal Error, Litigation, and Incentive to Obey the Law,

5 J. L. ECON. & ORG. 99, 99-101 (1989). See also Louis Kaplow & Steven Shavell, Accuracy in the

Determination of Liability, 37 J. L. & ECON.1, 10-11 (1994).

32

contractual transaction costs.85 In a simplified description, C comprises; (i) the parties’

aggregate specification cost; and (ii) the parties’ aggregate expected cost deriving from

the incompleteness of the contract.

(i) The contract’s specification cost, S, includes (a) the parties’ aggregate cost for

figuring out and negotiating contingencies (the “negotiation cost”); (b) the parties’

aggregate cost for drafting contractual provisions that allocates the parties’ entitlements

as to each negotiated contingency (the “drafting cost”). Such a cost is basically a function

of the time the parties consume to write the relevant contractual provisions; and (c) the

parties’ aggregate cost for the lawyers eventually involved in the negotiation and drafting

process (the “legal cost”). Moreover, it is worth observing that S is increasing in n, the

number of contingencies specified by the contract.

(ii) The parties’ aggregate expected cost arising from the incompleteness of the

contract, D, is the cost parties expect to bear in the course of their contractual relationship

due to the impossibility of writing complete contracts. In particular, D includes:

(a) the (aggregate expected) inadequacy cost parties bear for specifying a

number of contingencies that no longer reflect the external state in an adequate manner.

Since parties specify only some among all foreseeable contingencies, they face a risk that

the contractual allocation of their entitlements might prove not sufficiently contingent on

85 Cf. Posner, Contract Interpretation, supra note 70 (carrying out a more general cost-benefit analysis

on contract interpretation and defining contractual transaction cost as “as obstacles to efforts voluntarily to

shift resources to their most valuable use.”)

33

the external state over time. Indeed, unspecified contingencies might materialize which

make the original allocation of the parties’ entitlements no longer efficient.86

Consider, for instance, the delivery example. Suppose that, on the day before the

agreed date of delivery, the seller concludes a contract with a second buyer living nearby

the first. Delivery to the second buyer is to occur two days after the first delivery. The

seller lives far away from the two buyers and, therefore, she bears significant transport

costs. Thus, for her, it would be less costly to deliver the two goods together. However,

she might be not able to reach the first buyer in time to propose her a delayed delivery

date. (Remember that the seller concludes the second contract the day before the delivery

date of the first.) In turn, fearing that the buyer might not accept her proposal, she would

deliver the good on the agreed date. Assuming that the seller was willing to offer the

buyer a discounted price in exchange for a delayed delivery, both parties will be worse

off. The cost they bear is an example of inadequacy cost;

(b) the (aggregate expected) opportunism cost parties bear due to the risk of

opportunistic behavior of one toward the other when an unspecified contingency

materializes.

Consider, for instance, the case of an installment sales contract that expressly

requires payment on delivery, failing which the seller is entitled to terminate the contract.

Since the first of several deliveries, however, the buyer fails to pay on delivery and,

instead, mails a check to the seller a few days later each delivery takes place. After

86 In this work, I use the expression “inadequacy cost” for what is ordinarily known as an “opportunity

cost” to avoid confusion with the “opportunism cost”. On opportunity cost, see Clifford W. Smith Jr. & Jerold

B. Warner, On Financial Contracting-An Analysis of Bond Covenants, 7 J.FIN. & ECON. 117 (1979).

(analyzing, among other things, the effect of rigid contractual terms in bond indentures). For an analysis of

“opportunity cost” closer to the one proposed in this paper, see Battigalli & Magli, supra note 79, at 799.

34

several deliveries (and delayed payments accepted by the seller), the market price of the

good far outstrips the contract price. Then, the seller terminates the contract claiming that

the buyer’s failure to pay on delivery entitles him to invoke the termination clause.87 The

cost the buyer suffers for the termination of the contract is an example of opportunism

cost of the contract.88

It must also be emphasized that D: (i) is decreasing in n. The inadequacy cost is

decreasing in n because the higher the number of contingencies specified by the contract,

the less likely that an unspecified contingency might materialize which makes the original

allocation of the parties’ entitlements no longer efficient. The opportunism cost is also

decreasing in n because the more contingencies the contract specifies, the lower is the risk

that the parties might behave opportunistically without breaching the contract; (ii) is

increasing in e, which indicates the error the court might make in interpreting the

contract; i.e., the enforcing court’s failure to find the “correct answer”.

To test the efficiency of good faith means, practically, to verify what interpretative

regime (i.e., literal or contextual) must be preferred in contract interpretation. Let’s then

start by considering a contract that does not include good faith and, therefore, mandates

the application of a strict literal interpretative regime. Under such a regime, the

contractual transaction costs the parties bear can be expressed as ( ) ( )C S n D n= + . e, the

court’s error can be assumed equal to zero.

87 The example in the text is based on the facts of Baker v. Ratzlaff, 1Kan App. 2d 285, 564 P.2d 153

(1977).

88 Of course, the example assumes (on the basis of the actual facts of the case cited above) that the

delayed payments did not jeopardize the seller’s economic interest.

35

Under a strict literal interpretative regime, in fact, the contract is considered

complete. If the parties’ entitlements as to the occurrence of a particular contingency are

not specified, the parties cannot claim any right as to that contingency (i.e., the

entitlement does not exist). From this perspective there is no room for court error because

everything the court must do to find the “correct answer” is to read the contract.89 Take

the delivery example. If the parties do not specify a delivery time, the good can be

delivered on the specified date at any time, even at 3 a.m.

Thus, under a strict literal interpretative regime, the parties’ expected cost from

incompleteness is exclusively function of the number of contingencies they specify.

However, parties do not specify all foreseeable contingencies, since the cost of doing so

89 See Eric Posner, A Theory of Contract Law Under Conditions of Radical Judicial Error, 94 NW. U.

L. REV. 749, 752-753 (reporting the classical theory of contract interpretation and complete contracts.) Under

the “classical” (or “Willistonian”) or "textualist," theory of interpretation, contracts have "plain meanings"

that are apparent to judicial interpreters. See supra notes 5, 60. See also Schwartz & Scott, Contract Theory,

supra note 59, at 572.

For neo-formalists the contract is, indeed, complete. The logic is the following: (i) the contract

provides for general rules; (ii) specification creates special rules which are state-contingent; (iii) if a non-

specified contingency occurs, the contract is not incomplete because the general rules apply.

Suppose, for instance, that among the general rules, there are the price x and the good y. In one special-

set of rule is instead provided that, should a occur, the price will be x+b and the good y+d. Thus, the special

set modifies the general rules. Indeed, under the general rules, the seller has the right to x and bears the

obligation y; the buyer has the right to y and bears the obligation x. Under the contingency a, the seller has the

right to x+b, and bears the obligation y+d. Then, should c, a contingency similar to a, occur, what is the

applicable rule? A neo-formalist court would always apply the general rule because is what the contract says

and the parties entitlements cannot be ex post modified by the adjudicator. On the contrary, a liberal court

might apply the special set “in the name of good faith”, if she “thinks” that the application of the special set is

what the parties would have agreed upon. However, the parties might in fact want the court to apply the

general rule. By interpreting the contract “in good faith”, the court ends up modifying the parties’

entitlements, not necessarily in the most efficient way. On the contrary, under the literal interpretative

approach, the court could not make mistakes; she would just need to apply the contract as written.

36

could well outweigh the expected cost deriving from contractual incompleteness. Rational

parties specify n , the number of contingencies that minimizes C, that is, they specify

additional contingencies until the marginal expected cost deriving from the contract’s

incompleteness equals the marginal cost of specification.90

3.1.1. The Continuum between Rigid and Discretionary Contractual Terms

The contractual provision of discretionary terms is one of the devices that parties

adopt to reducen and, therefore, minimize C. Discretionary terms serve, indeed, two

different functions. First, by attributing discretion to one of the parties to set the terms of

performance, they adapt the original allocation of contractual entitlements to changing

external states (i.e., the flexibility function).91 This, in turn, reduces the number of

contingencies parties must specify for containing the expected inadequacy cost. Consider,

for instance, requirements contracts, in which the buyer commonly has discretion to vary

the quantity of the orders in accordance with her operational needs (e.g., production

process modifications, variation in product demand, etc.) Doing so may yield a result

more suitable to the parties’ contractual ends, by allowing them to adapt the terms of their

agreement to the occurrence of unspecified contingencies, than would providing for

90 Analytically, to find the optimal number of contingencies, n , we differentiate ( ) ( )C S n D n= + ,

with respect to n . Thus, we have dC dS dD

dn dn dn

= + . Setting the derivative equal to zero, n is yielded at

dS dD

dn dn

= !

91 Cf. Goldberg, supra note 47, at 321 (discussing, in particular, the attribution of discretionary powers

in requirements and other long-term contracts.)

37

costly supply and demand contingencies.92 Similarly, in the delivery example, if the

contract had given the seller some discretion about the delivery date, both parties could

have saved inadequacy cost.

In addition, attributing discretion to one of the party may serve as a deterrent

against the risk of opportunistic behavior of the other party, usually the one that is more

likely to behave opportunistically due to the transactional insecurity affecting the contract

(i.e., the deterrence function).93 This, in turn, reduces the number of contingencies that

must be specified to contain the expected opportunism cost. Hence, not only the

discretion-exercising party benefits from the inclusion in the contract of discretionary

terms. Both parties, in fact, save specification cost. In addition, the party who is not

attributed discretion ordinarily benefits from a reduction in the price of the discretion-

exercising party’s performance. Consider, for instance, the case of lending agreements.

The attribution to the bank of a power to accelerate the loan, not only reduces the risk of

the borrower’s opportunistic behavior due to contract’s transactional insecurity,94 but also

yields a lower interest rate.

The provision of discretionary terms, however, may have a reverse effect. The

discretion-exercising party might abuse her power to expropriate the counterparty’s

92 Id.

93 Cf. Fischel, supra note 15, at 136-137 (discussing, in particular, lender-borrower relationships and

the attribution of discretionary powers to lender.)

94 Id. Fischel considers the attribution of discretionary powers (to lenders) as a form of bonding

mechanisms, which “are not ‘unfair’ to the borrower in any meaningful sense. The opposite is true, because

the borrower obtains credit on more advantageous terms as a result of the protections afforded the lender by

the loan covenants.” For a similar analysis of discretionary terms, see Klein, “Unfair” Contractual

Arrangements supra note 72.

38

expected contractual benefits, rather than to ensure contractual flexibility or deter

opportunism. Thus, the attribution of discretionary powers gives rise to a risk of reverse

cheating.95 For instance, the buyer could reduce her orders not because her operational

needs are diminished, but to profit from a fall in the market price of the good.96 Likewise,

the bank could accelerate the loan to profit from a sudden rise in the market interest rate.

Finally, in the delivery example, the seller could unreasonably postpone the delivery date

if she had absolute discretion to decide the delivery date.

It is not accidental that, in actuality, a party is rarely attributed absolute discretion.97

In the delivery example, it is very unlikely that the buyer might agree to give the seller

absolute discretion as to the date of delivery. In fact, discretion and specification are two

continuous variables (measurable by the number of contingencies that are addressed in

writing by the contract) that the parties use in writing the contract to reduce C. Rational

95 The expression is borrowed by Benjamin Klein. Klein, “Unfair” Contractual Arrangements, supra

note 72, at 359-360.

96 The case law on the abuse of discretionary powers is more than copious. Limiting the research to

requirements and outputs contracts, see, among the innumerable cases, Loundenback Fertilizer Co. v.

Tennessee Phosphate Co., 121 F. 298 (6th Cir. 1903) (buyer reduced order for profiting of a sudden decrease

in the market price of the goods); Orange & Rockland Utils., Inc. v. Amerada Hess Corp., 59 A.D. 110, 397

N.Y.S. 2d 814 (1977) (buyer increased order to resell products on the market and profit from a sudden

increase in the market price); New York Cent. Ironworks Co. v. United States radiator Co., 174 N.Y. 331,

335-336, 66 N.E. 967, 968 (1903) (where the court, in dicta, established that the discretion-exercising party in

a requirements contract cannot use her discretion “to speculate on the contract.”)

97 Lending agreements represent an exception. Loan covenants often attribute to lenders discretionary

powers that can be exercised upon the lender’s sole (i.e., unrestricted) discretion. Borrowers’ willingness to

grant lenders (absolute) discretionary powers is justified by the interest rate reduction, from which they

benefit in reason of the cut-off of transaction costs resulting from the attribution of such powers. See Fischel,

supra note 15, at 136-137.

39

parties will thus seek the (optimal) balance between discretion and specification that

minimizes C.

In the delivery example, for instance, the parties may agree that the seller will have

discretion as to the delivery date, provided, however, that the delivery takes place within

a certain time (e.g., two months). In this way, the buyer would not bear the risk that the

seller might postpone the delivery of the goods indefinitely, but the parties would still

benefit from the reduction of the inadequacy cost deriving from the attribution of

discretion to the seller. Similarly, in a requirements contract, the parties might agree to

give the seller discretion as to the quality of the goods to be supplied, but specify that the

seller must get such goods from a particular supplier, one who is known for the good

quality of her products. In this way, the buyer would be protected from the risk that the

seller might deliver goods below a certain quality. Still, by constraining discretion, parties

might bear a higher inadequacy cost. For instance, the seller could find goods of the same

quality as those produced by the supplier selected by the buyer, but at a cheaper price. In

turn, the parties could find more convenient to add an additional contingency that

specifies their entitlements should this kind of situation occur. This, however, might

originate other problems, which the parties will again solve by specifying more or

attributing discretion, in a continuum, until the optimal balance of specification and

discretion is reached.

3.2. THE EFFICIENCY CONDITION FOR THE GOOD-FAITH OBLIGATION

In the view of law-and-economics scholars, good faith is another device at parties’

disposal to reducen and, therefore, minimize C. By filling-in contractual gaps, good faith

makes a contract more complete. Thus, given a same number of contingencies, the

40

expected cost from incompleteness that parties bear when a contract includes good

faith,GFD , would be lower than the cost borne under a contract not including good faith:

GFD D< . As a corollary, for the same expected cost of incompleteness, the inclusion of

good faith in the parties’ agreement would save specification costs and, therefore, reduce

C.

Under this view, the contract needs to address a lower number of contingencies in

order to reach the optimal balance between specification and discretion. Good faith,

indeed, makes specific provision more state contingent and discretionary terms more

specific. Consequently, it reduces both the contract’s inadequacy and opportunism costs.

Consider, for instance, the example of the installment sales contract. The inclusion of

good faith in the parties’ agreement would prevent the seller to terminate the contract. By

making the contract more state contingent (on the practice of the seller of accepting

payment by mail, rather than on delivery), good faith prevents the seller from behaving

opportunistically. In the case of the seller who finds cheaper goods of the same quality as

those provided by the supplier indicated by the buyer, good faith would permit the seller

to buy those cheaper goods. Therefore, it would save parties the cost of making the

contract more state contingent. Similarly, in the loan agreement example, good faith

would not allow the bank to accelerate the loan for profiting from a sudden rise in the

market interest rate. Still, the buyer, in the requirements contract example, could not

reduce her orders to profit from a sudden fall in the market price of the good.

This view of good faith, however, is misleading. As discussed above,98 good faith is

not a perfect substitute of specification, but yields a less precise definition of the parties’

98 See supra Par. 2.2.

41

entitlements. When the occurrence of a particular contingency is addressed through good

faith, there is, indeed, a higher risk that the court might fail to find the “correct answer”.

This consideration, however, does not radically exclude the efficiency of good faith. The

specification test proposed by law-and-economics scholars should simply be modified to

take into account the probability of court error when a contingency is unspecified. For

instance, in the case of the delivery example, parties should specify their entitlements as

to the 5.45 a.m. delivery, but would not need to do the same as to the 3 a.m. delivery. In

the latter case, it is indeed unlikely that the court might make errors, notwithstanding the

less precise definition of parties’ entitlements arising from good faith. Following this

logic, given a same number of specified contingencies, good faith would still prove

efficient.

The picture just described, however, is not complete. The inclusion of good faith in

the parties’ agreement gives raise to a second, compelling, problem. Under a good faith

interpretative regime, courts are free to determine in their own judgment what good faith

means not only as regards unspecified, but also specified contingencies. This may lead to

a judicial redistribution of the parties’ entitlements “in the name of good faith”. Thus, in a

good-faith interpretative regime, the parties’ expected cost of incompleteness is only

partially a function of the number of contingencies that are specified by the contract. The

court’s discretion in enforcing the contract may, in fact, increase the parties’ expected

cost even if they specify the same number of contingencies that they would specify absent

good faith.

The transaction costs under a good-faith interpretative regime,GF

C , can thus be

expressed as ( ) ( ),GF GF

GFC S n D n e= + . As in the literal interpretative regime, then,

42

rational parties will seek to specify a number of contingencies,GFn , which minimizes

such costs.99 Letting ( ) ( )GFS S n S n! = " , and ( ) ( ),

GF GFD D n e D n! = " , it can be

concluded that the inclusion of good faith in the parties’ agreement is efficient only when,

the inequality S D! " ! is satisfied (the “efficiency condition for good faith”). Finally, the

good-faith trade-off may be restated as the trade-off between the specification-cost saving

and the increased expected cost of incompleteness deriving from a good faith

interpretative regime.

I claim that the contracting parties are in the best position to verify whether the

condition of efficiency for good faith is satisfied. Parties have an informational advantage

over courts. They can observe information that courts might not observe. And, above all,

parties can verify such information.100

However, rather than mandating the exclusion of good faith, as too vague a concept

that increases the likelihood of court error, this considerations should lead “to place the

burden of choosing the optimal contract squarely on the shoulders of parties.”101 Thus,

parties themselves should choose the interpretative stile to be adopted by courts to

99 Analytically, we differentiate ( ) ( ),GF GFC S n D n e= + , where ( )

GFe n , with respect to n . Hence,

we have GF GF

GF

dC dedS D D

dn dn n e dn

! != + +

! !. Setting the derivative equal to zero,

GFn is yielded at

GF

GF GF

dedS D D

dn n e dn

! != "

! !" . In particular GF

GF

deD

e dn

!

! tell us how parties can contain the court error through

specification. Under a good-faith interpretative regime, an efficient equilibrium between specification and

expected court error might still be achieved.

100 See supra Par. 1.3.

101 Eric Posner, A Theory of Contract Law Under Conditions of Radical Judicial Error, 94 NW. U. L.

REV. 749, 752-753.

43

enforce their contracts. They should be free to exclude good faith and opt for a literal

interpretative regime, or include it and opt for a good-faith interpretative regime, because

they know better which interpretative style is more likely to yield the “correct answer”.

Courts, in turn, should commit to parties’ sovereignty and follow their instructions.

Under this view, parties have, basically, two different orders of decisions to make.

Once they have verified the efficiency condition for good faith, and decided whether to

include it in their contract, they should choose the interpretative style of good faith that

courts should endorse. From a practical viewpoint, they should decide which evidentiary

base courts should use in the contextual interpretation of the contract.

3.4. DETERMINANTS OF THE EFFICIENCY CONDITION: NATURE OF THE PARTIES AND

TRANSACTIONAL ENVIRONMENT

I argue that two basic factors determine the outcome of the good-faith trade-off:

namely, the nature of the parties and the transactional environment in which their

contractual relationship takes place.

3.4.1. The Distinction between Unsophisticated and Sophisticated Parties

Parties may have either an unsophisticated nature or sophisticated. Sophisticated

parties include corporate entities and other business forms (such as limited partnerships,

for instance) which operate in commercial context and “are expected to understand how

to make business contracts.”102 The latter expression indicates that such parties are repeat

102 Schwartz & Scott, Contract Theory, supra note 59 (the authors set additional limits to their

definition of sophisticated parties; for instance, corporate entities are considered as such only if they have at

least five employees. In the approach proposed, instead, also a partnership of two people could be a

sophisticated party if the characteristics that are hereinafter described are satisfied.) On the dichotomy

44

players in the industry in which they operate, regularly conclude contracts of one or more

particular commercial types, and know well the economic substance of their exchanges.

In addition, sophisticated parties enjoy economic, informational, and, most of the time,

organizational resources. They are able to invest money, time and human resources in the

transactions they conclude. For example, a sophisticated party will be typically assisted

by a lawyer (or even a group of lawyers) in her business affairs. Furthermore, they have

the ability (i.e., the economic and organizational means) to obtain and process

information. In turn, they are able to asses their risk adequately and to write contracts that

contain such a risk.

Unsophisticated parties are defined residually as those that are not sophisticated.

Hence, these parties will normally tend to be individuals or other commercially

unsophisticated entities103 that conclude sporadic and heterogeneous transactions, mostly

in the context of simple transactional environments (such as one-time sales of relatively

low-value, common, goods).104 In addition, unsophisticated parties have limited economic

between sophisticated-unsophisticated parties, see also Benjamin Hermalin & Michael L. Katz, Judicial

Modification of Contracts between Sophisticated Parties: A More Complete View of Incomplete Contracts

and Their Breach, 9 J. L. ECON. & ORG. 230 (1993); J.H. Verkerke, Legal-Ignorance and Information

Forcing Rules, American Law & Economics Association, Annual Meeting, Paper 22 (2004).

103 For instance, a family business organized in the form of an unlimited partnership and with a small

number of partners (all siblings) would be considered an unsophisticated party (as long as the partnership

satisfies the other characteristics typical of unsophisticated parties).

104 See Eggleston, Posner & Zeckhauser, supra note 67, at 15-16 (comparing one-time sales and

requirements contracts.) In more general terms, the distinction between relational and discrete contracts may

prove useful to identify complex and simple environments. Relational contracts are agreements characterized

by “continuing highly interactive contractual arrangements” lasting over-time. Discrete contracts, instead, are

agreements performed simultaneously and, in contrast to relational contracts, characterized by: short duration,

limited personal interaction, precise measurement of objects of exchange, requirement of only a minimum of

45

and informational resources. As a result, they might both have a scarce understanding of

the economics of their exchanges and not be able to assess their risk adequately. In

addition, their bounded rationality may well lead to cognitive errors and to inefficient

allocations of entitlements.

(A) Unsophisticated Parties

The different characteristics of sophisticated and unsophisticated parties, and of the

contracts they conclude, yield opposite predictions on the efficiency of good faith.

Contracts between unsophisticated parties will tend to be “necessarily incomplete.”105

Due to their limited economic resources (including time) and their informational deficit,

not only tend unsophisticated parties to specify a relatively low number of foreseeable

contingencies, but many contingencies are, in fact, unforeseeable for them. For the same

reason, the cost of “completing the contract”, i.e., of specifying one additional

contingency, is also high at the margin.106 Thus, good faith permits such parties to save

future cooperation between the parties, failure to require benefit or burden sharing. Although discrete

contracts are difficult to find in actuality, actual contracts can be seen as a continuum of relational and discrete

contracts. When the parties’ agreement has more characteristics of a discrete rather than a relational contract,

the underlying environment can thus be considered relatively “simple.” The seminal work on the dichotomy

relational-discrete contracts is due to Professor Macneil. See Ian R. Macneil, Restatement (Second) of

Contracts and Presentiation, 60 VA. L. REV. 589, 594 (1974); Ian R. Macneil, Contracts: Adjustment of

Long-Term Economic Relations Under Classical, NeoClassical, 72 NW. U.L. REV. 854, 886 (1978). See also

Goetz & Scott, supra note 54. For a more recent discussion of the matter, see Richard E. Speidel, The

Characteristics and Challenges of Relational Contracts, 94 NW. U. L. REV. 823 (2000).

105 See Gillian Hadfield, Judicial Competence and the Interpretation of Incomplete Contracts, 23 J.

LEGAL STUD. 159, 159-161 (1994)

106 The function of contractual specification cost can vary conceivably across parties. Sophisticated and

unsophisticated parties have different fixed and variable costs of specification. Sophisticated parties can bear

easily the fixed cost of specification (i.e., the legal fees for the drafting of the contract); by contrast

unsophisticated parties will tend to hire a lawyer only when the dollar value of the transaction is very high (if

46

not only specification cost, but the cost that they should bear to obtain the information

necessary for figuring out contingencies.

In addition, when contingencies are unforeseeable, and parties do not have other

legal tools to address such contingencies (such as complex mechanics of renegotiation or

bonding mechanisms),107 the ex ante deterrent effect of good faith is not frustrated by the

possibility of court error. In this context, not including good faith basically legitimates

parties to behave opportunistically at the occurrence of unspecified contingencies. Then,

due to the large number of contingencies that are unforeseeable for an unsophisticated

party, she is better off by choosing good faith and accepting that her counterparty might

still behave opportunistically (anticipating the possibility of court error in the ex-post

enforcement of the contract) than, by excluding good faith, giving the counterparty “the

right” to do so.

they have the necessary economic means). Similarly, the marginal cost of specification for the ones and the

others will tend to be different. Due to their bounded rationality and informational deficit, unsophisticated

parties will usually face higher cost to figure out, negotiate, and write additional contingencies than

sophisticated parties do.

107 Sophisticated parties, unlike unsophisticated, will normally be able to limit the risk arising from

unforeseeable contingencies through several forms of bonding mechanism. For instance, sophisticated parties,

as repeat players, can write self-enforcing contracts, where the capital loss that can be imposed on the

potential cheater by the withdrawal of expected future business is sufficient to deter cheating (reputational

risk). Other bonding mechanisms used by sophisticated parties include, for instance, (i) high-price premium

clauses which limit the risk of opportunistic behavior of the counterparty through different incentives and

make the potential cheater worse off if she cheats; and (ii) forfeitable-at-will collateral bonds paid by the

potential cheater. The pivotal work on bonding mechanism is due to Michael C. Jensen & William Meckling,

Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure, 3 J. FIN. ECON. 305, 308

(1976) (defining bonding mechanisms, in general, as devices that protect one party (i.e., the principal) from

the misbehavior of the other (i.e., the agent) by imposing penalties on the latter for shortfall in her

performance.). From this viewpoint, as above discussed, also the contractual attribution of discretionary

powers may represent a bonding mechanism. See supra note 93.

47

Finally, when the transactional environment tends to be relatively simple, the court

will need to verify less information to find the “correct answer”. In turn, the asymmetry

of information between parties and courts will tend to be less relevant. In such a context,

even generalist courts108 might be able to identify the value-maximizing action of the

contractual relationship. In the case of unsophisticated parties, it is likely courts might be

able to complete the contract with a term that parties themselves would have agreed

upon.109

From the above, I predict that, in the case of unsophisticated parties, the

inequality S D! " ! tends to be satisfied. Hence, the inclusion of good faith in contracts

between such parties maximizes the ex-ante value of the contractual relationship.

(B) Sophisticated Parties

Conditions are very different with sophisticated parties. They have both the

economic and informational resources necessary to foresee and specify contingencies

and, therefore, to write contracts that contain their risk(s). Their agreements tend, in fact,

to be complex, which means (i) rich in state contingencies; (ii) providing for a high

108 By generalist courts, I mean courts without a specialized knowledge of the underlying transactional

environment and, more generally, lacking commercial competence.

109 See Schwartz, supra note 62. Take the example of a rental agreement, which provides that the

tenant must bear the maintenance costs of the property. Following a violent storm that unhinges doors and

breaks windows, however, the tenant refuses to pay the necessary repayments. The landlord brings suit

against the tenant. In such a case, the court’s decision to make the landlord to pay for the extraordinary

maintenance cost and the tenant partially reimburse such cost to the landlord would be a value-maximizing

decision. The landlord is the cheapest cost-avoider of extraordinary maintenance costs and, if parties had

written the contract making her to bear such costs, she would have been able to ask a higher price.

48

variance in the parties’ entitlements as to each state contingency; and (iii) requiring a

significant, and specific, cognitive load to be understood.110

In transactions between sophisticated parties, the likelihood that courts might fail to

find the “correct answer” naturally increases, due to the complexity that characterizes the

parties’ contract. Thus, at a first sight, good faith seems not to pass the efficiency test for

sophisticated parties. I maintain, however, that to verify the test the underlying

transactional environment must first be considered.

3.4.2. The Distinction between Non-Idiosyncratic and Idiosyncratic Transactions

(A) Non-Idiosyncratic Transactions

By non-idiosyncratic transactions, I mean transactions between parties who deal

regularly in a given market and are characterized by widely accepted customary practices.

111 When a transaction has these characteristics, apparently complex contracts are, in fact,

relatively simple. In non-idiosyncratic transactions, contracts tend to include boilerplate

provisions, often based on standard agreements supplied by industry-specific trade

organizations, and to make large use of implied references to long established customary

110 The definition of contractual complexity is borrowed by Eggleston, Posner, & Zeckhauser, supra

note, 67 at 122. (considering each component of complexity measurable along a continuum so that contracts

may be more or less complex).

111 See generally Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating

Cooperation Through Rules, Norms, and Institutions, 99 MICH. L. REV. 1724 (2001); Lisa Bernstein, The

Questionable Empirical Basis of Article 2's Incorporation Strategy: A Preliminary Study, 66 U. Chi. L. Rev.

710 (1999) (defining customary practices as “industry-specific meanings of words and unwritten industry-

wide ‘trade usage’ or ‘commercial standards’” and analyzing practices of contracting parties across different

industries. Bernstein argues that parties may prefer to have their disputes governed by private rules and

procedures supplied by private institutions on the basis of widely accepted industry-specific commercial

standards, rather than (generalist) courts).

49

practices. This is the case, for instance, of credit default swap or interest rate swap

agreements,112 which look like extremely complex financial contracts, but, in fact, simply

reproduce the standardized clauses of the ISDA agreement, a model agreement accepted

substantially by all the market operators.

Thus, in non-idiosyncratic transaction the likelihood that courts (even generalist

ones), might fail to find the “correct answer” is relatively low, because such answer tends

to be substantially the same across a large number of similar transactions. By referring,

for instance, to data about past transactions, courts are ordinary able to verify information

and enforce the parties’ expected contractual benefits at a low mistake rate.

In addition, in this kind of transactional environment, the specific content of good

faith is determined, basically, by the customary practice that is widely accepted by all

transactors of a specific industry. From this perspective, an objective standard of good

faith applies in non-idiosyncratic transactions. Good faith means commercial

reasonableness as determined by easily identifiable and industry-specific customary

practice.113 Hence, it is unlikely that courts might make more errors when good faith is

included in the parties’ agreement.

112 The International Swaps and Derivatives Association (ISDA) is a global trade association

representing participants in the derivatives industry. Such business covers swaps and options across all asset

classes (interest rate, currency, commodity and energy, credit and equity). The ISDA agreement is the

standard model used for drafting these kinds of contract. The agreement is regularly updated by the

association to reflect and incorporate market rules accepted by the operators. Thus, swap contracts are

generally concluded by signing a confirmation letter which incorporates by reference (part of) the provisions

contained in the “master agreement”.

113 This is the good-faith standard steadily applied by American courts in good-faith claims arising in

commercial settings. See, for instance, Eastern Air Lines, Inc. v. Gulf Oil Corp., 415 F. Supp. 429 (S.D. Fla.

1975), in which the court held that the industry practice had established a course of performance and dealing

50

From the above, it can be concluded that, when a transaction is non-idiosyncratic,

the specification cost saving implied by good faith will tend to outweigh the expected cost

due to judicial error. Hence, in non-idiosyncratic transactions, parties are better off by

including good faith in their contracts.

(B) Idiosyncratic Transactions

Conditions are different with idiosyncratic transactions, which are transactions

characterized by the specific nature of the parties’ investment or the high level of

specialization acquired by the parties’ relationship over the years. In such a context,

contractual agreements tend to be highly complex, reflecting an underlying transactional

environment that is per se complex. Consider, for instance, project-finance transactions

for the development of thermoelectric power stations. The life of such projects, their

enormous dollar value, the complexity of the underlying financial arrangements and

technical agreements result in contracts of hundreds of pages, which third parties (i.e.,

parties that have not been involved in the negotiation and drafting of such contracts) can

difficultly understand.

In addition, idiosyncratic transactions widely differ, and it is practically impossible

to individuate any commercial standard or trade usage. Each idiosyncratic transaction

tends, in fact, to be a unicum and not comparable to another, even when transactions take

place within the same industry. Thus, it is impossible to compare a project-finance

transaction, taking place in a particular location, to another. Different environmental,

engineering, financial, and legal aspects will tend to exclude any resemblance between

the two projects. In this transactional environment, then, the possibility of court error that had implicitly become part of the contract and was to be respected in good faith, that is, pursuant to the

standard of conducts established by that industry-practice.

51

tends to be high, since courts lack the specific competence required to understand the

parties’ agreements and are, therefore, unable to verify many of the information necessary

to enforce the parties’ substantive bargain.

It follows that including good faith in idiosyncratic transactions might well

jeopardize the economic efficiency of the parties’ exchange, by further increasing the

likelihood of court error. In this kind of transactions, the ex-ante value of the contractual

relationship is maximized by a literal judicial interpretative stile. Parties know better than

courts how to allocate their entitlements and court should defer to parties’ knowledge,

rather than attempting to guess information they do not have and are costly to acquire and

process. Thus, in idiosyncratic transactions, good faith should be excluded by the parties’

contracts and courts should abide by the letter of this contract as if it was a complete one.

4. CONCLUSIVE REMARKS

4.1. WHY GOOD FAITH SHOULD BE A DEFAULT RULE

Good faith should be a default rule of law. It should not be a mandatory rule, as

traditionally argued,114 because it does not prove efficient in all cases. In particular, I have

predicted in this work that imposing good faith on sophisticate parties, in idiosyncratic

transactions, would reduce the value of their contractual surplus. However, good faith

114 See Ian Ayres, Default Rules for Incomplete Contracts, THE NEW PALGRAVE DICTIONARY OF

ECONOMICS AND THE LAW, Vol. A-D 585 (1998) (classifying good faith as a typical immutable rule of law (i.e.

mandatory rule) as opposed to default rules.) See also Ian Ayres & Robert Gertner, Filling Gaps in Incomplete

Contracts: An Economic Theory of Default Rules, 99 YALE L. J. 87, 87 (1989).

52

should still be provided by law being the majoritarian default. Unsophisticated parties, who

are the majority, are likely to opt for including good faith in their contracts.115

Good faith should be a rule that parties are free to contract around. The default should

apply only when the parties’ contract is silent on good faith. In this case, parties should be

presumed to be unsophisticated and, therefore, a good-faith interpretative regime should be

applied to their contracts. However, when the default regime applies, courts would not be

constrained by the parties’ indications (if any) on the interpretative stile of good faith.

Instead, courts would be free to determine which evidentiary base use in the ex-post

enforcement of the parties’ entitlements.

4.2. HOW PARTIES SHOULD WRITE THEIR CONTRACTS

Good faith should be an instrument at parties’ disposal to write better contracts and

maximize the ex ante value of their contractual relationship. For this, it should be included

only when it satisfies the condition of efficiency S D! " ! . Otherwise, parties are better off

by excluding good faith. From a practical viewpoint, this delineates a first-order choice at

parties’ disposal: whether to opt for a contextual (good-faith-based) interpretative regime or

a literal one. By excluding good faith, parties signal to courts that the contract is the only

115 Even if unsophisticated parties did not represent the majority (which is unrealistic, though), good

faith should nevertheless be provided by law since efficiency mandates that a legal rule be available, even

though many will decide to opt out of it, when the cost of bargaining for that rule for those that require it is

higher than the cost incurred to bargain around the rule. Indeed, for unsophisticated parties bargaining “for

including good faith” would represent a high cost since such a bargaining would require, in practice, the

specification of all the contingencies that are implicitly addressed by good faith. See David Charny,

Hypothetical Bargains: The Normative Structure of Contract Interpretation, 89 MICH. L. REV. 1815, 1842-43

(1991).

53

evidentiary base to be used in the ex post enforcement of their expected contractual

benefits, and that the contract must be interpreted as if it were complete.

From a practical viewpoint, parties have already found ways to limit the courts’

activism in contract interpretation. They include in their agreements merger clause, which

signal to court their preference for the exclusion of pre-contractual negotiations from the

available evidentiary base. They also include general clauses directing courts to ignore

extra-contractual evidence, such as course of dealings, or no-oral modification clauses,

which ask court to disallow any contract modification that is not in writing. Still, they may

choose jurisdictions with strict interpretative laws. Finally, parties may provide for

arbitration rather than judicial enforcement.

Under the regime I propose, the current instruments at parties’ disposal can be

combined with the exclusion of good faith to give courts even a stronger signal on the

parties’ choice of a literal interpretative regime. In addition, in some circumstances,

opting out of good faith might not be enough to ensure that courts will follow the letter of

the contract. For instance, in the context of idiosyncratic transactions, generalist courts

might be not able to follow that letter due to the high complexity of the parties’

agreements. Moreover, when contracts are highly specified, as complex contracts are,

they might include more conflicting terms than simple contracts do. Hence, at the

occurrence of a particular contingency, courts might well misinterpret a term for another,

regardless of the interpretative regime parties have chosen. Circumstances might arise

where the level of specification of the contract is so high that the expected error cost of

54

generalist courts is the same under a good faith or a literal interpretative regime.116 In

such a case, then, the trade-off decision is not between including and excluding good

faith, but rather between expected error cost of generalist courts and the cost of better-

quality adjudicators (i.e., specialized but far away courts or arbitration).117

Another option at parties’ disposal is to exclude or include good faith only partially.

This mean that parties can specify in their contracts the clauses as to which they want good

faith to apply or not to apply. Rather interestingly, this is what parties do in practice.118 For

116 Cf. Hadfield, supra note 105, at 171-172,175-178 and (discussing analytically the effects of courts’

competence.)

117 The choice of more specialized courts may assure a better-quality enforcement, but is also likely to

raise the costs of contracting when they are located in other jurisdictions, such as the cost of hiring local

counsels, travel expenses, etc. The same consideration applies to arbitration, which may be even more

efficient (since more adjudicators are even more specialized), but may involve considerable costs. A less

expensive method of containing the possible damages arising from the lack of (specific) competence of

adjudicators, although not as effective as the two above in ensuring high-quality enforcements, is the

contractual waiver of the right to a jury trial. In this way, “[the risk of] bias … likely to be outcome

determinative [is virtually eliminated] and complex issues are better understood. See Federal Court Rules in

Favor of DaimlerChrysler AG - Dismisses Investor’s Claim That Merger of Daimler-Benz AG and Chrisler

Corporation was Misrepresented to be a “Merger of Equals”, Skadden, Arps, Slate, Meagher & Flom LLP &

Affiliates, memorandum available online at

http://www.skadden.com/content/Publications/Publications1028_0.pdf , last visited May, 8, 2005 [hereinafter,

Skadden Memo]

118 I analyzed a limited sample of 200 contracts, in a temporal range going from 1990 to 2005. In

particular, I examined the following kinds of contracts: Joint Ventures Agreements, Merger Agreements,

Spin-offs Agreements (“Recombination Transactions”); Intellectual Property Agreements; Lease Agreements;

Sale (Purchase of Goods) Agreements; Sale (Purchase of Services) Agreements (“Business Transactions”);

Benefit/Incentive Plans Agreements; Compensation/Employment Agreements; Contractors/Consulting

Agreements; (“Compensation/Employment Transactions”); Guarantees/Security Agreements; Loan

Agreements; Purchase/Sale of Financial Assets Agreements; Trust Agreements (“Finance Transactions”);

Franchise Agreements (“Franchise Transactions”); Settlements Agreements (“Settlements Transactions”);

55

Shareholder/Stockholder Agreements; Subscription Agreements; Underwriting Agreements; Voting

Agreements; and Warrants & Warrant Agreements (“Security Transactions”).

From the research on the sample agreements, the following data have emerged:

(1) parties write contract as if the implicit obligation of good faith did not exists. They

tend to restate the good-faith requirement as to specific clauses. The practice is inconsistent with the

Restatement (Second) of Contract, the U.C.C., and the common law decisions establishing the

mandatory nature of the good-faith rule;

(2) the good-faith obligation is very often mentioned as a synonym for best effort (i.e.,

reasonable best effort);

(3) in long-term contracts, the good-faith obligation is often used as a device to agree

to try finding a contractual solution before any legal action is commenced by one party against the

other. The following clause is a typical example “[i]f any dispute arises under this agreement that is

not settled promptly in the ordinary course of business, the parties shall seek to resolve any such

dispute between them, first, by negotiating promptly with each other in good faith in face-to-face

negotiations.”;

(4) when discretionary powers are attributed by contract, parties tend to restate their

good-faith obligation explicitly, especially when the contract attributes one of the parties a power of

unilateral determination of the contractual performance;

(5) parties almost always mention good faith when they agree on generic cooperation

clauses. The following expressions are typical: “cooperate in good faith with the other party” or,

more elegantly, “[p]arties will consult and cooperate with one another, and will consider in good

faith the views of one another”;

(6) in Business Transactions, and Finance Transactions, parties usually refer to good

faith when they represent and warrant each other;

(7) very seldom parties refer to good faith when they regulate standards of conduct;

(8) in a few Credit Agreements, I have found a reference to good faith in exculpatory

clauses. Through such clauses, parties agree that, as long as they acted in good faith, they will not be

liable notwithstanding the failure to fulfil a particular contractual obligation. The following wording

represents an example of this kind of exculpatory clauses: “[t]he [d]epositary shall not be responsible

for any failure to carry out any instruction to vote any of the deposited [p]referred [s]hares or for the

manner or effect of any such vote made, as long as any such action or non-action is in good faith;”

(9) in many Litigation Agreements, I have found that parties refer to good faith as the

standard obligation for the pre-trial conduct, regardless of the kind of contractual breach involved.

56

example, in renegotiation clauses, parties tend to restate good faith. This, however, is totally

useless under the current regime, since good faith is an implied obligation of every contract.

Similarly, in financial contracts, parties expressly exclude good faith sometimes.119 This is

also totally useless because of the mandatory nature of good faith.120 On the contrary, under

the proposed good faith regime, this practice could be an efficient method for the parties to

better address costs of specification.

When parties decide to include good faith in their contracts (whether totally or

partially), they can choose the interpretative stile they want courts to adopt in interpreting

good faith. They can thus specify which meaning should be attributed to good faith in that

particular context. In addition, they can select the evidentiary base courts should use in the

contract’s interpretation. For instance, they can limit the evidentiary base to their practice

under the current agreement or decide to include also practices developed under previous

agreements. They can include all written pre-contractual documentation (i.e., memoranda,

Source: Contracting and Organizations Research Institute (CORI). CORI is a digital library which

contains 26,928 contracts. Most of the contracts in the collection are executed agreements made available in

public disclosure filings or filed with a regulatory agency. (See at http://cori.missouri.edu/pages/ksearch.htm,

last visit May 15, 2005).

119 [I]f the [enforcement] action of [the bank] … shall be permitted or required pursuant to any

provision hereof … such action shall be required to be in writing and may be withheld or denied by [the bank]

…, in its sole discretion, with or without any reason, and without being subject to question or challenge on the

grounds that such action was not taken in good faith. (Source: Financing Agreement signed by Fortress

Investment Group LLC in 2002.)

120 Pursuant to Section 1-102 (3) U.C.C., the duty of good faith cannot ”be disclaimed by agreement”;

however, parties can determine the standards of good faith if those standards are “not manifestly

unreasonable.” Section 1-102 (3) does not clarify what “manifestly unreasonable” means. Thus, if the above-

mentioned practice of the parties will be deemed lawful by a court cannot be easily predictable. Courts,

however, have so far adopted a basically restrictive approach toward the “private re-definitions” of good faith.

See BURTON & ANDERSEN, supra note 4, passim.

57

prior drafts, letters, etc.) or limit the evidentiary base only to some among these documents.

Thus, under the proposed regime, not only parties would be able to determine which

interpretative regime courts should adopt in enforcing their contracts, but also the

interpretative stile of their good-faith obligation.

4.3. HOW COURTS SHOULD INTERPRET CONTRACTS

Courts should delegate to parties both the decisions of whether to include or exclude

good faith in their contracts and that of the interpretative regime to enforce those contracts.

Parties have more and better information than courts and, therefore, are in a better position

to evaluate the efficiency condition for good faith. In turn, few exceptions should be

admitted to the commitment to “party sovereignty … [as to] the interpretative style an

adjudicator should use to find the substantive terms [of the agreement].”121

From this perspective, courts could disavow the parties’ decision to exclude a good-

faith requirement from their contract only if:

(i) at least one of the parties is unsophisticated; and/or

(ii) one of the party misrepresented information determinant in the decision of

excluding good faith.

When one of the parties is sophisticated and the other is unsophisticated, there are

two problems that mandate a more careful judicial scrutiny of the parties’ decision to opt

out of good faith and, possibly, permit courts to disallow private autonomy. The parties’

relationship may be affected by bargaining inequality. As discussed above, significant

differences in terms of both economic and informational resources distinguish sophisticated

from unsophisticated parties. Sophisticated parties have notably larger means at their

121 Schwartz & Scott, Contract Theory, supra note 59, at 569.

58

disposal, which they can potentially use in a coercive way, for example to obtain more

favorable economic terms, when the counterparty is an unsophisticated one. Thus, when the

decision to opt out of good faith appear prima facie to be the result of coercive pressure

exercised by the sophisticated party on the unsophisticated one, courts should disavow the

parties’ “decision” and apply the default regime.

In addition, even when sophisticated parties do not abuse their economic means

coercively, they still enjoy a significant informational advantage on unsophisticated parties.

This means that many information (about current and future states of the world) are

observable and, above all, verifiable only by the sophisticated party. Thus, it may well be

that the unsophisticated party may agree to opt out of good faith on the basis of limited

information, while should the party had enjoyed the same information at disposal of the

sophisticated one, she would not have agreed to such a decision. I suggest, however, that

courts should adopt a restrictive approach in disallowing parties’ agreements on the basis of

asymmetry of information only. To avoid jeopardizing party autonomy, the burden of proof

should fall on the unsophisticated party to show that if she had known that information, she

would not have opted out of good faith.

Finally, when the choice to opt out of good faith is determined by misrepresentation

of one party to the other, the court could disallow the parties’ choice even when they are

both sophisticated. Also in this case, however, especially if parties are both sophisticated,

courts should adopt a restrictive approach and limit their intervention to apparent cases of

misrepresentation by one party to the other.122

122 The recent decision of the United States district Court for the District of Delaware in the litigation

brought by the billionaire Kirk Kerkorian against DaimlerChrysler AG (Daimler) constitutes a practical

example of the restrictive approach I propose here, although the case does not raise an issue of good faith, but

59

rather one of securities fraud. In 2001, Kerkorian (i.e. Tracinda Corporation, his investment vehicle) brought

suit against Daimler alleging that the 1998 merger between Daimler-Benz AG and Chrysler Corporation was

falsely represented as a “merger of equals". On April 7, 2005, after a fierce legal battle between opposite

counsels, the court rejected Kerkorian’s claim holding that Tracinda had, in fact, not rely on the alleged oral

representations supposedly made directly to Kerkorian. The court applied an “absence of reliance” test based

essentially on the sophisticated nature of Kerkorian (who was advised by his own advisers, had designees on

Chrysler’s board) and ruled that in transactions between sophisticated parties reliance cannot be presumed but

must be demonstrated by the party bringing legal action.

SIMPLE http://www.unisi.it/lawandeconomics/simple.html

ELENCO DEI CONTRIBUTI RICEVUTI FINO AL NUMERO ODIERNO

2005

41 Veronica Grembi Guido Calabresi e l'analisi economica del diritto 40 Roger Van den Bergh, Yves

Montangie Competition in professional services markets: are latin notaries different?

39 Antonio Nicita L'Analisi Economica dei Contratti 38 Angelo Castaldo, Antonio Nicita Essential Facility and Efficiency in European Antitrust.

Some Lessons from GVG/FS in the Railway Sector 37 Pier Luigi Parcu Italian Authorities and Regulatory Reforms: Towards a European Framework? 36 Massimiliano Vatiero Atti dell’AGCM tra efficienza, equilibrio e fairness 35 Roberto Galbiati, Antonio Nicita,

Giorgio Nizi Competition And Remedies In Pay-Tv Markets: Recent Lessons From Uk, Australia And Italy

34 Massimiliano Vatiero Pareto concorrenza e Pareto efficienza 33 Roberta Karme The Challenge of Equivalency--US and EU Disclosure Requirements After Sarbanes-Oxley

and the Financial Services Action Plan 32 Laura Ferrari Bravo & Antonio

Nicita La Distribuzione Dei Farmaci In Italia. Verso Riforme Proconcorrenziali?

31 Antonio Nicita & Ugo Pagano Incomplete Contracts and Institutions 30 Maria Alessandra Rossi Software Patents: a Closer Look at the European Commission’s Proposal 29 Massimo D'Antoni, Roberto

Galbiati Deterrence And Information: The Optimal Use Of Monetary And Nonmonetary Sanctions Revisited

2004

28 Antonio Nicita, Ugo Pagano Institutional complementarities, corporate governance and financial-technological equilibria

27 Laura Ammannati Tutela della concorrenza e accesso al mercato dei servizi pubblici locali dell'energia: il caso del gas

26 Elisabetta Cervone EU conduct of business rules and the liberalisation ethos.The challenging case of investment research

25 Antonio Nicita Hold-up, competition and vertical integration:another look at fisher body/general motors?

24 Antonio Nicita, Ernesto Savaglio Minimal liberty and the 'coasian liberal': boundaries and complementarities between the state and the market

23 Antonio Nicita, Ugo Pagano Law and Economics in retrospect 22 Antonio Nicita, Matteo Rizzolli Much Ado About The Cathedral: Property rules and liability rules when rights are incomplete

21 Pier Luigi Parcu The dominant position and its abuse: suggestions from game theory 20 Emilio Asaro Collusion and auctions: a survey

2003

19 Francesco Farina Game-theoretic solutions to endogenous contractual incompleteness 18 Simone Sepe La cartolarizzazione: aspetti giuridici ed economici

17 Francesco Denozza Consumer transaction costs at the intersection of antitrust and intellectual property 16 Thomas S. Ulen Firmly grounded: economics in the future of the law

15 Luca Fiorito Transaction cost economics before Coase: L.K. Frank on industrial integration 14 Roberto Galbiati Durable goods monopolies: some insights on time inconsistency 13 Antonio Nicita Concorrenza e sanità: il 'pendolo delle riforme' in Europa

12 Laura Ammannati, Roberto Galbiati, Giorgio Nizi

Regolazione e concorrenza nei servizi pubblici locali

11 Marcello Clarich Servizi pubblici e diritto europeo della concorrenza: l'esperienza italiana e tedesca a confronto

10 Giuseppe Bellantuono Contratti incompleti e norme sociali

9 Donatella Porrini Financial internalization of environmental damage: legal framework for economic solutions

8 Ruggero Paladini Law & Economics ed economia pubblica

7 Maurizio Franzini, Antonio Nicita Externality, efficiency and fairness: understanding inertia in coase theorem 6 Lorenzo Sacconi Incomplete contracts and corporate ethics: a game theoretical model under fuzzy

information 5 Antonio Nicita Esternalità, transazioni e ambiente. Una rivisitazione del teorema di Coase 4 Arman Khachatrian Theory of the firm. A critical survey

3 Giovanni Ramello Diritto d'autore e tutela della concorrenza nell'esperienza antitrust europea e italiana 2 Antonio Nicita, Ugo Pagano Le istituzioni economiche del capitalismo alla prova della globalizzazione 1 Laura Ammmannati Diritto e Mercato. Una rilettura delle loro attuali relazioni alla luce della nozione di

'transaction' di Commons

simple Siena Memos and Papers on Law & Economics

La collana di quaderni SIMPLE nasce nell’ambito della ricerca collegata alla Scuola di Dottorato di Ricerca in Diritto ed Economia (sezione Law and Economics), nata dalla collaborazione tra l’Università di Siena, l’Università di Trento, il Liuc Castellanza e la Scuola Superiore in Economia e Finanza. Il dottorato L&E si propone di formare studiosi e dirigenti di istituzioni economico-sociali nazionali ed internazionali e di imprese per le quali sia richiesta una preparazione interdisciplinare con particolare attenzione all’area giuridico-economica. Il percorso didattico, al quale contribuiscono docenti provenienti da diverse università italiane ed estere, nonché esperti di istituzioni pubbliche e private, si propone di costruire un bagaglio di conoscenze e strumenti comuni nelle aree di diritto ed economia, partendo da basi disciplinari differenziate, come quelle offerte nella facoltà di economia, giurisprudenza e scienze politiche. In particolare, l’attività didattica e di ricerca offerta dal dottorato è rappresentata dalle seguenti macro-aree:Contratti, Diritti di Proprietà, Responsabilità; Organizzazioni economiche e Corporate Governance; Concorrenza, Regolazione e Antitrust; Istituzioni, Organizzazioni e Scelte pubbliche; Regolazione e Organizzazioni internazionali. Nell’ambito di questo percorso formativo sono previsti inoltre moduli di formazione di base per economisti e giuristi e moduli/seminari specialistici, tra i quali: diritti di proprietà intellettuale, finanza d'impresa, sistemi finanziari, analisi dell'impatto della regolamentazione, etica e impresa, diritto penale ed economia del crimine, tutela ambientale e tutela del consumatore. L’ iter formativo si divide in tre segmenti: uno di frequenza obbligatoria delle lezioni della durata di un anno e mezzo, un secondo di eventuale formazione all’estero e/o di stage presso istituzioni o enti di ricerca, di durata variabile da un semestre ad un anno, l’ultimo, focalizzato alla preparazione della tesi. L’obiettivo formativo è diretto a condurre lo studente ad una preparazione interdisciplinare orientata ad una successiva specializzazione tematica.

http://www.unisi.it/lawandeconomics