CONTRACT LAW: GENERAL THEORIES
Professor of Law, Stanford Law School
© Copyright 1998 Richard Craswell
When contracts are incomplete, the law must rely on default rules to resolve anyissues that have not been explicitly addressed by the parties. Some default rules(called "majoritarian" or "market-mimicking") are designed to be left in place bymost parties, and thus are chosen to reflect an efficient allocation of rights andduties. Others (called "information-forcing" or "penalty" default rules) aredesigned not to be left in place, but rather to encourage the parties themselvesto explicitly provide some other resolution; these rules thus aim to encourage anefficient contracting process. This entry describes the issues raised by suchrules, including their application to heterogeneous markets and to separatingand pooling equilibria; it also briefly discusses some non-economic theories ofdefault rules. Finally, this entry also discusses economic and non-economictheories about the general question of why contracts should be enforced at all.
JEL classification: K12
Keywords: Contracts, Incomplete Contracts, Default Rules.
This section describes research bearing on the general aspects of contract law.Most research in law and economics does not explicitly address these generalaspects, but instead proceeds directly to analyze particular rules of contract law,such as the remedies for breach. That body of research is described below insections 4100 through 4800.
There is, however, some scholarship on the general nature of contract law's"default rules," or the rules that define the parties' obligations in the absence ofany explicit agreement to the contrary. The phrase, "complete contingentcontract" is sometimes used to describe an (imaginary) contract that would spellout in complete detail the exact legal rights and duties of each party under everypossible state of affairs. While no real contract ever achieves this level ofcompleteness, the concept is still useful to define one endpoint of a spectrumof completeness. If any contract ever succeeded in reaching this endpoint, thelaw's default rules would then be irrelevant, as no issue would ever arise thatcould not be settled by the terms of the contract itself. Indeed, the same is trueof the interpretation of most other legal documents, such as wills (see section5830) or legislative enactments (see section 9200).
As long as legal documents fall short of this level of completeness, the lawmust have some set of presumptions or default rules, in order to resolvedisputes that are not settled by the terms of the document itself. The law could,of course, simply refuse to enforce any contract (or any will, or any statute) thatfell short of absolute completeness. But such a rule would itself be a "defaultrule": it would be a legal rule defining the obligations (or lack of obligations) thatresult when a contract does not itself specify what rules should govern. As longas actual contracts fall short of full completeness, then, the existence of defaultrules is not so much a choice as a logical necessity. The only question is whatthe content of those default rules ought to be.
In choosing the content of default rules for contractual relationships, it isoften useful to distinguish default rules chosen to increase efficiency if they areallowed to remain in force (as discussed in subsections 2-4) from those chosento increase efficiency if many parties contract around the default rule (discussedin subsections 5-9). In some cases, the steps required to contract around adefault rule could themselves increase efficiency, by inducing one party toreveal private information (as discussed in subsections 6-8). In other cases, itmay be too hard for the law to identify a single rule that would increaseefficiency if allowed to remain in force, so the default rule may instead beselected purely for its simplicity or ease of administration (as discussed insubsection 9). The choice between these approaches depends in part on theability of courts or lawmakers to identify efficient default rules; this issue isaddressed in subsection 10. The possible effect of default rules on the parties'own preferences is discussed in subsection 11, while subsection 12 discussessome non-economic theories of default rules.
Finally, there is also a good deal of scholarship on the general question ofwhen contracts ought to be enforceable, either through informal social sanctionsor more by formal legal mechanisms. Subsection 16 introduces these issues, anddiscusses the choice of enforcement mechanisms. Subsections 18-20 focus onthe question of whether promises should be enforceable at all (whatever themechanism), with subsection 18 discussing noneconmic theories of enforcementand subsections 19-20 discussing the economic theories.
A. "Majoritarian" or "Market-Mimicking" Default Rules
One way to select a default rule is to identify the rule that would be mostefficient if that rule were allowed to remain in place (i.e., if the parties did notspecify some other rule in their contract). For example, if the expectation measureof damages were determined to be the most efficient remedy for breach ofcontract, this approach to selecting a default rule would argue for makingexpectation damages the default remedy.
Much of the economically-oriented research on specific topics of contractlaw falls into this category. For example, many analyses of the remedies forbreach aim to identify the remedy that would produce the most efficient resultif that remedy were allowed to govern the parties' relationship (i.e., if the parties'contract did not stipulate that some other remedy would govern). The same istrue of many analyses of implied excuses such as impracticability and mistake,or of other terms such as implied warranties. This work is described at morelength in sections 4500 (unforeseen contingencies), 4600 (remedies), and 4700(warranties). This subsection addresses only those issues common to all defaultrules of this sort.
Default rules selected on this basis--that is, on the grounds of their efficiency ifallowed to remain in force--are sometimes described as the rules that the partiesthemselves would have chosen, if they had taken the time to agree explicitly ona rule to govern their relationship. In most cases, parties to a contract have aninterest in maximizing the efficiency of their relationship, so the rule the partiesthemselves would have chosen will be the same as the rule that would be mostefficient if allowed to remain in place. Thus, default rules chosen on this basishave also been labeled "market mimicking default rules," or default rules basedon the principle of "hypothetical consent." To be sure, if there are third-partyeffects, or if the parties to the contract are imperfectly informed or are subject toany other market failures, the rule that would be chosen by the actual partiesmight no longer coincide with the rule that would in fact be most efficient. Forgeneral discussions of the relationship between efficient default rules andhypothetical consent, see Posner and Rosenfield (1977, p 89) , Ayres and Gertner(1989, pp 89-93) , Coleman, Heckathorn, and Maser (1989) , and Craswell (1992) .Subsection 4 discusses some additional complexities that arise in heterogeneousmarkets, in which the same rule would not be chosen by every pair ofcontracting parties.
This basis for selecting default rules has been supported by two economicarguments, both of which involve transaction costs. (Some noneconomicarguments will be discussed in subsection 12 below.) The first economicargument applies when it would be prohibitively expensive for the parties tomake their contract more complete by specifying the rule they want to govern aparticular contingency. This is especially likely for extremely low-probabilitycontingencies, where the expected benefits of specifying a rule in advance arelikely to be low. In such a case, any rule the law selects as a default rule willinevitably be left in place by the parties, so the only way to get the benefits ofwhatever rule is most efficient is to make that rule the default rule. The secondeconomic argument applies if it would be costly, but not prohibitively expensive,for the parties to make their contract more complete by specifying their own rule.In such a case, selecting a default rule that matches whatever rule the partiesprefer may save some parties from having to incur those transaction costs, thusproducing all the benefits of the most efficient rule with lower total transactioncosts.
These arguments become more complicated, however, if a "market-mimicking" default rule is being applied to a market that is imperfect in somerespect. For example, if a remedy of expectation damages would in fact be mostefficient, but if imperfect information has led the parties to believe that someother remedy would be more efficient, should the law adopt as its default remedythe one that is in fact most efficient, or should it adopt the less efficient remedythat the (imperfectly informed) parties would choose if left to their own devices?The transaction-cost argument discussed in the preceding paragraph couldsuggest that, if the cost of contracting around the default rule is sufficiently lowthat the parties are likely to do so, then the law might as well adopt the less efficient remedy as the default rule, because that is the remedy the parties willcontract for anyway. Alternatively, if the costs of transacting around the defaultrule are so high that the parties are likely to leave the default rule in place,perhaps the law should adopt the remedy that in fact is most efficient, as that willgive the parties the benefit of the more efficient rule. But this latter approach iscomplicated by the fact that the identity of the rule that is most efficient (ifallowed to remain in force) may itself change if the parties are imperfectlyinformed, or if there are other market imperfections. Moreover, in some cases thepresence of information asymmetries or other market imperfections may call foran entirely different approach to the selection of default rules, in which thedefault rules are intentionally chosen not to be left in force by many parties. Thisapproach is discussed below in subsections 5-9.
Market-mimicking or "majoritarian" default rules raise additional issues inmarkets characterized by heterogeneity, where different rules would be efficientfor different contracting pairs. If a single default rule must be chosen to governall contracts in such a market, the rule selected will determine which contractingpairs can do best by leaving the default rule in place and which pairs couldpotentially do better by incurring the transaction costs needed to specify someother rule. It is sometimes assumed that the most efficient single default rulewould be that which was most efficient for most of the contracting pairs [hencethe label "majoritarian" default rule, coined by Ayres and Gertner (1989) ]. But iftransaction costs differ across heterogeneous parties, this conclusion actuallydepends on the exact levels of transaction costs that each member of eachcontracting pair would have to incur to contract around whatever default rule thelaw adopts [see Ayres and Gertner (1989, pp 112-15) ].
If different rules would be efficient for different contracting pairs, the lawmust also to decide the extent to which its default rules should be "tailored," orcustomized to match the rule that would be most efficient for each individualcontracting pair [see Ayres (1993) , Goetz and Scott (1985) , Baird and Weisberg(1982) ]. If the law adopts a single, untailored default rule, that will be mostefficient for only some of the contracting pairs: all other pairs will have to incureither (1) the transaction costs of contracting around the default rule, or (2) theefficiency loss from leaving in place a default rule that is less than efficient fortheir contract. A set of default rules tailored to each individual contracting paircan in theory eliminate or reduce these costs, but such "tailoring" will usuallyintroduce other costs. For instance, individually tailored rules will usually bemore complex, thus increasing the drafting costs that must be incurred by thelegislature or other lawmaking body. Moreover, it is often impossible to spell outa complete set of individually tailored rules in advance, so the law will insteadhave to rely on vague standards to be applied by courts on a case-by-case basis(e.g., an implied excuse in cases where performance is no longer commercially"reasonable"). Vague standards such as these usually entail higher litigationcosts; they also introduce the possibility of case-by-case error in the applicationof the standard, and may make it hard for the parties to predict what rule will beapplied to their relationship. In short, the question of how finely to tailor adefault rule--and, indeed, whether to tailor it at all-- raises most of the sameissues that are raised whenever the law faces a choice between specific rules andvague standards (see section 9000).
B. "Information-Forcing" or "Penalty" Default Rules
In many transactions, the two parties begin with differing amounts ofinformation. In some cases, they may be differently informed about the relevantlegal rules, or about the risks involved in the transaction. They may also bedifferently informed about the characteristics of the other party--for example, theseller may not know how much the buyer will lose if the seller's product turnsout to be defective. Differences such as these are often referred to as"information asymmetries." While information asymmetries are often addresseddirectly through disclosure regulations and the like (see section 5110), they canalso have implications for the law's choice of default rules. Subsection 6 belowdiscusses cases where the two parties are differentially informed about thegoverning legal rule. Subsections 7-8 then discuss cases where the two partiesare differentially informed about some other aspect of the transaction.
In some cases, the choice of default rule may help correct one party'sinformation about the legal rule itself. For example, suppose that the defaultremedy for breach provides for only a small monetary payment. Suppose alsothat the seller (the potential breacher) knows this, but the buyer (the potentialnonbreacher) does not. Suppose that the buyer instead thinks, incorrectly, thatthe law's default rule provides for a very large payment in the event of breach.
If left uncorrected, this information asymmetry could lead to either of twoproblems. First, if a larger remedy would be more efficient for this contractingpair, they may not alter their contract to provide for the larger and more efficientremedy, because the buyer will think that he or she already has the benefit of alarger remedy. Second, even if the smaller remedy provided by the default ruleis in fact most efficient for this contracting pair, the buyer's ignorance about theactual rule may keep him or her from optimally adapting his or her behavior tothat rule. For example, the buyer may purchase insufficient insurance, or takeinsufficient precautions to reduce the losses that would be caused by breach.(The effect of legal remedies on the incentives governing decisions such asthese is discussed at more length in section 4600.)
By contrast, suppose now that the default rule were changed to provide fora very large payment in the event of breach. If this remedy is left in place (i.e., ifthe parties do not contract around it), then both parties will optimally adapt theirbehavior: the buyer will adapt because this rule matches what he or she thinksthe rule is; while the seller will adapt because, by hypothesis, he or she iscorrectly informed about whatever the default rule is. And if this default remedyis not left in place (i.e., if the parties contract around it by stipulating a smallerremedy), the act of stipulating some other remedy should serve to inform thebuyer about what remedy will apply in the event of breach. Indeed, if the defaultrule is one that the parties will be certain to contract around--for example, if it ischosen to be highly unfavorable to whichever party is better informed about thelaw--that will virtually ensure that the other party will also become correctlyinformed about the resulting rule, by the very process of stipulating to someother rule in their contract. Hence, these default rules are sometimes describedas "information forcing" rules ( Scott (1990, p 609) ), or as "penalty defaults"slanted against the better-informed party to induce him or her to contract aroundthe default rule ( Ayres and Gertner (1989, p 97) ). A number of analyses ofparticular topics in contract law have recommended default rules chosen on thisbasis, or have suggested that actual legal rules could be understood as servingthis function. See, e.g., Ayres and Gertner (1989, pp 98-99, 104-06) , Goldberg(1984, pp 295-96) , Muris (1983, pp 390) , Verkerke (1995, pp 885-90) , Isaacharoff(1996, pp 1793-95) .
One difficulty raised by default rules slanted against the better-informedparty is that it may not be clear which party is better informed. This raisesanother "tailoring" question of the sort discussed supra in subsection 4: shouldthe party against whom the default rule is slanted be determined in advance fora broad category of cases, or individually on a case-by-case basis?
Another difficult issue, even in cases where it is clear which party is least well-informed, concerns the comparison between the benefits and the costs ofcorrecting the information asymmetry. The benefit of correcting the asymmetrywill vary from case to case, depending on the information involved, so it isdifficult to generalize. The costs of correcting the asymmetry clearly include thedirect costs of contracting around the original default rule (which will depend inturn on the procedures that are required to contract around a default rule, asdiscussed in subsection 14 below). The costs of correcting the asymmetry alsodepend on the risk that the parties will simply forget to contract around, therebyunintentionally leaving in force a default rule that was chosen not to be efficientfor these contracting parties.
In many markets there is heterogeneity among the potential parties on a singleside of a proposed transaction. For example, sellers may differ in the reliabilityof their products, or buyers may differ in the losses that they would suffer if theproduct they purchase fails to perform. At the outset of the transaction, thisinformation is often known to one party but not to the other--for example, eachbuyer may already know how much losses he or she would suffer, but sellersmay have no way of knowing how much any individual buyer has at stake. Thisis another example of information asymmetry.
When this asymmetry is present, contract law's default rules can again createincentives for one party to take actions that will reveal information to the otherparty. For example, if the default remedy for breach of contract provides for onlya small monetary payment, buyers who would lose a larger amount in the eventof breach may have an incentive to try to contract around that rule, bynegotiating for a clause stipulating a larger payment in the event of breach. Bycontrast, if the default remedy is already set at a larger payment, buyers whowould lose less may be the ones with an incentive to contract around that rule,by negotiating for a smaller stipulated damage to get a more favorable price). Inother words, the choice of default rule will determine which set of buyers--buyers with little at stake, or buyers with lots at stake--have an incentive to tryto contract around the default rule. As a result, sellers may be able to infersomething about the amount that any given buyer has at stake by observingwhether that buyer does or does not try to contract around the default rule. Andif buyers also differ in the gains they would get from transacting around thedefault rule, or in the costs they would face in doing so, the resulting equilibriummay depend critically on which default rule the law adopts. (The welfareconsequences of these differences are discussed below in subsection 8.)
These differences are often analyzed using signaling models from gametheory (see generally section 0550). If sellers have no information about howmuch any given buyer would lose in the event of breach, the result may be a pooling equilibrium in which all buyers are charged an identical price. But if thisasymmetry can somehow be overcome, the result may be a separatingequilibrium in which buyers who face large losses in the event of breach willhave a right to collect those losses, but will pay a higher price (to compensatethe seller for its greater potential liability), while buyers who face lower losseswill pay a lower price. One way to achieve this separating equilibrium is to selecta default rule that induces one of these classes of buyers to signal the amountthey have at stake, by their actions in trying to contract around the default rule.For formal models of this effect see, e.g., Ayres and Gertner (1989, 1992) , Johnston (1990) , Allen and Gale (1992) , Hviid (1996) .
Unfortunately, it is often difficult to assess the welfare implications of defaultrules that produce separating equilibria. Part of the difficulty is that, while therewill usually be efficiency gains from eliminating this information asymmetry, thatwill not always be the case. Gains could arise if information about eachindividual buyer's potential losses may let some sellers make customizedadjustments to the reliability of their products. Gains could also arise if thisinformation let sellers calculate a price that better reflected the true risk of sellingto that particular buyer, thus sending buyers the correct signals about whetherto purchase that product (see Ayres and Gertner (1989) , Quillen (1988) ). In somecases, though, information about the buyer's potential losses may be of norelevance to sellers--as, for instance, when sellers operate mass businesses thatcannot practicably adjust their actions or prices for individual buyers ( Danzig(1975) , Eisenberg (1992, pp 591-96) ). In such markets, there will be no efficiencygains at all from curing the information asymmetry. In still other markets, theremay be positive gains from curing the asymmetry but those gains may notoutweigh the costs of communicating the necessary information ( Bebchuk andShavell (1991) ).
Moreover, even when net efficiency gains are possible, the parties' privateincentives may lead them to act in a way that fails to achieve those gains. Forexample, if buyers face a price-discriminating monopolist who can charge higherprices to buyers who have more at stake in a particular transaction, buyers maybe reluctant to do anything to reveal the amount they have at stake for fear ofhaving to pay a higher price ( Wolcher (1989) , Johnston (1990) , Ayres andGertner (1992) ). In other cases, some buyers may get private benefits (such asa more favorable price) by distinguishing themselves from other buyers, thusleading to socially excessive signaling ( Rea (1984) , Aghion and Hermalin (1990) ).At present, it is difficult to generalize about when the signaling or separatingeffects of a default rule will improve overall efficiency (see Hviid (1996) ).
9. "Formalities" as Default Rules
Still another approach to selecting default rules aims entirely at ease ofadministration. That is, rather than trying to identify the default rule that would(1) be most efficient if left in place, or would (2) induce an efficient disclosure ofinformation, this approach aims for default rules that are easily administered bycourts, and easily learned and understood by contracting parties. In aninfluential early article, Fuller (1941) coined the term "formality" to refer to rulesdesigned to induce parties to specify their intentions in a way that could easilybe recognized by courts. In this respect, Fuller's rules were designed to force theparties to disclose information to the court, rather than (or in addition to) forcingone party to disclose information to the other contracting party.
For example, if the parties to a contract fail to specify the quantity of goodsthey intend to convey, many jurisdictions refuse to enforce any obligationwhatsoever, thus implicitly filling the gap with a quantity of "zero" (see Ayresand Gertner (1989, pp 95-96) ). This default rule clearly is not designed to matchwhat any contracting pair would have intended anyway. Instead, its purpose isto induce each pair to contract around the default rule by specifying the quantitythey prefer, thus sparing the court from having to guess about the parties'preferred quantity. Thus, unlike "majoritarian" or "market-mimicking" defaultrules, formalities are not necessarily intended to be left in place by the parties.
Unlike other "penalty" or "information-forcing" rules, however (seesubsections 5-8), formalities need not be slanted against any particular party. Forexample, the default rule for contracts that fail to specify any quantity of goodscould just as easily be set at any specific number, not necessarily zero. True, adefault rule of "zero" has certain administrative advantages over any othernumber: there is no way to breach an obligation to deliver a quantity of zero, andhence no need to ever measure damages. Still, the essential feature of a formalityis merely that some number be fixed in advance and be easily learnable by theparties, so that they can determine whether they will have to specify some othernumber in their contract. The formality could even be set at the number that mostparties would prefer (if that number were known), thus giving it one feature incommon with a "market mimicking" or "majoritarian" default rule. The only keyis that, since formalities are chosen for their simplicity and ease ofadministration, they will necessarily be "untailored" in the sense defined earlierin subsection 4. As a result, no matter what specific rule is adopted as thedefault formality, many (perhaps most) contracting pairs will find it in theirinterest to specify some other rule in their contract.
Because formalities are designed to be easily administered, they will at leasthave the virtue of reducing litigation costs in all cases when the parties fail tospecify some other rule in their contract, so the default rule remains in effect.This same ease of administration may also minimize another component oftransaction costs: the cost to the parties of becoming informed about the legalrule, and of predicting how that rule might be applied. Also, if the formality is notchosen to be deliberately unfavorable to either party, it may reduce the costsimposed when parties forget to specify some other rule in their contract: in thisway, formalities may reduce the cost of remaining uninformed about the law.(These two effects have offsetting influences on parties' incentives to becomeinformed about the law.)
The preceding subsections have discussed three bases on which default rulesmight be selected: "market-mimicking" or "majoritarian" defaults (subsections2-4), "information-forcing" or "penalty" defaults (subsections 5-8), and pureformalities (subsection 9). The choice among these various approaches dependsin part on the competence and capabilities of legal institutions. For example, thefirst two approaches--"market mimicking" or "majoritarian" defaults, and"information-forcing" or "penalty" defaults--place obvious demands on thoseinstitutions, either to identify the rule that would in fact be most efficient (if leftin place) for most contracting pairs, or to identify the default rule that wouldinduce the most efficient disclosure of information. These demands are onlyincreased if either of these approaches is to be adopted on a relatively "tailored"basis, requiring the legal decisionmakers to assess the effect on individualmarkets or even individual contracting pairs. Thus, it is obviously anoversimplification to analyze how either of these approaches would work if itwere to be administered by an omniscient legal decisionmaker. The real questionis how well either of these approaches will work when administered by actualcourts and legislatures.
Two views of this question have developed in the contracts literature. Thefirst begins by positing that certain rules--often, the rules that would have beenefficient in a first-best world with perfect legal institutions--are simply beyondthe capability of real legal institutions. This assumption is often made when therule in question depends on information that, although observable by one oreven both of the parties, cannot be demonstrated or verified publicly in court.For example, if the default remedy for breach were based on the net profits thebreacher made, this would require a court to be able to measure the breacher'srevenues and costs, which might in some cases require more accountingexpertise than real courts possess. Scholars often posit that such a rule wouldbe unworkable in order to focus their analysis on alternative rules: rules that,while they might be less efficient in a first-best world, place fewer demands onthe court or other legal decisionmaker (e.g., Hermalin and Katz (1993) ). For ageneral discussion and defense of this approach, see Schwartz (1992, pp 279-80 ,and 1993, pp 403-406 ).
By simply positing that certain rules are "unworkable," though, thesescholars implicitly assume an unfavorable balance between (1) the sum of allcosts that would be imposed if courts tried to apply the "unworkable" rule, withthe large number of errors that would entail; and (2) the sum of all costsassociated with whatever rule is second-best. Other scholars have attempted amore finely-grained analysis by explicitly modeling the costs associated withjudicial implementation of an "unworkable" rule. However, those costs dependin part on the nature and the probability of various errors courts might make inapplying such a rule, and there is no consensus (and, regrettably, no empiricaldata) on how best to model such an error function. For two applications of thisapproach to contract issues, see Hadfield (1994) , Hermalin en Katz (1991) and Allen and Gale (1992) . The general discussion of legal errors and uncertain rulesin other bodies of law (see section 0790) is also relevant here.
In some situations, the law's choice of default rules could alter contractingparties' beliefs or preferences, thereby changing the value of the costs andbenefits associated with different rules. For example, if the law adopts an impliedwarranty making sellers liable for all defects in their products (unless explicitlydisclaimed), this could conceivably lead buyers to increase the value they placeon such a warranty. At the same time, the adoption of the opposite default rule(no implied warranty) might lead buyers to the opposite view, if it caused themto reduce the value they place on such a warranty. In that event, it would bepossible for either default rule (an implied warranty, or the absence of an impliedwarranty) to be left in place by the parties, if the law's adoption of each defaultrule changed the parties' preferences sufficiently. It would also be possible forboth default rules to be perfectly efficient, at least when judged by thepreferences the parties would have once the law adopted either default rule.
The possibility that default rules might influence parties' preferences is onlyjust beginning to be explored. Experimental tests of this possibility can be foundin Schwab (1988) and Korobkin (1998) ; the latter also discusses many of thepotential policy implications. (Other experimental work regarding the effect onpreferences of legal rules generally is discussed in section 0570). Briefdiscussions of the implications for default rules in particular can also be foundin Charny (1991, pp 1835-40) ; and in Schwartz (1993, pp 413-15) , who refers todefault rules with this effect as "transformative default rules."
While default rules have received less attention in scholarship outside of lawand economics, there are several non-economic theories that deserve mention.For general discussions in the legal literature, see Charny (1991) , Barnett (1992) ,and Burton (1993) .
The first, and least well-developed (at least as a general theory), depends onthe identification of certain rules as morally superior, as a sort of "merit good."That is, just as it is sometimes argued that all citizens ought to have certaingoods (education, health care, etc.), it is sometimes said that all citizens oughtto have certain contract rights, such as the right to complete compensation inthe event of a breach; or that the law should especially promote certain kinds ofrelationships, such as those based on long-term cooperation. Arguments of thissort could be particularly compelling if the law's choice of default rule influencedcitizens' own beliefs about the value of certain relationships, as discussed abovein subsection 11.
A second non-economic theory rests on the idea of hypothetical consent.There is an entire family of non-economic theories that traces the binding forceof contracts to individual autonomy, and to the fact that the contractualobligation was freely chosen by the contracting party (see subsection 17 below).To be sure, these theories might seem to have little to say about the content ofdefault rules, which (by definition) fill in the content of obligations that were not explicitly chosen by the parties (see Craswell (1989a) ). However, some haveargued that, even in the absence of explicit consent, a party should still bebound to the rule that he or she would have consented to if he or she hadexplicitly negotiated an agreement on that point. Interestingly, this argumentconverges with the "market-mimicking" or "majoritarian" approach discussedearlier in subsections 2-4. As a result, some economically-oriented scholars havesought to rest their recommendations on this philosophic position, as well as onmore conventional economic grounds (e.g., Schwartz (1988, pp 357-60) ). In thephilosophical literature, however, it is disputed whether this invocation ofhypothetical consent carries any of the moral force of "real" consent, or whetherit adds anything to standard efficiency arguments. For discussions in the legalliterature of this issue, see Barnett (1992) , Charny (1991) , Coleman, Heckathornand Maser (1989) , Craswell (1992) , and the exchanges between Coleman (1980a , 1980b ) and Posner (1980 , 1981 ). In the philosophical literature, usefuldiscussions include Scanlon (1982) , Gauthier (1986) , and Brudney (1991) .
A third non-economic approach suggests that default rules should bedesigned to mimic the norms or customs that are already observed, either in thecommunity at large or in the parties' prior relationship. This approach is oftenemployed in the "relational contract" theory usually identified with Ian Macneil(1978 , 1980 , 1981 ); see also Brown and Feinman (1991) , Feinman (1993) , and Craswell (1993b) . In addition, some scholars working from philosophical theoriesof individual autonomy and consent have seen the prevailing norms andcustoms as acceptable sources of default rules, at least in the absence of anyexplicit agreement to the contrary ( Barnett (1992) , Burton (1993) ). In particularcases, this approach could of course converge with any of the economictheories discussed here, depending on whether the norms or customs happenedto be efficient (see the general discussion in section 0800). Among the issuesraised by this approach are questions about the manner in which the norms orcustoms are to be identified (see Bernstein (1996, pp 1787-95) ; Craswell (1998) ),and about the ability of courts or other legal decisionmakers to properly carryout this identification (see subsection 10 above).
A default rule, by definition, leaves parties free to specify some other rule togovern their relationship if they so choose. There is, however, very littlescholarly analysis of the steps the parties must take if they wish to specify someother rule. Most scholars implicitly assume that specifying some other rulewould require a valid provision to that effect in a valid contract. This couldsuggest that the rules governing those steps can be left to the law of contractformation and contract interpretation, two bodies of law that are addressed atmore length in sections 4300 and 4400.
Unfortunately, those bodies of law are themselves among the least analyzedportions of contract law. Moreover, the choices made by a legal system inselecting the rules of contract formation and interpretation can both beinfluenced by, and exert an influence upon, the choice of the original defaultrule. Subsections 14 and 15 briefly discuss some of the interactions betweenthese sets of rules.
Any attempt to displace a default rule will normally have to satisfy (at aminimum) all the requirements of an enforceable contract. For example,depending on the legal regime, the agreement may (or may not) have to be inwriting, and it may (or may not) have to be supported by consideration. Theserequirements, and the other rules governing contract formation, are discussedat more length in section 4300.
Should there be extra procedural requirements for parties who wish tocontract around a default rule? If the default rule is presumptively valid, andespecially if it is presumptively valid for moral or other non-economic reasons,it might be argued that parties who wish to choose a presumptively less validrule should have to take extra steps, if only to ensure (and to demonstrate to acourt) that this is what the parties really intend. For example, it might be arguedthat one party should not be able to displace a default rules by means of a singleclause buried in a 30-page document that the other party will not normally botherto read. Arguments of this sort are often made in connection with standard formcontracts and the legal doctrine of unconscionability (see section 4100). Someof the links with standard default rule analysis are noted briefly in Craswell(1993a, pp 12-14) .
Indeed, arguments of this sort may fit best in connection with "information-forcing" or "penalty" default rules. As discussed earlier (see subsection 6), such rules may be deliberately designed to induce the better-informed party tocontract around the default rule, in the hope that the process of contractingaround the rule will itself give the other party more information about the rightsthat he or she now has. The efficacy of this approach, however, depends onwhether the process of contracting around the rule really does inform the otherparty--and this, in turn, depends on the procedures that are required to contractaround the rule. Again: if a default rule can be altered merely by inserting aclause in a 30-page contract that nobody ever reads, the process of contractingaround the default will not inform the other party at all.
On the other hand, extra procedural requirements will usually increase thetransaction costs required to contract around the default rule. For example, ifaltering a default rule requires the clause in question to be pointed out andexplained to the other party, the process of contracting around the default mightindeed increase the other party's information. However, this requirement mayalso be costly to satisfy, especially if a single contract alters a large number ofdefault rules, and if the other party has no desire to sit through a lengthy set ofexplanations. In that event, the cost of altering the default rules via thisprocedure might turn out to be effectively prohibitive, so many parties wouldend up leaving the original default rules still in force. If so, this would requirereevaluation of the original decision to select a default rule that was designed not to be left in force (that is, a default rule that was chosen to be inefficientprecisely in order to induce the parties to contract around it). In this way, therules governing the process of contracting around a default rule can themselvesaffect the principles on which the original default rule should be chosen.
Contracting parties often use language which appears to address an issue thatwould otherwise be governed by a default rule, but which is vague orambiguous, and thus requires interpretation. For example, imagine a contractproviding that the remedy to be paid in the event of a breach should be anamount that would "reasonably compensate" the nonbreacher. Should thecourts treat this vague language as leaving a gap in the contract, to be filled bythe normal default remedy? Or should they instead apply some other set ofprinciples?
In conventional legal terminology, default rules are said to apply only whenthere is an actual gap in the contract, while questions of vague contractuallanguage are usually described as questions of interpretation. There is, however,no consensus (and very little theory) on what should count as a "gap" (see Ayres and Gertner (1989, pp 119-20) ). Fortunately, this distinction often will notmatter, because the approaches used to interpret contracts (see section 4400)have much in common with the approaches used to select default rules. In manycases, for example, vague or ambiguous language is interpreted so as to fitwhatever the parties probably would have agreed to if they had discussed thematter, thus producing the same result as the "majoritarian" or "market-mimicking" default rules discussed in subsections 2-4 (see also Goetz and Scott(1985) ). In other cases, vague or ambiguous language is interpreted against theparty who drafted it, just as in the case of a "penalty" or "information-forcing"default rule designed to induce more careful and explicit communication (seesubsections 5-9).
Notice, though, that some decision must still be made to determine when theparties have taken enough steps to make their language sufficiently clear toavoid the rule construing ambiguities against the drafter, or when their languageis sufficiently clear to avoid being interpreted in accordance with the court's ideaof what most parties would have wanted. This is the question of precisely howfar the parties must go to contract around an otherwise applicable default rule(or, in this case, around an otherwise applicable rule of interpretation). Asdiscussed in subsection 14, whatever requirements the law adopts here will haveto be considered in deciding what default rule (or what interpretive presumption)to adopt in the first place.
The analysis of default rules takes it as given that contracts are enforceable, andconcerns itself with determining the content of the enforceable obligation. Thissubsection addresses the opposite issue: given a contract of (let us assume)definite content, when should that contract be legally enforceable?
One branch of this question asks whether contracts should be enforced bya legal system, or whether non-legal enforcement mechanisms might be superior(see, e.g., Bernstein (1992 , 1996 ), Charny (1991) ). Non-legal enforcementmechanisms could include arbitration panels or trade association boards thatfunction much like a court, except for not being backed by official statesanctions. They could also involve much less formal mechanisms, such as thethreat of withholding business from anyone who had broken a promise in thepast. In both of these respects, they are similar to the non-legal mechanisms thatmight be used to enforce any other obligation (see sections 0780 and 0800), withwhich this literature has much in common.
By contrast, most of the literature on whether contracts ought to be enforcedtypically abstracts from the choice of the enforcement mechanism, and focusesinstead on the question of whether (and why) contracts ought to be enforced byany mechanism whatsoever. Non-economic theories are discussed briefly insubsection 17 below, followed by a discussion of the law-and-economicsliterature in subsections 18-19.
Moral philosophers have written extensively about the question of whether (orwhy) a promise should be morally binding. Some of their theories rest onutilitarian accounts that have much in common with the economic theories; thesewill be discussed in subsections 18-19. This subsection focuses instead on thenon-utilitarian or non-instrumental accounts. In the legal literature, generalsurveys of these accounts can be found in Atiyah (1981) , Barnett (1986) , Craswell (1989a, pp 491-503) , and Gordley (1991) .
One family of theories derives the obligation to keep one's promise fromconsiderations of individual freedom and moral autonomy. For example, Fried(1981) argues that if the law did not allow individuals to bind themselves withrespect to their future conduct, it would thereby fail to respect the individuals'status as autonomous moral agents. (For a variant account of when individualautonomy requires that promises be enforced, see Barnett (1986) .) Theseaccounts do not imply that promises ought to be kept in every circumstance: thistheory, like all of the others discussed in this subsection, allows for thepossibility of a set of implied conditions or implied excuses. Instead, the goal ofthese theories is to explain why promises are prima facie binding, and this familyof theories rests that binding force on the promise's status as the voluntarycommitment of a autonomous moral agent.
Another, different set of theories rests the obligation to keep a promise onthe substance of what has been promised. Perhaps the best-known theories inthis set focus on harm to others: if one's promise has led others to change theirbehavior in reliance on the promise, so that they would now be injured if thepromisor failed to perform, that promise is binding because it is wrong to harmother agents. Making a promise and then failing to keep it might also be seen asa form of lying, for such a promisor has misled others about the future. Othertheories in this set focus on reciprocity or restitution: if one has received abenefit and promised to pay for it, that promise is binding because it is only rightto pay for the benefits one has received ( Atiyah (1981, pp 34-36) ). Under sometheories, the fairness of the terms of the promise is also relevant to whether thepromise is binding (e.g., Gordley (1995) ).
Under any of these substance-based theories, the promise's status as thevoluntary commitment of a free moral agent is less significant, for the agent is(in some sense) promising to do what he or she ought to be doing anyway.Indeed, under many of these theories, if there is no independent reason for theagent to perform the action--e.g., if the agent has received no benefits, and if noone has yet relied on the promise--the obligation to keep the promise is seen asweaker or non-existent. These theories thus have difficulty explaining why theobligation should be stronger in the case of an individual who has made apromise than it is in the case of an individual who has acted identically but hasexplicitly disclaimed any promise (e.g., "I think I'm going to do x, but I warn youthat I'm not promising to do it, so you should rely only at your own risk," or"you can confer those benefits on me if you want, but I warn you now that Ihave no intention of paying you anything for them"). Atiyah (1981, pp 184-202) has argued that an explicit promise could serve as evidence, perhaps evenconclusive evidence, that the underlying obligation really is a just one thatought to be enforced. If one asks why an individual's promise ought to be givensuch evidentiary weight, however, it is difficult to avoid falling back either onautonomy-based explanations of the sort discussed earlier in this subsection, oron utilitarian theories of the sort to be discussed in subsections 18 and 19.
Economic theories of why promises should be enforced can be divided into twocategories. One set of theories, to be discussed in this subsection, focuses onthe utility created by the eventual performance of the promise. The other set, tobe discussed in subsection 19, focuses on the utility created by the ex anteincentives that the promise sets up.
The first economic argument for enforcing promises rests on the utility thatwill be produced when the promise is eventually performed. Since voluntarytransactions generally increase the welfare of all parties to the transaction,whenever a promise is voluntary it could be argued that welfare will usually beincreased if the promise is carried out. Viewed in these terms, the economicargument for enforcing promises is very similar to the economic argument forfree exchange and free markets generally (see section 5000).
This argument, however, could imply that a promise should not be bindingif conditions have changed sufficiently that the promised transaction would notincrease both parties' welfare. For example, if a promisor's costs have increasedto the point where it is no longer efficient for him or her to perform the promise,this theory could suggest that the promisor ought to be released with noobligation to pay any damages at all, because this particular efficiency rationalefor enforcing promises no longer applies. Under this theory, the only way tosave the prima facie obligation to keep one's promises (even after conditionshave changed) is to argue for the virtues of a general rule over a case-by-caseinquiry into the efficiency of any given transaction--in philosophical terms, byshifting from act-utilitarianism to rule-utilitarianism. That argument, in turn, mustrest on empirical claims about the ability of courts to determine whether andwhen performance of the promise was still efficient.
More generally, there is an important difference between permitting free exchange and permitting (or enforcing) binding promises. An exchange can takeplace instantaneously, but a promise necessarily involves a commitment to actin a certain way at some time in the future. Once this temporal elements isrecognized, it can be seen that enforcing promises does not simply transferexisting goods from one owner to another. Instead, the enforcement of promisescreates a new good: it allows people to exchange "wheat to arrive on September1" (for example), where without a binding promise they could only exchangeactual bushels of wheat. Viewed in these terms, the economic case forenforcement rests on the proposition that this new good (i.e., the goodrepresented by a future commitment) is a socially useful good which peoplefrequently will want to exchange. To explain why such a commitment is usefuland valuable, economists have focused more on the ex ante effects created bysuch a commitment, as discussed in subsection 19 below.
Most economic justifications for enforcing promises have focused on the ex anteeffects that would be created by a general rule of enforceability. In essence,these economic theories analyze promises as effectuating a present transfer notof goods and services, but of rights and duties. (For a similar view expressed innon-economic terms, see Barnett (1986, pp 291-300) .) Such a transfer of rightsand duties may itself produce efficiency gains, independently of whether it isefficient to actually carry out the promise.
The efficiency gain that is analyzed most often stems from the effect on thepromisee's incentive to rely on the promised behavior (see Goetz and Scott(1980) ). If the law denied promisees any compensation whenever a promisorcould show that performance of the promise had become inefficient, this wouldshift more of the risk of a change in conditions to the promisee, and thus wouldreduce the promisee's incentive to rely. To be sure, if the law instead guaranteesthat promisees will be compensated whenever the promisor fails to perform, thiscould create incentives for excessive reliance on the part of the promisee (see thediscussion of reliance incentives in connection with remedies for breach, insection 4600). If the incentives for excessive reliance can be constrained by otherlegal doctrines, however, the net effect on the promisee's reliance incentivescould still be positive. The argument that enforceability is needed to induce oneparty to take the risky step of beginning his or her own performance (when theother party's return performance is not due until later) is really a special case ofthis argument, since beginning performance is one of the many ways in whichparties may rely on a contract.
More generally, there are many other ex ante effects that could be producedby a rule requiring compensation even when conditions have changed to makeperformance unprofitable. If the promisor and promisee differ in their attitudestoward risk, a rule requiring compensation may achieve a better allocation of riskbetween the two parties (see, e.g., Polinsky (1983) ). A rule requiringcompensation may also increase the promisor's incentive to tell the truth aboutthe conditions that will affect his or her performance, thus increasing thepromisee's information about those conditions (see Shavell (1991) , Craswell(1989b) , Katz (1996, pp 1289-91) ). A compensation requirement may also give thepromisor an incentive to affect those conditions directly, if the probability ofperformance rests on factors that are within the promisor's control. Generallyspeaking, all of the economic analysis of efficient remedies for breach (seesection 4600 below) is relevant here, since to make a promise enforceable is toset a non-zero remedy for the promise's breach. The less-developed economicanalysis of the conditions under which formation of a binding contract shouldbe inferred (see section 4300) is also relevant here.
Richard Craswell is a professor of law at the Stanford Law School. Jack L.Goldsmith, Eric A. Posner, Cass R. Sunstein, and two anonymous refereesprovided helpful comments. Financial support was provided by the Lynde andHarry Bradley Foundation and the Sarah Scaife Foundation at the University ofChicago Law School.
Aghion, Philippe and Hermalin, Benjamin (1990) , 'Legal Restrictions on Private Contractscan Enhance Efficiency', 6 Journal of Law, Economics, and Organization , 381-409.
Allen, Franklin and Gale, Douglas (1992) , 'Measurement Distortion and MissingContingencies in Optimal Contracts', 2 Economic Theory , 1-26.
Ayres, Ian and Ayres, Robert (1992) , 'Strategic Contractual Inefficiency and the OptimalChoice of Legal Rules', 101 Yale Law Journal , 729-73.
Ayres, Ian (1993) , 'Preliminary Thoughts on Optimal Tailoring of Contractual Rules', 3 Southern California Interdisciplinary Law Journal , 1-18.
Ayres, Ian and Gertner, Robert (1989) , 'Filling Gaps in Incomplete Contracts: An EconomicTheory of Default Rules', 99 Yale Law Journal , 87-130.
Baird, Douglas G. and Weisberg, Robert (1982) , 'Rules, Standards and the Battle of theForms: A Reassessment of '2-207', 68 Virginia Law Review , 1217-62.
Barnes, David W. and Stout, Lynn A. (1992) , Economics of Contract Law, Minneapolis,West Publishing.
Barnett, Randy E. (1992) , 'The Sound of Silence: Default Rules and Contractual Consent', 78 Virginia Law Review , 821-911.
Bebchuk, Lucien Ayre and Shavell, Steven (1991) , 'Information and the Scope of Liabilityfor Breach of Contract: The Rule of Hadley v. Baxendale', 7 Journal of Law,Economics, & Organization , 284-312.
Bernstein, Lisa (1996) , 'Merchant Law in a Merchant Court: Rethinking the Code's Searchfor Immanent Business Norms', 144 University of Pennsylvania Law Review , 1765-1821.
Bernstein, Lisa (1993) , 'Social Norms and Default Rule Analysis', 3 Southern CaliforniaInterdisciplinary Law Journal , 59-90.
Bernstein, Lisa (1992) , 'Opting Out of the Legal System: Extralegal Contractual Relationsin the Diamond Industry', 21 Journal of Legal Studies , 115-57.
Birmingham, Robert L. (1968) , 'Legal and Moral Duty in Game Theory: Common LawContract and Chinese Analogies', 18 Buffalo Law Review , 99-117.
Brown, Patricia A. and Feinman, Jay M. (1991) , 'Economic Loss, Commercial Practices,and Legal Process: Spring Motors Distributors, Inc. v. Ford Motor Company', 22 Rutgers Law Journal , 301-60.
Charny, David (1991) , 'Hypothetical Bargains: The Normative Structure of ContractInterpretation', 89 Michigan Law Review , 1815-79.
Charny, David (1990) , 'Nonlegal Sanctions in Commercial Relationships', 104 Harvard LawReview , 375-467.
Coleman, Jules, Heckathorn, Douglas D. and Maser, Steven M. (1989) , 'A BargainingTheory Approach to Default Provisions and Disclosure Rules in Contract Law', 12 Harvard Journal of Law and Public Policy , 639-709.
Craswell, Richard (1989a) , 'Contract Law, Default Rules, and the Philosophy of Promising', 88 Michigan Law Review , 489-529.
Craswell, Richard (1989b) , 'Performance, Reliance, and One-Sided Information', 18 Journalof Legal Studies , 365-401.
Craswell, Richard (1992) , 'Efficiency and Rational Bargaining in Contractual Settings', 15 Harvard Journal of Law and Public Policy , 805-37.
Craswell, Richard (1993a) , 'Property Rules and Liability Rules in Unconscionability andRelated Doctrines', 60 University of Chicago Law Review , 1-65.
Craswell, Richard (1993b) , 'The Relational Move: Some Questions from Law andEconomics', 3 Southern California Interdisciplinary Law Journal , 91-114.
Craswell, Richard and Schwartz, Alan (eds.) (1994) , Foundations of Contract Law , NewYork, Oxford University Press.
Craswell, Richard (forthcoming, 1998) , 'Do Trade Customs Exist?', in Krauss, Jody andWalt, Steven, The Jurisprudential Foundations of Corporate and Commercial Law, Cambridge, Cambridge University Press.
Danzig, Richard (1975) , 'Hadley v. Baxendale: A Study in the Industrialization of the Law', 4 Journal of Legal Studies , 249-84.
D'Ursel, Laurent (1985) , 'L'Analyse Économique du Droit des Contrats' [EconomicAnalysis of Contract Law], 14 Revue interdisciplinaire d'études juridiques , 45-88.
Eisenberg, Melvin Aron (1992) , 'The Principle of Hadley v. Baxendale', 80 California LawReview , 563-613.
Farber, Daniel A. (1983) , 'Contract Law and Modern Economic Theory', 78 NorthwesternUniversity Law Review , 303-339.
Feinman, Jay M. (1983) , 'Critical Approaches to Contract Law', 30 UCLA Law Review, 829-60.
Feinman, Jay M. (1990) , 'The Significance of Contract Theory', 58 University of CincinnatiLaw Review , 1283-1318.
Feinman, Jay M. (1993) , 'Relational Contract and Default Rules', 3 Southern CaliforniaInterdisciplinary Law Journal , 43-58.
Gillette, Clayton P. (1993) , 'Cooperation and Convention in Contractual Defaults', 3 Southern California Interdisciplinary Law Journal , 167-88.
Gillette, Clayton P. (1990) , 'Commercial Relationships and the Selection of Default Rulesfor Remote Risks', 19 Journal of Legal Studies , 535-81.
Goetz, Charles J. and Scott, Robert E. (1985) , 'The Limits of Expanded Choice: An Analysisof the Interactions Between Express and Implied Contract Terms', 73 California LawReview , 261-322.
Goetz, Charles J. and Scott, Robert E. (1980) , 'Enforcing Promises: An Examination of theBasis of Contract', 89 Yale Law Journal , 1261-1322.
Goldberg, Victor P. (ed.) (1989) , Readings in the Economics of Contract Law , Cambridge,Cambridge University Press.
Goldberg, Victor P. (1984) , 'An Economic Analysis of the Lost-Volume Seller', 57 SouthernCalifornia Law Review , 283-97.
Hadfield, Gillian K. (1994) , 'Judicial Competence and the Interpretation of IncompleteContracts', 23 Journal of Legal Studies , 159-84.
Harris, Donald (1983) , 'Contract as Promise--A Review Article Based on Contract asPromise: A Theory of Contractual Obligation by Charles Fried', 3 International Reviewof Law and Economics , 69-77.
Harris, Donald and Veljanovski, Cento G. (1984) , 'The Use of Economics to Elucidate LegalConcepts: The Law of Contract', in Daintith, Terence and Teuber, Gunther (eds.), Lawand the Social Sciences, Florence, European University Institute.
Hermalin, Benjamin E. and Katz, Michael L. (1991) , 'Moral Hazard and Verifiability: TheEffects of Renegotiation in Agency', 59 Economica , 1735-1753.
Hermalin, Benjamin E. and Katz, Michael L. (1993) , 'Judicial Modification of ContractsBetween Sophisticated Parties: A More Complete View of Incomplete Contracts andTheir Breach', 9 Journal of Law, Economics, & Organization , 230-255.
Hillman, Robert A. (1988) , 'The Crisis in Modern Contract Theory', 67 Texas Law Review ,103-136.
Holderness, Clifford G. (1985) , 'A Legal Foundation for Exchange', 14 Journal of LegalStudies , 321-344.
Horwitz, Morton J. (1974) , 'The Historical Foundations of Modern Contract Law', 87 Harvard Law Review , 917-956.
Hviid, Morten (1996) , 'Default Rules and Equilibrium Selection of Contract Terms', 16 International Review of Law and Economics , 233-45.
Johnston, Jason Scott (1990) , 'Strategic Bargaining and the Economic Theory of ContractDefault Rules', 100 Yale Law Journal , 615-64.
Köhler, Helmut (1989) , 'Zur Ökonomischen Analyse der Regeln über die Geschäftsgrundlage'[On the Economic Analysis of the Rules of the Implicit Basis of a Contract], in Ott,Claus and Schäfer, Hans-Bernd (eds.), Allokationseffizienz in der Rechtsordnung , Berlin,Springer, 148-162.
Korobkin, Russell (1998) , 'The Status Quo Bias and Contract Default Rules', 83 Cornell LawReview , 608-87.
Kronman, Anthony T. (1980) , 'Contract Law and Distributive Justice', 89 Yale LawJournal , 472-511.
Kronman, Anthony T. (1985) , 'Contract Law and the State of Nature', 1 Journal of Law,Economics & Organization , 5-32.
Kronman, Anthony T. and Posner, Richard A. (eds.) (1979) , The Economics of ContractLaw , Boston, Little, Brown and Company.
Lowry, S. Todd (1976) , 'Bargain and Contract Theory in Law and Economics', 10 Journalof Economic Issues , 1-22.
Macneil, Ian R. (1980) , The New Social Contract: An Inquiry Into Modern ContractualRelations , New Haven, Yale University Press.
Macneil, Ian R. (1981) , 'Economic Analysis of Contractual Relations: Its Shortfalls and theNeed for a 'Rich' Classificatory Apparatus', 75 Northwestern University Law Review ,1018-63.
Macneil, Ian R. (1978) , 'Contracts: Adjustment of Long-Term Economic Relations UnderClassical, Neoclassical, and Relational Contract Law', 72 Northwestern University LawReview , 854-905.
Marini, Giovanni (1990) , Promessa ed affidamento nel diritto dei contratti , [Promise andReliance in Contract Law], Napoli, Jovene.
Perron, Edgar du (1990) , 'De rechtseconomische analyse van het verbintenissenrecht' [ALaw-and-Economics Analysis of the Law of Obligations], 39 Ars Aequi , 770-776.
Poughon, Jean-Michel (1990) , 'Une constante doctrinale: l'approche économique ducontract', 12 Droits. Revue française de théorie juridique , 47-59.
Romani, Franco (1985) , 'Appunti sull' analisi economica dei contratti' [Notes about theEconomic Analysis of Contract Law], Quadrimestre , 15-29.
Schwab, Stewart J. (1988) , 'A Coasean Experiment on Contract Presumptions', 17 Journalof Legal Studies , 237-268.
Schwartz, Alan (1993) , 'The Default Rule Paradigm and the Limits of Contract Law', 3 Southern California Interdisciplinary Law Journal , 389-419.
Schwartz, Alan (1992) , 'Relational Contracts in the Courts: An Analysis of IncompleteAgreements and Judicial Strategies', 21 Journal of Legal Studies , 271-318.
Scott, Robert E. (1990) , 'A Relational Theory of Default Rules for Commercial Contracts', 19 Journal of Legal Studies , 597-616.
Simpson, A.W.B. (1979) , 'The Horwitz Thesis and the History of Contracts', 46 Universityof Chicago Law Review , 533-601.
Symposium , 'Default Rules and Contractual Consent', 3 Southern CaliforniaInterdisciplinary Law Journal , 1-444.
Verkerke, J. Hoult (1995) , 'An Empirical Perspective on Indefinite Term EmploymentContracts: Resolving the Just Cause Debate', 1995 Wisconsin Law Review , 837-917.
Wang, William K.S. (1982) , 'Reflections on Contract Law and Distributive Justice: A Replyto Kronman', 34 Hastings Law Journal , 513-527.
Witt, Ulrich (1986) , 'Evolution and Stability of Cooperation without EnforceableContracts', 39 Kyklos , 245-266.
Wonnell, Christopher T. (1986) , 'Contract Law and the Austrian School of Economics', 54 Fordham Law Review , 507-543.
Atiyah, P. S. (1979) , The Rise and Fall of Freedom of Contract, Oxford, Clarendon Press.
Atiyah, P. S. (1981) , Promises, Morals, and Law , Oxford, Clarendon Press.
Barnett, Randy E. (1986) , 'A Consent Theory of Contract', 86 Columbia Law Review , 269-321.
Brudney, Daniel (1991) , 'Hypothetical Consent and Moral Force', 10 Law and Philosophy ,253-70.
Buckley, F.H. (1988) , 'Paradox Lost', 72 Minnesota Law Review , 775-827.
Burton, Steven J. (1993) , 'Default Principles, Legitimacy, and the Authority of a Contract', 3 Southern California Interdisciplinary Law Journal , 115-66.
Coleman, Jules L. (1980a) , 'Efficiency, Exchange, and Auction: Philosophic Aspects of theEconomic Approach to Law', 68 California Law Review , 221-49.
Coleman, Jules L. (1980b) , 'Efficiency, Utility, and Wealth Maximization', 8 Hofstra LawReview , 509-51.
Fried, Charles (1981) , Contract as Promise: A Theory of Contractual Obligation , Cambridge,Massachusetts, Harvard University Press.
Fuller, Lon L. (1941) , 'Consideration and Form', 41 Columbia Law Review , 799-824.
Gauthier, David Peter (1986) , Morals by Agreement , Oxford, Clarendon Press.
Gordley, James (1995) , 'Enforcing Promises', 83 California Law Review , 547-614.
Gordley, James (1991) , The Philosophical Origins of Modern Contract Doctrine, Oxford,Clarendon Press.
Isaacharoff, Samuel (1996) , 'Contracting for Employment: The Limited Return of theCommon Law', 74 Texas Law Review , 1783-1812.
Katz, Avery (1996) , 'When Should an Offer Stick? The Economics of Promissory Estoppelin Preliminary Negotiations', 105 Yale Law Journal , 1249-1309.
Muris, Timothy J. (1983) , 'Cost of Completion or Diminution in Market Value: TheRelevance of Subjective Value', 12 Journal of Legal Studies , 379-400.
Perloff, Jeffrey M. (1981) , 'Breach of Contract and the Foreseeability Doctrine of Hadleyv. Baxendale', 10 Journal of Legal Studies , 39-63.
Polinsky, A. Mitchell (1983) , 'Risk Sharing Through Breach of Contract Remedies', 12 Journal of Legal Studies , 427-44.
Posner, Richard A. (1981) , 'A Reply to some Recent Criticisms of the Efficiency Theoryof the Common Law', 9 Hofstra Law Review , 775-94
Posner, Richard A. (1980) , 'The Ethical and Political Basis of the Efficiency Norm inCommon Law Adjudication', 8 Hofstra Law Review , 487-507.
Posner, Richard and Rosenfield, Andrew M. (1977) , 'Impossibility and Related Doctrines inContract Law: An Economic Analysis', 6 Journal of Legal Studies , 83-118.
Quillen, Gwyn D. (1988) , 'Contract Damages and Cross-Subsidization', 61 SouthernCalifornia Law Review , 1125-41.
Rea, Jr., Samuel A. (1984) , 'Arm-Breaking, Consumer Credit and Personal Bankruptcy', 22 Economic Inquiry , 188-208.
Scanlon, Thomas (1982) , 'Contractualism and Utilitarianism', in Sen, Amartya andWilliams, Bernard (eds.), Utilitarianism and Beyond , Cambridge, Cambridge UniversityPress.
Schwartz, Alan (1988) , 'Proposals for Products Liability Reform: A Theoretical Synthesis', 97 Yale Law Journal , 353-419.
Shavell, Steven (1991) , 'An Economic Analysis of Altruism and Deferred Gifts', 20 Journalof Legal Studies , 401-21.
Wolcher, Louis E. (1989) , 'Price Discrimination and Inefficient Risk Allocation Under theRule of Hadley v. Baxendale', 9 Research in Law and Economics , 9-31.
© Copyright 1998 Richard Craswell