Nonstate Institutions

The Courts

The state is an important institution. However, when it is a monopoly institution, it will be a weak one. This of course does not imply that the state itself is weak: on the contrary, it is the strength that the state derives from a monopoly right to define property rights that undermines them. This is partly because politicians have such strong incentives to prey upon some sections of the population in order to attract the support of others, but even in the absence of predatory behavior property rights are weakened when the state has a monopoly over property-right specification. The (rather implausible) state which aims selflessly to maximize welfare by acting as an enterprise association will of necessity interfere with the free transfer of assets, and the exercise of economic initiative. This weakens property rights, and we have argued that in so doing, it will undermine economic efficiency.

It follows that a strong institutional framework must incorporate a non-state-intermediated means of establishing and transferring property rights. Provided it serves to undermine state predation, the exact form of the alternative institution is not necessarily important. Indeed, Acemoglu and Johnson's work (2005) is supportive of the thesis that, provided the state cannot simply steal assets, the precise mechanism by which they are transferred is not particularly important.

Notwithstanding these observations, the most important mechanism for establishing entitlements and transferring them is the court system. In the contractual context we divide the courts' functions into two categories, although there is some overlap between the two. The first function is enforcement. The courts enforce statutory contractual legislation (for example, prohibiting certain types of contract), and they enforce private bilateral contracts. Within the boundaries established by legislation, the courts act independently of the state, although they rely upon the support of the state to enforce their decisions.

The relationship between the state and the courts is central. A strong court system upon which individuals can rely for enforcement enhances their ability to commit to economic contracts, and hence strengthens property rights and increases economic efficiency.15 However, the state has a natural tendency arbitrarily to interfere in the operations of the courts in response to political exigencies. However noble its motives, in doing so it will weaken property rights and ultimately, welfare.16 In a strong institutional environment, the state's rapaciousness is ultimately checked by the law, which places limits upon the legislature's actions.17

If the only role of the courts was enforcement, it would be hard to understand why contracting parties ever went to court. If the outcome of a case was predictable, then they would avoid the costs of litigation by simply agreeing to settle outside the court.18 Indeed, most technical

15 We argue in n. 23 below that the effectiveness of the court in enforcing contracts depends upon the damages standard which it adopts. From a purely institutional perspective, expectation damages appear to be most efficient.

16 This is hardly a fresh insight. Hume (1777: 375) famously argued that 'Public utility requires that property should be regulated by general inflexible rules', and that an inevitable consequence of this is that laws 'deprive, without scruple, a beneficent man of all of his possessions, if acquired by mistake, without a good title; in order to bestow them on a selfish miser who has already heaped up immense stores of superfluous riches'.

17 See the discussion on page 59.

18 This is not to say that courts would be irrelevant. The anticipated costs of litigation have an effect upon incentives to break contracts: as MacLeod (2006) notes, economics assumes, either implicitly or explicitly, that legal actions will not occur: this is one of the important distinctions between legal and economic contract scholarship.19

So when parties know in advance how the courts will behave in every state of the world, bargaining in the shadow of the courts should prevent court actions from ever occurring. In practice though, contracts do not cover every possible contingency: in the real world, the parties to a contract cannot enumerate every possible state of the world and specify precisely the actions that they will take in each state.20 In other words, contracts are inevitably incomplete.21

Incomplete contracts are the subject of an important microeco-nomic literature.22 In most of this literature, the parties to a contract are unable for technological reasons to record their response to an event, although they appreciate that it is likely to be important. At a later stage agreement on the uncontracted contingency becomes the standards required by a contract have to be high enough to ensure that it is worth going to court for enforcement. Djankov, la Porta et al. (2003) show that court costs vary greatly across jurisdictions, and argue that they are inefficiently high in developing countries. A theoretical literature examines the effects of court verification costs upon the behavior of contracting parties: see, for example, Townsend (1978) and Gale andHellwig (1985).

19 The costs of litigation can be used to explain 'nuisance suits', brought by one party in the hope that the other will settle (see, for example, Bebchuk, 1988). Court litigation may then be an attempt to establish a reputation which will protect against such suits. This still leaves the majority of court actions unexplained.

20 Indeed, in cases where they expect to rely upon non-court intermediated contracts, they may choose not to do so. See the discussion of reputational contracting on page 52.

21 According to Schwartz (1992), even a contract which provides clear rules governing the interaction between the parties can be incomplete if its terms do not vary with the state of the world. Hence, a fixed price supply contract could be deemed incomplete if it fails to specify price variations in states of the world where one party makes or loses an unusually large amount of money. Schwartz shows that the courts will often complete these contracts, provided they can do so through the application of a general rule which can be used in future cases. This is a subtle point. A contract which the courts can complete in this sense could have been completed ex ante by the contacting parties. That they did not is probably evidence of intent: as we noted on page 58, and Schwartz acknowledges on page 317, even when it is motivated by efficiency concerns ex post legal activism can be ex ante inefficient. As Schwartz states, in some cases such activism reflects a wider social goal. From our perspective, whether it is motivated by ex post efficiency or by social concerns, court alteration of contracts with unambiguous consequences undermines property rights and hence reflects weaker institutions. See the discussion of the purposive State on page 43.

22 See, for example, Bolton and Scharfstein (1990), Grossman and Hart (1986), Hart and Moore (1995), Hart and Moore (1989), Hart and Moore (1988), Hart and Moore (1990), and Hart and Moore (1994). Hart (1995) provides a clear survey of the major ideas.

possible, and the parties bargain over their actions. The bargaining is anticipated up front and, to the extent that agents will be forced in bargaining to share the fruits of their labors, they are less willing to work hard.

The economic literature on incomplete contracts has advanced our understanding of the dynamics of contracting relationships, of the efficiency consequences of incomplete contracts, and of the consequences of contractual incompleteness for industry structure. But, in common with the economic literature on complete contracts, it assigns no active role to the courts: agents always anticipate in these models the results of court enforcement, and they bargain in the shadow of the court to obtain agreement, without ever entering into litigation.

In reality, bargaining of this type is stymied by genuine disagreement over the meaning of a contract, and hence over the likely actions of the courts. In these situations, litigation may be the best way to resolve a dispute. The second role of the courts is therefore to interpret ambiguous or incomplete contracts, and it is this which explains costly court actions.

How should a court interpret an ambiguous contract? Strong institutions facilitate the free exchange of property rights, and in particular, they allow agents to assume binding commitments.23 Hence, their decisions should not overrule the freely expressed will of the contracting parties at the time of contracting. When it is not clear from contract documentation what this was, the courts should attempt as far as possible to divine it.24 This might be accomplished through the

23 This discussion is related to a debate in the legal contracting literature: namely, what level of compensation should the courts require when a contract is broken? The law distinguishes between expectation damages, under which the promisee receives compensation for what would have been received under the promise; reliance damages, in line with harm or costs caused by the broken promise; and benefit damages, where the promisor pays the benefits received from breaking the promise. Scholars who interpret contracts in the liberal tradition as promises tend to argue morally for expectation damages (Fried, 1981); others, who argue that contracts reflect socially sanctioned modes of behavior, make a case for reliance and benefit damages (Atiyah, 1979). Of course, it is cheaper to break a promise when one faces only reliance or benefit damages. Since we argue that strong property rights require agents to be able to make credible long-term commitments, we suggest that expectation damages are a feature of strong institutions, and that benefit and reliance damages are in general associated with weak institutions.

24 Schwartz (1992: 271 n. 8) discusses this process. In a strong property rights environment, the courts base their decisions on what the parties probably expect of each other. As Schwartz notes, this is distinct from a decision based upon 'reasonable expectations', which is normative phrase, reflecting a view of how people ought to behave, rather than how they do. See also the discussion on page 191 of contract interpretation.

interpretation of a legal code to which the parties had access; or it might simply reflect the assessment that a reasonable and neutral third party would make. The latter approach is the essence of the common law approach to contracting which prevails in the United States and the United Kingdom. The common law uses social mores to adjudicate over uncontracted outcomes for which legislation does not exist. At least in theory,25 the court's role in the common law is therefore to discover the law, rather than to create it.26

So the courts have two roles: enforcement and interpretation. The court's enforcement role relies upon the cooperation of the state. In strong systems that are governed by the rule of law, the mere existence of the courts will be sufficient to preclude their use for enforcement when contractual meaning is clear. The courts therefore act, and their design therefore only matters, when an incomplete contract requires interpretation. In short, the structure of the courts matters because contracts are incomplete.

The enforcement activity of the courts is logically prior to their interpretation role. To see why, imagine that the courts were unable effectively to enforce agreements, or that in doing so they made completely arbitrary and unpredictable decisions. In this case, the parties to a contested contract would not use the courts to intermediate over incomplete or contested contracts: if neither believed that the courts were reasonable, neither could expect the court to uphold his interpretation. In this case the courts would never be used for enforcement, or for interpretation. The irrelevance of weak courts would not reflect contractual completeness: either a private sector alternative to the courts would arise or, more likely, economic activity would be greatly curtailed. The interpretative role of the courts therefore rests upon their legitimacy and effectiveness as an enforcement mechanism.

Our attempts to explain the importance of court structure are related to seminal work performed by Coase (1960), who discussed the role of institutions as devices for allocating property rights and costs. Coase argues that when property rights are clearly assigned

25 Legal positivists dismiss this as a fiction, claiming that laws are commands which reflect conscious decisions made by judges, sovereigns or legislatures. Two important works in this vein are Austin ([1832] 1995) and Hart ([1961] 1997). Oliver Wendell Holmes' contribution to this school of thought is discussed on page 191.

26 Hayek (1973) argues that property right institutions under the common law reflect tried and tested social mores and customs and that they are valuable because they codify existing practices, whose successes we can seldom adequately explain. His work suggests a possible bridge between North's emphasis (1990) upon social and cultural rules, and our concentration upon property rights.

and bargaining is costless, economic actors will always agree upon the welfare-maximizing course of action. He concludes that the assignment of property rights is important only when bargaining is impeded. This observation was captured in Stigler's famous (1966) formulation of the 'Coase Theorem': in the absence of transactions costs, institutions are not important.27

Demsetz (2003) criticizes the Coasian emphasis upon transactions costs, arguing that they should not be distinguished from other types of economic costs: property right assignments are an endogenous response to all relevant economic effects, and singling out transactions costs for special attention is appropriate only insofar as state actors have an advantage in resolving them. Hence, like us, Demsetz is concerned with the property rights institutions that support a particular allocation of economic resources, rather than with the costs of agreeing to performing a transfer between agents.

The courts affect economic outcomes either when they are ineffective or in some way circumscribed, or when contracts are incomplete. In either case, agents can be expected to search for extralegal mechanisms to establish and to trade property rights. This is the primary function that we identify with the modern investment bank and, at a more general level, it is the subject of the next subsection.

Extra-Legal Institutions Many forms of property rights are secured if the courts operate in a predictable fashion and without the interference of the state. Sometimes, however, the courts may be politically undermined: they may have little ability to make nuanced judgments, or they may be prevented from enforcing some contracts. Equally, the courts may be undermined by technological limitations. For example, it may be impossible to establish rights to some assets, such as the price-relevant information in which investment banks deal, because they lose their value as soon as they are publicly revealed. Entrepreneurs whose ability to contract is reduced for political or technical reasons will naturally attempt to create alternative, private sector, property rights

27 It seems that Coase regrets the ubiquity of his eponymous theorem. He argues (Coase, 1988: ch. 6) that mainstream economists have been misled by a poor understanding of the Coase theorem into assuming that institutional structure does not affect economic outcomes.

institutions for themselves. It is to these 'private law' institutions that we now turn.

We illustrate the importance of extra-legal institutions with an example which is a response to extreme contractual incompleteness. When several people decide to combine their skills in a productive activity, they may be unsure before they begin precisely what proportion of their time will be devoted to their various tasks. Similarly, it will be hard for them to predict which customers they will attract, to whom they will sell their product, and which of them will be responsible for the relationships. They will be aware that at some stage an unforeseen opportunity may arise, but they clearly will not know what it will be, and they will not anticipate its effects upon their working relationship.

In this situation, it is clearly unreasonable to expect the productive relationship to be governed by a formal contract. The people concerned will make minimal legal commitments, and they will commit to respect some non-contractual mechanism for making decisions. This type of arrangement is common: we usually refer to it as a firm. Coase (1937) first recognized that the firm is a natural response to contracting problems.28 Although firms have a legal existence, and their actions are to some extent circumscribed by statute,29 within the boundaries set by the law agents are free when they create a firm to adopt whichever authority structures they choose. In this sense, every corporation is a web of extra-legal contracts, and the corporation is therefore an extra-legal institution.

How does the corporation perform its enforcement role? We have already noted the demise of indentured servitude contracts: in their absence, the firm cannot compel an employee to act against her wishes. Similarly, it cannot imprison an employee. In fact, the employer's only sanction is to deny the employee something. The simplest way to do this is to fire her; a more subtle one is to deny her promotion. These sanctions impose a cost upon the employee whose severity varies depending upon circumstance. Firing an employee

28 Coase was ahead of his time. His work spawned a related literature from the late 1960s: see Alchian and Demsetz (1972), Williamson (1985), Jensen and Meckling (1976), and Cheung (1983).

29 By sinking costs to create standardized and predictable corporate law, the State probably facilitates the creation of the private property structures. Of course, there is a countervailing tendency to interfere for the usual reasons in corporate laws, as in others. This topic is important, but it is beyond the scope of this book.

is costly when she will struggle to find an equally attractive job;30 denying her promotion is costly when she cannot get it elsewhere.31

When an employee subjects herself to the extra-legal authority of the firm she relies upon the firm to keep promises to which it is not legally bound. Why does she trust the firm to reward her for keeping her side of the bargain, and to punish her only when she does not? The answer is at the heart of extra-legal contracting: the parties to extralegal contracts rely upon their reputations. The employee will only trust a firm which has in the past kept its promises. Firms that renege upon their promises cannot expect in the future to be trusted, and so will lose their ability to contract without the courts. To the extent that extra-legal contracts are necessary in its business, this harms the firm.

Hence extra-legal contracts are credible to the extent that it is costly to break them.32 The costs of reneging upon an agreement are often reputational. They therefore rely upon a future stream of reputationintensive business: if technological or legal advances lower the importance of reputation, reputationally enforced extra-legal contracts are more difficult to enforce.33

Reputations are created and maintained in the context of a repeated trading relationship. Such relationships are usually characterized by a mixture of formal legal contracting, and informal extra-legal relational contracts. Stewart Macaulay (1963) interviewed key business personnel and legal counsel for a number of (mostly Wisconsin) companies to document this type of relationship. He found that in the context of long-term continuing relationships, businessmen frequently committed themselves to significant exchanges quite casually.34 Since it

30 So the firm might optimally choose to pay the employee more than her outside option so as to add potency to the threat of dismissal. See Shapiro and Stiglitz (1984) and Stiglitz (1987).

31 This is a central idea in informationally opaque businesses such as investment banks. See Morrison and Wilhelm (2004), Morrison and Wilhelm (2005b), and chapter nine.

32 The classic economic exposition of extra-legal contracting underwritten by repeated trade is due to Klein and Leffler (1981). They show that contractual performance in a purely bilateral trading relationship requires prices to be far enough above cost.

33 See the discussion on page 58, and also the discussion of Bankers Trust on page 246 in chapter eight.

34 'For example, in Wisconsin requirements contracts—contracts to supply a firm's requirements of an item rather than a definite quantity—probably are not legally enforceable. [Of the people quizzed about this] none thought that the lack of legal sanction made any difference. Three of these people were house counsel who knew the Wisconsin law before being interviewed' (Macaulay, 1963: 60).

is costly and difficult to write formal contracts, this was probably a cheaper and more precise way to deal. For example, the counterparts to these deals appear to have benefitted from a high degree of flexibility.35

It seems clear from Macaulay's work that the contracts he studied were reputational.36 The agreements concerned were part of a long-term relationship. If either party had lost its reputation for abiding by the spirit of the relationship, both would have had to fall back upon the law. This would have reduced the precision of the agreements between them, and would have imposed substantial costs upon both parties. In contrast, Macaulay found that carefully planned legal contracts were in practice used only for large one-off transactions (such as the sale of the Empire State Building), and for those in which bargaining power was extremely skewed. In both cases the costs of reneging on informal contracts might, for at least one party, have outweighed the reputational benefits of complying with them. The parties thus fell back upon formal legal processes.

Interestingly, Macaulay's findings suggest that parties prefer where possible to use extra-legal contracts, rather than to rely upon the courts. Litigation is a last resort, and it is evidence of a breakdown in relational contracting. This point is developed by Macneil (1974,1978), who points to the inevitable incompleteness of long-term contracts. As he argues, reputationally enforced norms of behavior are a far more effective basis for an economic relationship than a series of discrete contracts based upon classical contract law.

Given the importance of repeat business to reputationally enforced contracts, it appears at first blush that their effectiveness must be limited to bilateral economic relations with frequent and extended interactions. If a large firm expects to deal with a customer only once, it will face an overwhelming temptation to break the implicit agreements between them. But this is true only if the reputation that it has with one counterpart is not transferable to others; if in ripping off one customer it damages its chances of dealing with others, the

35 Macaulay's subjects were happy to break contracts, and did so frequently: 'all ten of the purchasing agents asked about cancellation of orders once placed indicated that they expected to be able to cancel orders freely subject to only an obligation to pay for the seller's major expenses such as scrapped steel' (Macaulay, 1963: 61).

36 He expands upon this point in a later article (Macaulay, 1985: 468): 'People in business often find that they do not need contract planning and contract law because of relational sanctions. There are effective private governments and social fields, affected by but seldom controlled by the formal legal system.'

firm will be more circumspect in its dealings with small and relatively insignificant customers.

Complex exchanges that rely upon implicit agreements are possible only if the counterparties can rely upon the reputational mechanisms which enforce them. It is therefore in the firm's interests (and society's) to find some way of extending the reach of its reputation as far as possible. This will be accomplished automatically if the firm's poor dealings with one counterpart are observable to others. For example, because the firm's employees form a close-knit group who observe one another's dealings with the firm, the firm's reputation as an honest employer is at risk whenever it deals with any of them. However, it may be harder for the firm's customers to see each other's dealings. This could reflect technological problems, such as geographic dispersion or an inability to document convincingly the details of their dealings with the firm; or it could be because proving malfeasance would require the release of commercially confidential information. In either case, the firm will struggle to build a single reputation with a sufficiently broad base of customers to sustain relational contracting.

The natural response to this problem is the professional body. A professional body that simply gathers information from customers and publicizes it automatically extends the reach of its members' reputations. In doing so, it enhances their ability to contract. A broad range of professional bodies may fill this role. For example, when information is commercially confidential, it may be necessary to gather it in secrecy and then simply formally to reprimand any miscreants. This is one of the functions of a self-regulatory organization. We relate it to the operation of the syndicate in chapter three.

Bernstein (1992, 2001) discusses the operation of professional bodies as noncourt contract enforcers in the diamond and the cotton industries, respectively. Her evidence is consistent with a role for the professional bodies as clearing houses for information. For example, she notes that failure to comply with a ruling of a cotton industry professional body results in expulsion from the body. This harms local and international business, not least because a firm expelled from a professional body is patently no longer bound by its rules. The associated costs are a sufficient threat to ensure compliance. Like Macaulay, Bernstein stresses the value of flexibility.37 She stresses the importance of both social links between industry participants, and also of what

37 For example, she states that 'although cotton industry trade rules and contracts provide clearly specified delivery times, it is understood that when making an adjustment is not too costly to the merchant, the adjustment will be made' (Bernstein, 2001: 23).

she refers to as 'information-intermediary based trust', which concerns formal methods for transmitting reputation information, such as circulars reporting the names of transactors who refused to arbitrate or to comply with an award against them. We identify a similar role for investment bank syndicates in chapter three.

Bernstein's evidence is contemporary. Milgrom, North, and Weingast (1990) argue that similar methods were employed in medieval Europe. They are concerned with the enforcement of contracts between agents who cannot call upon the law, who are very geographically dispersed, and for whom communication is very hard. They present a formal model which suggests that in this circumstance, a third party 'law merchant' can serve as a store of trader reputation. The law merchant maintains a database of information about traders, which they can access before transacting in exchange for a fee. The merchant fines agents who deal with counterparts against whom a judgment is outstanding, and records a failure to pay. Agents who paid up front for information revelation can take their disputes to the merchant for binding arbitration; failure to settle is recorded by the merchant. Milgrom, North, and Weingast show that with these rules, merchants will prefer to pay for access to the database, and will settle any judgments in order to avoid the costs of future lost business. They argue that this was effectively the system that governed the operation of the European Champagne fairs of the thirteenth century which dominated commercial activity in the entire Western world before the advent of the geographically dispersed nation-state.

The law merchant in Milgrom, North, and Weingast's work is the analog to the professional body which we analyzed above. By maintaining records about its counterparts, and providing access to them for a fee, it provides its members with a way of exposing their reputations in trade. Without this exposure, they would be unable to transact at all.

Greif, Milgrom, and Weingast (1994) examine another medieval example of a professional body. Between the tenth and fourteenth centuries, traveling European merchants found it hard to enforce the contracts that they wrote with the rulers of the cities and states in which they operated. Greif, Milgrom, and Weingast give several examples of merchants who were cheated, or even attacked. Trade in the absence of formal international law would therefore have been impossible without a system of private enforcement:38 Greif et al. show that the

38 Benson (1989) discusses the evolution of the Law Merchant, which governed commercial contracts in this period. He states (p. 647) that 'This body of law was merchant guild evolved as a professional body that provided the requisite enforcement. It maintained information about the treatment of its members, and enforced sanctions against a counterpart who cheated one of them. A trader who cheated with one member of the guild therefore lost his reputation with every member. The cost of losing trade links with every member of a guild was too great to bear and, as a result, guild members knew that they could trust their counterparties. Conversely, agents who were not members of a guild could not rely upon their counterparty's reputational concerns to ensure that enforcement occurred. Naturally, this gave guild members something of a competitive advantage.

The effectiveness of the professional body is only as good as its ability to communicate with its members. As a result, there is a technological upper bound to the size of a reputational network. This point is illustrated by Greif (1989, 1993, 1994), who studies another medieval example of collective enforcement. Greif examines the Maghribi, Jewish Mediterranean traders operating in the Muslim world at a time when the law courts were poorly developed, operated very slowly, and had very limited enforcement powers. Since individual traders sold very valuable cargos and operated on an infrequent basis, trade could occur only if the reputational costs of stealing from one trader extended to many others. The Maghribi formed a social grouping within which information sharing was feasible: they achieved reputa-tional reach by refusing as a group to trade with a transgressor. Greif provides evidence that this was a sufficiently potent threat to enforce contracts.39

Greif analyzes the problems of sustaining a coalition of this type. Members of the coalition must invest the time and effort required to disseminate trading experiences, and also to learn about transgressions against other members of the group. Furthermore, the costs of sanctions busting must be sufficiently high to discourage it. As a result, Greif notes that the size of the Maghribi trading network was constrained by the information technology available to its members, and he argues that its members had to be sufficiently wealthy to suffer voluntarily produced, voluntarily adjudicated and voluntarily enforced. In fact, it had to be. There was no other potential source of such law, including state coercion.'

39 A large literature shows that societies without formal law fall back upon repeated interaction between individuals who can rely upon bonds of ethnicity or kinsmanship to underscore their dealings, and who can expect through repeat dealings to overcome informational problems. See Landa (1981), Geertz (1978), and Posner (1980).

from the withdrawal of its distribution network. In this context, Greif contrasts the Maghribi with the Genoese traders, who were more atomistic and did not maintain the collective enforcement systems of the Maghribi.

Each of the examples we have provided concerns trade across borders when traders cannot rely upon international law to enforce contracts. In these circumstances, trade relies upon reputational enforcement. Trades will therefore occur only if the counterparts can find some way of placing their reputations at risk; at the same time, traders without reputations will be unable to trade. It follows that traders who have a reputational network have a competitive advantage: if they can, they will attempt to prevent their competitors from building such a network. Professional bodies that are monopoly preservers of corporate trading reputations therefore have a substantial amount of power, which they may attempt to use in order to introduce restrictive trading practices.40 We argue later in the book that these observations can help to explain the nineteenth-century emergence of the American investment bank.

Relationship Between Legal and Extra-Legal Property Rights Institutions

In practice, of course, extra-legal contracting occurs today in a world with well-developed legal institutions. Indeed, the formal law affects informal quasi-legal structures like the ones that we examine in this subsection. One important application of the formal law in informal extra-legal contracting concerns the creation of threat points.41 If the counterparties to a trade anticipate that their interaction will be governed through an informal extra-legal contract, they may nevertheless write a formal contract which incorporates some clearly specified costs of deviation from a standard pro forma trade agreement. In practice, trade will be governed by the extra-legal contract and the formal legal contract will not be invoked: its purpose is not to govern

40 For example, Banner (1998: 266-7) shows that the NYSE Board (the precursor to today's NYSE) started in 1817 to organize brokers into a cartel with respect to their commission levels. See page 135 of chapter five for a more extensive discussion of the NYSE as a reputation-enhancing professional body.

41 A very thoughtful early exposition of a related point is due to Mnookin and Kornhauser (1979: 968-9), who argue that divorce cases are usually negotiated before they reach court. They argue that the role of the formal law is to provide the parties with fallback negotiating positions. However, in their discussion the formal law is not determined, as it is in our discussion, via ex ante contracting between the parties concerned, but reflects external social norms.

the day-to-day terms of trade, but to increase the costs of breaking a relational contract.

For example, a supplier who is unable to deliver goods on time may expect some flexibility from his purchaser in day-to-day trading. However, when the purchaser regards the delay as too much, or as occurring too frequently, she may deem that the terms of the extralegal trade have been broken. In this case the supplier will lose a valuable future stream of business. He will in addition be sued, and will be forced to pay out under the terms of the formal legal agreement. It is important to note that the purchaser will have several opportunities to sue under the terms of this contract: she chooses not to do so in order to sustain the relationship upon which the extra-legal contract rests. The formal contract is intended to strengthen the informal contract, and not in general to govern the terms of trade.

A similar point is made by Bernstein (1996a), who criticizes the United States Uniform Commercial Code's fundamental principle that courts should seek to discover 'immanent business norms', and use them to decide cases. The consequence of this is that a formal contract could be overruled in court because it does not reflect past business practices between the litigants. But this is because past business practices reflected the terms of an informal, and undocumented, contract: the formal contract is intended to provide a basis for court action in the event that the informal contract is breached. The reliance upon informal contracts is evidence that the relationship cannot adequately be governed by formal legal contracts. Anything that undermines the formal contract's effectiveness as an enforcement tool reduces the applicability of relational contracts, and hence undermines trade. Hence, if the court extrapolates from past behavior into a specific interpretation of a contract, it prevents this valuable behavior from emerging in the first place.

Formal legal contracts also serve as an alternative to informal extralegal contracts. If an agent reneges upon a reputational deal, he can always fall back upon a relationship that is intermediated by formal contracts. These contracts rely upon the strength of legal institutions, and not upon reputation. If the contracts are sufficiently close to the relational ones, and the legal institutions are sufficiently strong, reputation therefore ceases to be important. As a result a substantial barrier to entry is removed.42 This has consequences for the structure of an

42 Johnson, McMillan, and Woodruff (2002a , 2002b) note this effect in post-Soviet countries: because they reduce the need for reputational capital in transactions, courts lower switching costs, and hence lower barriers to entry.

industry that previously relied upon traders with access to a large reputational network: see chapters eight and nine for a discussion in the context of investment banking.

When detailed contracting is sufficiently powerful, strong legal institutions thus obviate the need for reputation. To the extent that this leads to heightened competition, it is likely to be a good thing. However, legally enforceable contracts are unlikely to be as nuanced as reputationally intermediated agreements. A shift from the latter to the former may therefore be damaging. This point is highlighted by Baker, Gibbons, and Murphy (1994, 2002), who argue that technological advances that improve formal contracts sufficiently raise the value of relationships which rest upon them. This lowers the cost of reneging upon a reputational contract, and so renders the latter impossible to enforce, even though it generates superior outcomes.43

Was this article helpful?

+1 0
Internet Entrepreneurship Survival Guide

Internet Entrepreneurship Survival Guide

Master The Backwoods of Internet Entrepreneurship All Distilled into a Single Most Powerful Guide! Like a long pole, that can shift a great weight with little effort such is the case with succeeding in business. Your chances of succeeding-as an 'army of one' fall somewhere between zip, zilch and nill.

Get My Free Ebook


Post a comment