In “The Role of Institutions in the Revival of Trade: The Law Merchant, Private Judges and the Champagne Fairs,” Paul Milgrom, Douglass C. North, and Barry R. Weingast take as their starting point a circumstance unusual for legal work: circumstances where effective enforcement does not exist.
Their motivating example is the “law merchant” or lex mercatoria that emerged in Medieval Europe—a specialized uniform set of standards across a wide range of locations, although some research casts doubt on its actual existence as a legal code. The reach of any official sanction could be quite limited: If Antonio cheats Bassanio by providing inferior goods and then skips town, there’s little Bassanio can do. The cheat has slipped outside the bounds of the jurisdiction; Antonio is beyond the reach of the law.
While the scope of modern legal institutions is far broader than towns in 11th century Europe, there are plenty of situations where the law is effectively unenforceable. International law and limited jurisdiction are ready analogues. One of the great challenges in politics is that there is little in the way of enforceable law or contracts. Current members of Congress cannot bind themselves to enact a law next term—at least, not in any judicially enforceable way—and the incentive structures are inherently quite blunt. Likewise, procedural rules like standing, summary judgment, or class certification can effectively close the courthouse doors.
The most common example of legal sanctions with limited reach is ubiquitous: when the formal legal system is simply too costly to utilize.
The fundamental problem facing the merchants is a familiar one: there is incentive to cheat, thus undermining the gains of trade. In short, Antonio and Bassanio face a prisoner’s dilemma. Legal institutions can resolve this, of course, by changing the payoff for cheating.
Cooperation—that is, honest trade—can be sustained even in a prisoner’s dilemma through repeated interactions. This reputational approach, however, has problems scaling up. If Antonio knows he will never see Bassanio again (or if the odds of doing so are low) then the temptation to cheat reemerges.
Milgrom, North, and Weingast build on this well-known iterated prisoner’s dilemma framework. And they show how it can be scaled up—how the same reputation- and information-based behavior can be generated outside of formal enforcement institutions—even if the merchants never meet again.
The problem of scaling up the iterated prisoner’s dilemma is that each merchant only knows their own history. Bassanio knows if Antonio has cheated him, but is ignorant about exchanges with Portia or Lorenzo. The absence of information about past behavior ruins the strategies used to sustain cooperation in an iterated prisoner’s dilemma.
How the Law Merchant Works
To solve this problem, Milgrom, North, and Weingast introduce a new character: the Law Merchant. (While I think “The Law Merchant” is nicely readable, I do find the use of the same term for the specialized legal code and for this character or institution confusing.) The Law Merchant serves as a centralized ledger, a repository of past trades, as well as an adjudicator of disputes.
I don’t want to bury the lede: the key result of the paper is that “efficient trade does not require that every trader know the full history of the behavior of each other trader.” Indeed, the surprising insight is that the amount of information needed in the law-merchant system is rather small. This, in turn, keeps the costs of maintaining this institution modest.
Put another way, Milgrom, North, and Weingast show that the merchants need to be adequately informed, not completely informed. They also recognize that the merchants need to be motivated to provide the necessary information in order to sustain the institution.
Sketching their law-merchant game illustrates how the system works:
- A pair of merchants may query the Law Merchant about their current trading partner’s history. This query is costly. If a trading partner has any unpaid judgments (see below) against them, the Law Merchant reports that information. That a query was solicited is public information.
- The merchants engage in a trade, a prisoner’s dilemma game.
- Either merchant may appeal to the Law Merchant at some cost, but only if they queried the Law Merchant at Step 1.
- If either merchant makes such an appeal, then the Law Merchant issues a judgment if one of the merchants was cheated (i.e., they played “cooperate,” while their trading partner played “defect”).
- Payment of such a legal judgment is, however, entirely voluntary. There is no explicit enforcement mechanism.
- Any unpaid judgments are incorporated into the Law Merchant’s record.
Milgrom, North, and Weingast identify an equilibrium where this law-merchant system is self-sustaining. The equilibrium works over a robust range of parameter values, with the discount factor and the possible gains from cheating being critical ones. In equilibrium, the Law Merchant encourages queries by making them a condition for potential appeals. In equilibrium, merchants who do not query are constantly cheated by their partners.
One thing I like about the paper is how the authors pay attention to costs at every stage. Gathering and warehousing information is not costless. The equilibrium is also robust to some degree of corruption by the Law Merchant itself.
The law-merchant system solves an informational problem. What is noteworthy, and why I think this paper is valuable to consider today, is that it explicitly considers the transaction costs entailed in doing so. The system “not only centralizes this information but provides it in very simple form: all that needs to be communicated is whether there are any outstanding judgments.” Information, and especially attention, is not free. Sometimes just a little bit of knowledge is all we need. And maybe all we want.
Another feature of the law-merchant system is the ability to pay judgments to get out from underneath the sanction, even if it is never used in the equilibrium presented in the paper. It could be worth considering how that might be implemented more generally. That is, when would we want to allow people to operate with a “clean slate”? What would they have to do, and how could that be implemented?
The game is obviously artificial. They all are. But it reveals something useful and insightful.
The Modern Law Merchant?
A range of new technologies with decentralized ledgers recording transactions have proliferated. While game theory has apparently greatly influenced the thinking behind cryptocurrencies and the like, I have not found much work discussing the relationship between blockchain and, specifically, iterated prisoner’s dilemmas.
Blockchain (and similar technologies) is a ledger, a record of transactions. With Bitcoin, one can observe the entire history of transactions, which creates a huge, decentralized record. Faking such a record is a lot of work, which is seen as its primary benefit. But it also has the potential to scale up the iterated prisoner’s dilemma: Lorenzo may know nothing about Portia, but if he can pull up her entire economic history with ease, then he doesn’t have to. Huge amounts of information can, unsurprisingly, solve an informational problem.
A key shift here, though, is that things are less anonymous. In this iterated prisoner’s dilemma environment, Portia and Lorenzo are less interested in the exchange history of a particular good or token—i.e., the object of a potential trade—but in their respective histories of exchange. Aliases could be interposed, and they could develop reputations of their own. We see this already with both brands and with rankings on platforms like eBay.
A wealth of information is not bad, per se, although it can contort decisionmaking and have unintended consequences. But the law-merchant system illustrates that it is wildly unnecessary. It’s overkill. In many circumstances, we do not need an entire history of transactions to sustain cooperation or the voluntary punishment of cheaters.
The law-merchant system does, however, entail a degree of centralization. This is useful if there may be disputes as to what counts as “cheating,” but it may make it a poor fit for the stated goals of some applications of blockchain technology.
Recovering the Shadow of the Law
When studying law, we focus on disputes. We read cases: These are the units of analysis in U.S. law and where common law develops. Easy cases do not go to trial; they settle. That, of course, does not mean that the law is having no effect. Indeed, quite the opposite: The effect is so consequential that nobody needs to go through the rigamarole of testing it.
All sorts of behavior can be similarly shaped by the law in ways that enforcement is never actually carried out. Crimes deterred need not be prosecuted, contracts honored need not find their way inside the courthouse. Much human activity, especially economic activity, takes place in the shadow of the law (to borrow from another canonical law & economics piece).
All that presumes, however, that enforcement is available. If the law is too costly, cumbersome, or inconvenient to use, then its shadow shrinks. Litigation costs can thus mute the influence of the law. If a lawsuit is too costly to mount, then even if Bassanio can get ahold of Antonio, the law will do little to influence anyone’s behavior. Litigation costs can bring us back to the 11th century. In a world where litigation is costly to mount and other efficiency-saving tools are sharply limited, then law’s shadow contracts.
Is there perhaps an opening for a law-merchant-style institution to fill this gap? The law-merchant system has low transaction costs, presents information in an easily digestible way, and lacks formal enforcement. Arbitration could be of use here, but it has only some of those features and tends to be used in specialized contexts. But there might be some possibility of a more broad-ranging, informational story based on a reputation for fair dealing inspired by the Law Merchant.
Further Reading
- Lisa Bernstein, “Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions,” John M. Olin Program in Law and Economics Working Paper No. 133 (2001)
- Douglas Hill, “Errors of Judgment and Reporting in a Law Merchant System,” Theory and Decision, Vol. 56 (May 2004)
- Catherine Hafer, “On the Origins of Property Rights: Conflict and Production in the State of Nature,” The Review of Economic Studies, Vol. 73 (2006)
- Gillian Hadfield & Barry Weingast, “What is a Law – A Coordination Model of the Characteristics of Legal Order,” The Journal of Legal Analysis, Vol. 4 (July 2012)
