will cease trading following such an abuse. Conditioning future transactions between the ruler and many merchants on his conduct toward a particular merchant may be able to surmount the commitment problem without the aid of any formal organization. In Sicily, as in the other examples cited merchants imposed collective punishment on the city that included participation by merchants who had not been directly injured. The offenses reflected in these cases were often against an entire group of merchants. But rulers could also discriminate among merchants, abusing or not protecting them selectively, by confiscating the belongings of or withholding legal protection from some merchants without directly harming other merchants. Indeed, the Sicilian ruler increased the tariff only on Jewish traders; and in Constantinople during two attacks on the Genoese quarter, other Italian merchants were not harmed.
These examples suggest two interconnected reasons why, without a supporting organization, a multilateral reputation mechanism may be insufficient to surmount the commitment problem at the efficient level of trade. The first involves contractual ambiguities and asymmetric information. The second reflects the distinct incentives among different merchants generated by a multilateral response.
Long-distance premodern trade took place in a highly complex and uncertain environment. Unanticipated events and multiple interpretations of existing agreements were always possible under these circumstances, implying that the definition of a “contract violation” was often ambiguous. Different interpretations of facts by merchants, information asymmetry, and slow communication implied that without an organization that coordinated responses, merchants as a whole were not likely to respond effectively to the abuse of any group of merchants. Section 4.2 demonstrates formally that if the fraction of merchants who detect and react to an abuse against any group of merchants is only proportionate to the number abused, then a multilateral reputation mechanism is ineffective at the efficient volume of trade for the same reason that a bilateral reputation mechanism is ineffective: a threat by a group of marginal traders to withdraw their trade is barely significant once trade has expanded to its efficient level.
Expanding trade to the efficient level in the medieval environment required an organization that supplemented the operation of a multilateral reputation mechanism by coordinating the responses of a large fraction of the merchants. Only when a coordinating