748 JOURNAL OF POLITICAL ECONOMY tion mechanism,in which a merchant whose rights were abused ceased trading,or a multilateral reputation mechanism,in which the cheated merchant and his close associates ceased trading.Yet the historical records indicate that,by and large,the ruler-merchant relations were governed by neither bilateral nor informal multilateral arrangements. On the contrary,ruler-merchant relations were governed by adminis- trative bodies rooted outside the territory of the ruler,which held certain regulatory powers over their member merchants in their own territory and supervised the operation of these merchants in foreign lands.What roles could these administrative bodies theoretically play in overcoming the ruler's commitment problem?What roles did they play in fact? To investigate these questions,we utilize historical records to de- velop a series of game-theoretic models corresponding to different institutional arrangements.The theoretical analyses indicate that al- though some trade is possible even without supporting organizations, sustaining the efficient level of trade is more demanding.Without administrative bodies capable of coordinating and sometimes compel- ling merchants'responses to a ruler's transgressions,trade could not expand to its efficient level.The corresponding historical analysis then suggests that during the late medieval commercial revolution, a specific institution-the merchant guild-developed the necessary attributes to enforce agreements with rulers,thus overcoming the commitment problem and enabling trade expansion.Merchant guilds exhibited a range of administrative forms from subdivision of a city administration to an intercity organization.Yet these forms all shared the common function of ensuring the coordination and internal en- forcement required to surmount the commitment problem by permit- ting effective collective action.We emphasize two points at the outset. First,our argument concerns merchant guilds and not craft guilds.3 Second,we define merchant guilds according to their function rather than their "official,"late medieval name.Hence,as we discuss below, our theory applies to a wider range of medieval merchant organiza- tions than those labeled as merchant guilds. The evaluation of merchant guilds as supporting efficient trade is complementary to the view more common among economic histori- ans that merchant guilds emerged to reduce negotiation costs,to administer trade and taxation,to extract privileges from foreign cities,and to shift rent in their own city (see,e.g.,Gross 1890;Thrupp 1965;North and Thomas 1973).While the existence of merchant s Economists have long associated the latter with the monopolization of a given craft within a specific town.For a recent economic analysis of craft guilds,see Hickson and Thompson (1991).See also Gustafsson (1987).748 JOURNAL OF POLITICAL ECONOMY tion mechanism, in which a merchant whose rights were abused ceased trading, or a multilateral reputation mechanism, in which the cheated merchant and his close associates ceased trading. Yet the historical records indicate that, by and large, the ruler-merchant relations were governed by neither bilateral nor informal multilateral arrangements. On the contrary, ruler-merchant relations were governed by adminis￾trative bodies rooted outside the territory of the ruler, which held certain regulatory powers over their member merchants in their own territory and supervised the operation of these merchants in foreign lands. What roles could these administrative bodies theoretically play in overcoming the ruler's commitment problem? What roles did they play in fact? To investigate these questions, we utilize historical records to de￾velop a series of game-theoretic models corresponding to different institutional arrangements. The theoretical analyses indicate that al￾though some trade is possible even without supporting organizations, sustaining the efficient level of trade is more demanding. Without administrative bodies capable of coordinating and sometimes compel￾ling merchants' responses to a ruler's transgressions, trade could not expand to its efficient level. The corresponding historical analysis then suggests that during the late medieval commercial revolution, a specific institution-the merchant guild-developed the necessary attributes to enforce agreements with rulers, thus overcoming the commitment problem and enabling trade expansion. Merchant guilds exhibited a range of administrative forms from subdivision of a city administration to an intercity organization. Yet these forms all shared the common function of ensuring the coordination and internal en￾forcement required to surmount the commitment problem by permit￾ting effective collective action. We emphasize two points at the outset. First, our argument concerns merchant guilds and not craft guilds.3 Second, we define merchant guilds according to their function rather than their "official," late medieval name. Hence, as we discuss below, our theory applies to a wider range of medieval merchant organiza￾tions than those labeled as merchant guilds. The evaluation of merchant guilds as supporting efficient trade is complementary to the view more common among economic histori￾ans that merchant guilds emerged to reduce negotiation costs, to administer trade and taxation, to extract privileges from foreign cities, and to shift rent in their own city (see, e.g., Gross 1890; Thrupp 1965; North and Thomas 1973). While the existence of merchant 3 Economists have long associated the latter with the monopolization of a given craft within a specific town. For a recent economic analysis of craft guilds, see Hickson and Thompson (1991). See also Gustafsson (1987)