Polarity and Free Trade respect to the opening of markets between Table 2.The Tariff Game between Allies or among states. Column Intraalliance Exploitation:Incentives and Constraints Row Cooperate Defect Cooperate 1,3 3,1 The presence of both great and small Defect 2,4 4,2 powers within an alliance raises the issue of exploitation in sharp relief.Assuming Note:Payoffs are ranked from 1(best)to 4(worst); that "great"and "small"are good proxies row's payoffs are listed first. for market power,6 the tariff game be- tween allies is an asymmetric one:a large state can,and a small state cannot,use tariffs to improve its terms of trade.In for example,it has some private informa- this game,the dominant strategy of the tion about the elasticity of global large state is to defect to an optimum demand-and-supply curves and if small tariff;the dominant strategy of the small countries organizing to exert countervail- state is to cooperate or adhere to free ing power in world markets incur some trade (Conybeare 1987,35).Table 2 dis- costs in doing so.That a hegemon might plays the payoffs to the small (row)and to indeed have private information about the large(column)power.The logical out- global markets follows logically from its come of the game is CD:the small state incentives to become informed about abjures and the large state erects tariff them;a small country,in contrast,has lit- barriers.Under what conditions,then, tle incentive to acquire such information, will free trade occur within an alliance? because it assumes that it cannot affect its What will persuade the intraalliance hege- terms of trade.That significant transac- mon to forgo an optimum tariff? tion costs are incurred in the process of forming customs unions follows from the Incentives to exploit.Critics of hegemonic distributional effects (both within and stability theory argue that there are strong across the custom union's potential mem- incentives for any hegemon to use an opti- bers)evoked by the setting of uniform ex- mum tariff (Conybeare 1984,1987).On ternal trade barriers (McMillan 1986,67). economic grounds alone,however,a non- The trading practices of both Britain myopic,rational hegemon may reject and the United States suggest that the such a tariff for the same reason that a analogy to limit pricing is of more than domestic monopolist sets prices below just analytic interest.Midnineteenth- their short-run maximizing levels:to deter century Britain,according to one observ- competition.?By "limit"pricing,the er,maintained its tariffs at less than opti- monopolist seeks to convince potential mum levels in order to fix its "monopoly entrants that its costs of production are of manufactures on the rest of the world lower than they are in reality (Milgrom for a few more decades than its natural and Roberts 1982).The incumbent firm's term"(William Cunningham,cited in Mc- ability to sacrifice short-run gains in order Closkey 1980,304).The logic of limit to earn higher,long-run returns depends pricing apparently became clear to the on the existence of costs to entry as well as United States when it attempted,in the on asymmetric information about its costs 1930 Smoot-Hawley tariff,to turn the of production.8 terms of trade in its favor.This effort, A rational,nonmyopic hegemon may among other factors,provoked the con- set its tariff at less than the short-run opti- struction of trading blocs abroad and ap- mum level under analogous conditions:if, parently induced the United States to try 1251Polarity and Free Trade respect to the opening of markets between or among states. Intraalliance Exploitation: Incentives and Constraints The presence of both great and small powers within an alliance raises the issue of exploitation in sharp relief. Assuming that "great" and "small" are good proxies for market power,6 the tariff game be￾tween allies is an asymmetric one: a large state can, and a small state cannot, use tariffs to improve its terms of trade. In this game, the dominant strategy of the large state is to defect to an optimum tariff; the dominant strategy of the small state is to cooperate or adhere to free trade (Conybeare 1987, 35). Table 2 dis￾plays the payoffs to the small (row) and to the large (column) power. The logical out￾come of the game is CD: the small state abjures and the large state erects tariff barriers. Under what conditions, then, will free trade occur within an alliance? What will persuade the intraalliance hege￾mon to forgo an optimum tariff? Incentives to exploit. Gitics of hegemonic stability theory argue that there are strong incentives for any hegemon to use an opti￾mum tariff (Conybeare 1984, 1987). On economic grounds alone, however, a non￾myopic, rational hegemon may reject such a tariff for the same reason that a domestic monopolist sets prices below their short-run maximizing levels: to deter competition.' By "limit" pricing, the monopolist seeks to convince potential entrants that its costs of production are lower than they are in reality (Milgrom and Roberts 1982). The incumbent firm's ability to sacrifice short-run gains in order to earn higher, long-run returns depends on the existence of costs to entry as well as on asymmetric information about its costs of production.8 A rational, nonmyopic hegemon may set its tariff at less than the short-run opti￾mum level under analogous conditions: if, Table 2. The Tariff Game between Allies Column Row Cooperate Defect Cooperate 1, 3 3, 1 Defect 2, 4 4, 2 Note: Payoffs are ranked from 1 (best) to 4 (worst); row's payoffs are listed first. for example, it has some private informa￾tion about the elasticity of global demand-and-supply curves and if small countries organizing to exert countervail￾ing power in world markets incur some costs in doing so.9 That a hegemon might indeed have private information about global markets follows logically from its incentives to become informed about them; a small country, in contrast, has lit￾tle incentive to acquire such information, because it assumes that it cannot affect its terms of trade. That significant transac￾tion costs are incurred in the process of forming customs unions follows from the distributional effects (both within and across the custom union's potential mem￾bers) evoked by the setting of uniform ex￾ternal trade barriers (McMillan 1986, 67). The trading practices of both Britain and the United States suggest that the analogy to limit pricing is of more than just analytic interest. Midnineteenth￾century Britain, according to one observ￾er, maintained its tariffs at less than opti￾mum levels in order to fix its "monopoly of manufactures on the rest of the world for a few more decades than its natural term" (William Cunningham, cited in Mc￾Closkey 1980, 304). The logic of limit pricing apparently became clear to the United States when it attempted, in the 1930 Smoot-Hawley tariff, to turn the terms of trade in its favor. This effort, among other factors, provoked the con￾struction of tra- blocs abroad and ap￾parently induced the United States to try