The debate on the efficacy of rules versus discretion has received a great deal of attention in the theory of economic policy, in macroeconomics, public finance and in trade policy. But the existing literature has focussed almost exclusively on theoretical aspects of this debate. To date there has been virtually no empirical study of how relevant the distinction between rules and discretion is in the real world. Trade policy lends itself particularly well to an empirical investigation of these issues, for two reasons. First, the theory yields very sharp predictions of how a trade policy chosen under discretion differs from that chosen under rules. Second, and perhaps more important, trade policy in the U.S. is implemented under a variety of institutional arrangements. A major difference among some of these arrangements is the commitment technologies that they provide. Hence, by comparing the policies implemented within these different environments, one can examine whether the capacity to undertake binding policy commitments matters or not. This project represents an empirical exploration into the effects of government discretion on trade policy decision-making. The theoretical literature on time-consistent trade policy suggests that government discretion may be counter productive, and that reliance on a simple set of trade policy rules may be preferred to endowing trade policy makers with a high degree of policy discretion. This research proceeds on the premise that tariff decisions made in the context of escape clause actions (section 201 of the United States Trade Act of 1974) and the determination of exclusions from the tariff reductions negotiated under the Tokyo Round form a natural experiment that can be used to assess the impact of policy discretion on trade policy decisions. The former decisions were made in an environment in which the government possessed a high level of discretion relative to the allocation decisions of those factors of production affected by the government choice. The latter decisions were implemented only after a substantial time lag, and were thus made in a setting in which the government possessed little discretion relative to the relevant factor allocation decisions. The aim of this study is to analyze differences in the government decisions made in each of the two environments, and to ask whether or not systematic differences exist which would suggest that trade policy discretion leads governments to be trapped in sub-optimal time-consistent tariff equilibria. The answer to this question has very important implications for economic policy and for trade policy in particular. A finding that the two environments lead to systematically different policies would imply that policies conducted under rules tend to dominate policies chosen under discretion; as a consequence a institutions providing a commitment technology would be valuable. The opposite conclusions would be reached if instead the two environments led to similar policies or to policies that differed on dimensions other than those predicted by theory.
