Would the right tariff aggregator for policy analysis please stand up?

Abstract

The conventional use of import-weighted averages in tariff aggregation has little theoretical basis and wastes valuable information typically available in the detailed structure of protection. We build on the Anderson–Neary Trade Restrictiveness Index (TRI) and define ways in which a detailed set of tariffs may be aggregated consistently to provide measures of the impact of tariffs at the sectoral level. Under the assumptions needed for aggregation, we provide distinct aggregators for expenditure, input costs, tariff revenues, and the overall restrictiveness of tariffs in particular commodity groups. We illustrate the nature of the differences between the measures and show how they may be used in applied general equilibrium models to avoid wasting the valuable information available below the high level of aggregation necessarily used in such models.