Disaggregated import-demand elasticity estimates based on import unit-value indexes
are used in virtually all trade policy simulation models. However, unit-value indexes
have been criticized especially by Kravis and Lipsey (1974). To examine the effect
of using unit-value indexes on estimates of disaggregated import-demand elasticities,
this paper compares regression results using unit-value indexes with results using
U.S. Bureau of Labor Statistics import-price indexes for several detailed trade categories
based on quarterly data for 1978-88. Results show that using unit-value indexes does
not greatly affect estimated import-demand elasticities.