Import Price-Elastcities: Reconsidering the Evidence

Recent geography and trade empirical studies based on monopolistic competition [Hummels, 1998; Hanson, 1999; Head and Ries, 1999] suggest high levels of trade price-elasticities (between 3 and 11). However, direct estimations of price-elasticities in trade equations, using price indexes at the aggregate or industry levels, lead to much lower values than those predicted by the prior studies and the theory (usually around unity). In this article, we show that these inconclusive results may be due to an econometric misspecification of the trade equations, to measurement errors in import price indexes as well as endogeneity problems. We re-estimate import price-elasticities from gravity-like equations using methods of transformed least squares and instrumental variables. Our study is based on compatible bilateral trade and activity data from the OECD and INSEE1 for 14 import countries, 16 trading partners, 27 industries and 23 years. When suitable instrumental variables are used, we find relatively high price-elasticities, usually ranging from 1 to 7, the highest estimates corresponding to industries producing homogeneous goods. Our results constitute a first step towards a reconciliation of the theory and the evidence.