GEP 19/02: Gravity

Summary

In less than two decades, the field of international trade has seen the emergence of a vast literature on the correct estimation of theory-grounded gravity models of bilateral trade, and, by extension, on the real effects of international agreements on trade. A consensus has now emerged that regional trade agreements have a sizeable impact on the flow of goods. The standard approach in this literature however still models the trade gravity equation as the common relationship across trading pairs and hence provides a single estimate for the trade effect of a regional trade agreement across all country-pairs and time periods. This estimate may however not be very informative for individual policymakers as it likely represents an average of potentially very different impacts across country-pairs.

In this Nottingham School of Economics working paper Rodolphe Desbordes and Markus Eberhardt investigate this issue using quarterly data from 20 advanced economies over the past 55 years. They adopt the preferred empirical implementation in this literature (a Poisson pseudo maximum likelihood model) but estimate the gravity relationship at the pair-level using new insights from the common factor model literature to capture unobserved trade network and globalisation effects. The average effect of entering regional trade agreements is found to be an increase in trade of just over 20%. However, this average hides substantial heterogeneity across countries, with direct consequence for heterogeneity across time periods. Beyond these findings, the empirical approach introduced by the authors offers solutions to issues currently discussed in frontier gravity model research.

GEP discussion paper 19/02: Grav_ity, by Rodolphe Desbordes and Markus Eberhardt.

Download the paper in PDF format

Authors

Rodolphe Desbordes and Markus Eberhardt

View all GEP discussion papers | View all School of Economics featured discussion papers