Search for a command to run...
What are the barriers that separate nations? While recent work provides intriguing clues, we have remarkably little concrete evidence as to the nature, size, and shape of barriers. This paper offers direct and indirect evidence on trade barriers, moving us toward a comprehensive geography of trade costs. There are three main contributions. One, we provide detailed data on freight rates for a number of importers. Rates vary substantially over exporters, and aggregate expenditures on freight are at the low end of the observed range. This suggests import choices are made so as to minimize transportation costs. Two, we estimate the technological relationship between freight rates and distance and use this to interpret the trade barriers equivalents of common trade barrier proxies taken from the literature. The calculation reveals implausibly large barriers. Three, we use a multi-sector model of trade to isolate channels through which trade barriers affect trade volumes. The model motivates an estimation technique that delivers direct estimates of substitution elasticities. This allows a complete characterization of the trade costs implied by trade flows and a partition of those costs into three components: explicitly measured costs (tariffs and freight), costs associated with common proxy variables, and costs that are implied but unmeasured. Acknowledgments: Thanks for the gracious provision of data go to Jon Haveman, Rob Feenstra, Azita Amjadi and the ALADI secretariat. Thanks for helpful suggestions on previous drafts go to seminar participants at the Universities of Chicago, Michigan, and Texas, Boston University, NBER and the 4th Annual EIIT Conference at Purdue University. Finally, Julia Grebelsky and Dawn Conner provided outstanding research assistance. This research was funded by a grant from the University of Chicago’s Graduate School of Business.