Understanding interstate trade patterns

Abstract

This paper models and estimates bilateral trade patterns of U.S. states in a CES framework and identi…es the elasticity of substitution across goods, elasticity of substitution across varieties of each good, and the good-speci…c elasticity of distance measures by using markup values obtained from the production side. Compared to empirical international trade literature, the elasticity of substitution estimates are lower across both goods and varieties, while the elasticity of distance estimates are higher. Although home-bias e¤ects at the state level are signi…cant, there is evidence for decreasing e¤ects over time.JEL Classi…cation: F12, R12, R32

Border Effects Without Borders: What Divides Japan's Internal Trade?

“…Table in the Appendix presents the results for 2000, 2005, and 2010. Depending on the sector, varies from 2.03 for “manufacturing” in 2010 to 4.79 for “miscellaneous products” in 2005, which is in line with the findings of Yilmazkuday (), who computes elasticities of substitution for trade within the United States that range from 1.61 to 5.99, with an average value of 3.01. Pooling over all sectors implies an average trade cost elasticity of about , which is a somewhat smaller value than the mean or the preferred estimate of 3.19 or 4.51 that Head and Mayer () report in their meta study .…”

Does distance reflect more than transport costs?

“…D P k denotes productspecific dummies and D C h denotes city-specific dummies which are to capture any idiosyncratic aspects of the price of a given city. 6 Notice that the regressor ofε ij in Regression 2 is a residual and thus it is susceptible to the so-called generated regressor problem that OLS-based standard errors are invalid (e.g., Yilmazkuday, 2012). As noted by Pagan (1984, p. 242), however, standard inference is still valid if unlagged residuals are used in a regression as in our case.…”

Price dispersion across US districts of entry

“…9 One of the highest contribution of trade costs is for plate of iron (with SITC code of 6741) where the contribution of freight-related costs is more than 10%, mostly due to the heavy structure of the product. 6 In this second methodology, we decompose the summation of the right hand side of the expression into its components. 7 All estimated g 's are signi…cant at the 10% level.…”

“…Since this …rst methodology has an implicit assumption that the right hand side variables are independent from each other, we also consider a second variance decomposition methodology according to the following approximation: where we have ignored the covariance terms to focus on the pure e¤ects of right hand side variables on price dispersion. 6 …”