How does history matter? Hysteresis in Canadian trade

Should trade economists account for hysteresis when estimating gravity models, as [Eichengreen, B., & Irwin, D. A. (1998). The role of history in bilateral trade flows. In J. A. Frankel (Ed.), Regionalization of the world economy. Chicago: University of Chicago/NBER] recommend? Are the effects on trade of key variables, such as a national border, mismeasured when hysteresis is not accounted for? Using a unique panel data set on Canada's international trade over 1988-1996, and building on recent advances in fixed effects approaches to gravity estimation, we find strong evidence that history does matter in determining trade. But exporter and importer fixed effects, as opposed to lagged dependent variables, appear adequate to capture history. Omitting lagged dependent variables matters little to measured parameters like the border effect. After accounting for history we find Canada's overall border effect to be large with little evidence of diminishment in this period.