Learning by exporting in Turkey: an investigation for existence and channels

Using a rich longitudinal database at the plant level, I shed new light on the causal nexus between exports and productivity for Turkey, a middle-income country. I find evidence for both self-selection into exporting and learning-by-exporting. My main focus is on post-entry effects. To test this hypothesis I follow recent empirical literature and I apply the Propensity Score Matching and a Difference-in-Difference estimator. I find a higher labour productivity and TFP growth for exporting firms in the entry year and some years following the entry. Exports seem to place firms on a superior productivity path. My main contribution is to show the strict linkage between export and import activity: export starters often start also importing. Learning by exporting effects hold when I control for the role of imports and I verify larger productivity gains for firms which start exporting and importing at the same time. Finally, in order to verify if post-entry effects are not only scale eects but work through competition channel and/or technology transfers, I look for a heterogeneity according to the sectoral productivity gap between the domestic market and foreign trade partners. I verify a different timing of efficiency improvements between comparative advantage and disadvantage sectors.