Abstract
Using aggregate manufacturing data and a cross-section of 40 countries, we estimate the effects of trade, competition and country size on productivity. Endogeneity of trade and competition is accounted for using instruments that are based on entry barriers and geographical characteristics of the countries. We establish several empirical regularities: (i) Both trade and competition are statistically and economically significant determinants of productivity. (ii) The pro-competitive effect of trade accounts for a quarter of trade's total productivity effects. (iii) Country size appears to play no role, once trade and competition are controlled for. (iv) There is no evidence for a nonlinear relation between competition and productivity or for the hypothesis that larger countries gain less from trade.
Notes
1Since there are no PPP measures at the manufacturing level, we obtain our variable by multiplying aggregate real openness with the country-specific shares of manufacturing trade in total trade. Using aggregate openness instead hardly affects the results.