Data and Code for: The Intensive Margin in Trade: How Big and How Important?

In the benchmark trade models with monopolistic competition featuring a constant trade elasticity, variation in bilateral trade flows happens either entirely on the intensive margin of exports per exporting firm (Krugman) or entirely on the extensive margin of the number of exporting firms (Melitz-Pareto). Using the World Bank's Exporter Dynamics Database featuring firm-level exports from 50 countries, we find that around 50% of variation in exports is along each margin, implying that the trade elasticity may not be constant, and gains from trade may differ from those in benchmark models. We show that moving from a Pareto to a lognormal distribution gives a positive role for both margins, and we use likelihood methods to estimate a generalized Melitz model with a joint lognormal distribution for firm productivity, fixed costs, and demand shifters. Using "exact hat algebra'' we quantify how trade costs affect trade flows and welfare in the estimated model. We find similar welfare effects to those in the Melitz-Pareto model but significant differences in the implied trade flows.