Exporter Survival with Uncertainty and Experimentation

Two facts distinctively separate exporter dynamics from firm dynamics. One is the strikingly low survival rate of new entrants into export markets. The second is that new entrants survive less than re-entrants. We argue that these two facts are critical to discipline exporter dynamics models because many sources of firm heterogeneity (e.g. fixed costs) do not affect survival rates when firms time their entry decision optimally. We extend a standard exporter dynamics model by positing that firms experiment to resolve an uncertain component in foreign-market profitability. We estimate the model using customs data from Peru. Despite its parsimony, having only four relevant parameters, the model matches the survival profile of entrants and re-entrants. It is also sufficiently rich to deliver predictions about many exporter dynamics facts highlighted in the literature. Finally, we exploit variation across products and markets to provide additional evidence supporting the model’s experimentation mechanism.