Working Paper: CEPR ID: DP15236
Authors: Federico Ciliberto; Ina C. Jakel
Abstract: In many countries, exports are highly concentrated among a few "superstar" firms. We estimate the export decisions of superstar firms as the result of a complete information, simultaneous, discrete choice, static entry game. We employ a dataset on the universe of Danish trade transactions by firm, product and destination. We also obtain detailed information on applied, preferential tariff protection from the MAcMap-HS6 database. We find evidence of strong negative competitive effects of entry: in the absence of strategic competitive effects, firms would be 53.2 percentage points more likely to export to a given market. Next, we run two counterfactual exercises. We show that failing to account for the strategic interaction among superstar exporters leads to: (i) overstating the probability that firms would start exporting to a market following tariff elimination by a factor of two; and, (ii) overstating the probability that firms would stop exporting to a market if tariffs were imposed by a factor of more than five.
Keywords: export participation; strategic interaction; multiple equilibria; trade policy
JEL Codes: F12; F14; L13
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Entry in export markets (F10) | Strong negative competitive effects (D41) |
| Failing to account for strategic interactions (C72) | Overestimations in export probabilities (F14) |
| Tariff elimination (F13) | Probability of starting to export (F10) |
| Tariffs imposed (F14) | Probability of stopping to export (F10) |
| Presence of competitors (L19) | Profitability and export participation (F10) |
| Estimates not accounting for competitive effects (C13) | Overestimate entry response due to trade liberalization (F14) |