Estimating the landlocked penalty for international trade

This paper studies the effect of being landlocked on international trade flows in the manufacturing sector. The study overcomes the empirical challenge of using a country-specific variable ‘landlocked’, within a structural gravity framework in the presence of importer and exporter fixed effects. Results show that the impact of being landlocked for manufactured goods trade is highly negative and statistically significant. Such effect is higher for low-income landlocked countries compared to high-income landlocked countries. The impact of being landlocked on manufactured goods trade declined in 2005 compared to 1980. A series of robustness checks further confirm the estimates.