DP15633 Borders within Europe
|Author(s):||Marta Santamaría, Jaume Ventura, Ugur Yesilbayraktar|
|Publication Date:||January 2021|
|Keyword(s):||Border effect, European integration, Regional Trade|
|JEL(s):||D71, F15, F55, H77, O57|
|Programme Areas:||Public Economics, International Trade and Regional Economics, Economic History, International Macroeconomics and Finance, Macroeconomics and Growth|
|Link to this Page:||cepr.org/active/publications/discussion_papers/dp.php?dpno=15633|
Are country borders still an impediment to trade flows within Europe? Using a rich microlevel survey with 3 million annual shipments of goods by road across 269 European regions, we construct a matrix of bilateral trade flows for 12 industries from 2011 to 2017. We then use the causal inference framework to design an dentification strategy to estimate the causal effect of country borders on trade flows. Take two similar region pairs, the first one containing regions in different countries and the second one containing regions in the same country. The market share of the origin region in the destination region for the international pair is only 17.5 percent that of the intranational pair. We refer to this estimate as the average border effect. When we look at each industry separately, we find border effects that range from 12.3 to 38.9 percent. When we look at recent borders, i.e. created after 1910, we find a border effect of 28.8 percent, which is smaller than the average border effect but still quite large. The implication is clear: Europe is far from having a single market.