Harald Oberhofer, WU Vienna: "Everything You Always Wanted to Know About EU Membership Trade Effects but Were Afraid to Ask"

This paper investigates the impact of the selection of particular trade data for the estimated EU membership trade effects in gravity models. We obtain 543 estimates for EU membership trade effects from a standard specification of the gravity model using eleven different data sources and sample periods. These are used to perform a meta-regression analysis aiming to explore the sources for heterogeneity in the obtained estimates. The findings from the meta study highlight the crucial role of domestic trade flows for the magnitude of the EU membership trade effect. The estimated EU trade effect is around 25.3% larger in datasets that also account for trade diversion from domestic trade. Furthermore, the estimated EU membership trade effects also vary with the time period, country coverage, the use of interval data, the use of various HS revisions, and the level of disaggregation available in the data sources. Alternative estimation packages and the replacement of exports by import data as the outcome of interest do not significantly affect the estimated magnitudes of the EU membership trade effects.