To Adopt or Not to Adopt: Heterogeneous Trade Effects of the Euro

Two decades of research on the euro’s trade effects have produced estimates ranging from 4% to 30%, with no consensus on the magnitude. We find evidence that this divergence may reflect genuine heterogeneity in the euro’s trade effect across country pairs rather than methodological differences alone. Using Eurostat data on 15 EU countries (12 eurozone members plus Denmark, Sweden, and the UK as controls) from 1995-2015, we estimate that euro adoption increased bilateral trade by 29% on average (14.1% after fixed effects correction), but effects range from -12% to +79% across eurozone pairs. Core eurozone pairs (e.g., Germany-France, Germany-Netherlands) show large gains, while peripheral pairs involving Finland, Greece, and Portugal saw smaller or negative effects, with some negative estimates statistically significant and interpretable as trade diversion. Pre-euro trade intensity and GDP account for over 90% of feature importance in explaining this heterogeneity. Extending to EU28, we find evidence that crisis-era adopters (Slovakia, Estonia, Latvia) pull down naive estimates to 4.3%, but accounting for fixed effects recovers estimates of 13.4%, consistent with the EU15 fixed-effects baseline of 14.1%. Illustrative counterfactual analysis suggests non-eurozone members would have experienced varied effects: UK (+33%), Sweden (+22%), Denmark (+19%). The wide range of prior estimates appears to be largely a feature of the data, not a bug in the methods.

💡 Research Summary

This paper tackles the long‑standing puzzle of why empirical estimates of the euro’s impact on intra‑European trade vary so widely, ranging from 4 % to 30 % in the literature. The authors argue that the dispersion is not merely a methodological artifact but reflects genuine heterogeneity in treatment effects across country pairs. Using Eurostat bilateral trade data for 15 EU economies (the 12 euro‑zone members plus Denmark, Sweden and the United Kingdom) covering 1995‑2015, they apply a state‑of‑the‑art causal inference framework: causal forests combined with double‑machine‑learning (DML).

First, they estimate a conventional gravity model with pair‑fixed effects, obtaining a “raw” average treatment effect (ATE) of roughly +29 % in trade volumes after euro adoption. After controlling for country‑time and pair‑fixed effects, the average effect falls to 14.1 %. The causal‑forest approach then uncovers the full distribution of conditional average treatment effects (CATEs) at the country‑pair level. Effects span from –12 % to +79 %, indicating substantial variation. Core euro‑zone pairs such as Germany‑France and Germany‑Netherlands experience the largest gains, while peripheral pairs involving Finland, Greece or Portugal show modest or even negative impacts—some of which are statistically significant and can be interpreted as trade diversion.

Variable‑importance analysis reveals that pre‑euro trade intensity and the GDP of the two partners together explain more than 90 % of the heterogeneity. In other words, the more intensive the pre‑existing trade relationship and the larger the economies, the larger the trade‑boosting effect of a common currency.

The authors extend the analysis to the full EU‑28 sample. They find that the “crisis‑era” adopters (Slovakia, Estonia, Latvia) pull down naïve average estimates to about 4.3 %. However, once pair‑fixed effects are introduced, the estimate rises to 13.4 %, aligning closely with the 14.1 % benchmark from the EU‑15 sample. This demonstrates that the low naïve estimates are driven by a subset of late‑adopting, smaller economies rather than a universal lack of effect.

A counterfactual exercise is performed for the three non‑eurozone members. The model predicts that, had they adopted the euro, the United Kingdom would have seen a 33 % increase in bilateral trade, Sweden a 22 % increase, and Denmark a 19 % increase. These figures illustrate that the trade‑gain argument for euro adoption is highly contingent on a country’s existing trade network and economic size.

Methodologically, the paper contrasts three dominant approaches: (1) gravity models that estimate a single ATE, (2) synthetic‑control methods that provide precise estimates for a few high‑trade pairs but cannot scale, and (3) the causal‑forest/DML pipeline that delivers a full CATE distribution while allowing for valid inference through honest splitting and bootstrap aggregation. By doing so, the study reconciles why gravity‑based studies cluster around the lower end of the range (≈20 %) while synthetic‑control studies report higher effects (≈30 %+).

The contribution is twofold. First, it offers a comprehensive empirical picture of how the euro’s trade impact varies across the entire network of euro‑zone relationships, thereby resolving a key inconsistency in the existing literature. Second, it demonstrates the practical value of modern causal‑machine‑learning tools for policy‑relevant macro‑economic questions, showing that average effects can be misleading when policy decisions hinge on the distribution of outcomes.

In sum, the paper concludes that the euro’s trade‑enhancing effect is real but highly heterogeneous. Policymakers contemplating euro adoption should therefore assess not only the average expected gain but also the specific characteristics of their bilateral trade ties—especially pre‑existing trade intensity and economic size—to gauge whether the benefits will materialize for their country.