doi.org/10.71734/RB-84-en

Economic sanctions have become an integral instrument of foreign and security policy. Since 1950, there has been a significant increase in ongoing sanction regimes as well as newly imposed measures (Syropoulos et al., 2024). During this period, almost every country has been affected by sanctions or involved in imposing them at least once. At the same time, there has been growing interest in reliably determining their economic impact. Many studies treat sanctions as largely homogeneous, although they differ significantly in intensity, duration and in their evolution over time. In addition, the lifting of sanctions is often not captured separately, even though trade can develop very differently thereafter. In a new study (Kwon, Nagengast, Yoon and Yotov, 2026), we therefore use modern econometric methods that explicitly take these differences into account both while sanctions are in force and after sanctions are lifted.

In the new study, we combine a modern difference-in-differences approach with a gravity model of international trade (Yotov et al., 2016; Nagengast and Yotov, 2025; Wooldridge, 2025). We thus capture the effects of imposing and lifting sanctions separately. At the same time, the approach explicitly takes into account the fact that sanctions may differ significantly between countries and over time. This allows us to avoid the simplifying assumption of previous studies that sanctions have a broadly uniform impact. To this end, we link data from the Global Sanctions Database (Syropoulos et al., 2024) with bilateral trade data from 260 countries for the period from 1950 to 2019. The focus is on complete trade sanctions that were imposed and subsequently lifted. In total, we consider 795 sanctioned country pairs. These cases are spread across 37 groups, each with identical imposition and lifting years.

Our results indicate that complete trade sanctions dampen bilateral trade significantly more than previous studies suggest. As long as sanctions are in force, trade flows decline by just under 60 % on average. If conventional methods are applied to the same data, however, they imply a decline of only just under 40 %. One key reason for this deviation is that these methods do not adequately account for differences between individual sanction cases. As a result, cases with relatively small trade effects tend to be given too much weight in the estimation. In addition, conventional estimates sometimes mix the effect during the sanction phase with developments from the years after the lifting of sanctions (de Chaisemartin and D’Haultfœuille, 2023; Goldsmith-Pinkham et al., 2024). Our results suggest that this mixing alone explains around one-third of the deviation from previous estimates.

We also examine how trade relationships develop after sanctions are lifted. On average, our estimates show no statistically significant trade effect after sanctions are lifted. This contrasts with conventional estimates based on the same data, which point to positive effects after lifting. However, this average value masks important differences between the individual sanction cases. For sanctions lasting up to six years, trade returns to pre-sanction levels and, in some cases, rises beyond them once they have been lifted; these results are qualitatively in line with those obtained using conventional approaches. In the case of longer-lasting sanctions, by contrast, the effects tend to remain negative even after they have been lifted. 

The study shows that the trade effects of sanctions are systematically related to the characteristics of the respective country pairs and sanction cases. Trade losses are particularly pronounced among geographically distant partners. One possible explanation is that long-distance trade is more time- and cost-intensive, making it more vulnerable to disruption and making the associated business relationships harder to maintain. Trade losses are also above average for countries with a common border. Additional compliance and monitoring frictions at border crossings could play a role here. Institutional ties can counteract these effects. An existing regional trade agreement dampens losses during the sanctions phase. If both countries are members of the WTO, trade tends to recover more quickly after sanctions are lifted. However, sanctions aimed at preventing or ending a war or influencing territorial conflicts have particularly strong and, in some cases, persistent trade effects. Earlier non-trade sanctions between the same countries are also associated with stronger trade effects during the sanction phase. In addition, sanctions that ultimately fail to achieve their declared objective show persistent negative trade effects, especially after lifting.

The results underscore the extent to which economic sanctions differ in their intensity and impact. A comprehensive assessment of sanctions should not only take into account their direct trade effect, but also how long sanctions remain in place, the institutional setting in which they operate, and how trade develops after they are lifted. For instance, if both countries are members of the WTO, this can support the recovery after the lifting of sanctions. At the same time, the study shows that, on average, more precise estimation methods imply stronger sanction effects than previous studies suggest and can make differences between individual sanction cases more evident. This could also be relevant for analysing the effects of other bilateral policy measures that are introduced and subsequently lifted and that impact cross-border flows such as migration and direct investment.

de Chaisemartin, C. and X. D’Haultfœuille (2023), Two-way fixed effects and differences-in-differences estimators with several treatments, Journal of Econometrics, Vol. 236, No 2, 105480.

Goldsmith-Pinkham, P., P. Hull and M. Kolesár (2024), Contamination Bias in Linear Regressions, American Economic Review, Vol. 114, No 12, pp. 4015‑4051.

Kwon, O., A. J. Nagengast, J. Yoon and Y. V. Yotov (2026), From imposition to lifting: Estimating the effects of sanctions over their lifecycle, Deutsche Bundesbank Discussion Paper No 06/2026.

Nagengast, A. J. and Y. V. Yotov (2025), Staggered Difference-in-Differences in Gravity Settings: Revisiting the Effects of Trade Agreements, American Economic Journal: Applied Economics, Vol. 17, No 1, pp. 271‑296.

Syropoulos, C., G. Felbermayr, A. Kirilakha, E. Yalcin and Y. V. Yotov (2024), The global sanctions data base – Release 3: COVID‑19, Russia, and multilateral sanctions, Review of International Economics, Vol. 32, No 1, pp. 12‑48.

Wooldridge, J. M. (2025), Two-way fixed effects, the two-way Mundlak regression, and difference-in-differences estimators, Empirical Economics, Vol. 69, pp. 2545–2587.

Yotov, Y. V., R. Piermartini, J.-A. Monteiro and M. Larch (2016), An Advanced Guide to Trade Policy Analysis: The Structural Gravity Model, Geneva: United Nations and World Trade Organization.