Working Paper: CEPR ID: DP13579
Authors: Thiemo Fetzer; Carlo Schwarz
Abstract: Are retaliatiory tariffs politically targeted and, if so, are they effective? Do countries designing a retaliation response face a trade-off between maximizing political targeting and mitigating domestic economic harm? We use the recent trade escalation between the US, China, the European Union (EU) and the North American Free Trade Agreement (NAFTA) countries to answer these questions. We find substantial evidence that retaliation was directly targeted to areas that swung to Donald Trump in 2016 (but not to other Republican candidates running for office in the same year). We further assess whether retaliation was optimally chosen using a novel simulation approach constructing counterfactual retaliation responses. For China and particularly, for Mexico and Canada, the chosen retaliation appears suboptimal: there exist alternative retaliation bundles that would have produced a higher degree of political targeting, while posing a lower risk to damage the own economy. We further present evidence that retaliation produces economic shocks: US exports on goods subject to retaliation declined by up to USD 15.28 billion in 2018 and export prices have dropped significantly. Lastly, we find some evidence suggesting that retaliation is effective: in areas exposed to retaliation Republican candidates fared worse in the 2018 Midterm elections, and similarly Presidential approval ratings, especially among Democrats, have declined.
Keywords: trade war; tariff targeting; political economy; elections; populism
JEL Codes: F13; F14; F16; F55; D72
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Political support for Trump in 2016 (D79) | Retaliatory tariffs targeting (F19) |
| Retaliatory tariffs targeting (F19) | Economic harm (F69) |
| Retaliatory tariffs targeting (F19) | Decline in U.S. exports on goods subject to retaliation (F19) |
| Retaliatory tariffs targeting (F19) | Worse performance of Republican candidates in 2018 midterm elections (D79) |
| Economic harm (F69) | Political outcomes (D72) |