The end of 2025 and start of 2026 saw the world trading system shatter the rules and informal norms that had sustained globalization for over half a century. What will be the impact? Is the trading system, as President Trump claims, rigged against the United States? Will abandoning the norms of multilateral cooperation in favor of bilateral dealmaking and unilateral tariff threats lead to gains?

In this paper, the authors study these and related questions. They rely on the idea that the tariffs a country chooses to impose reveal how much it values the welfare of its trading partners. For any given import, there is a tariff that would maximize what a country can extract from its trading partners, which the authors coin the opportunistic tariff: the tariff that would maximize what a country extracts from its trading partners, high enough to shift income toward itself, but not so high that lost trade cancels out the gain . A country charging less than that is implicitly placing a positive welfare weight: how many cents on the dollar a country values a transfer to a foreign trading partner compared to a transfer to itself on their trading partner.

The authors systematically measure the size of these welfare weights by combining data on global tariffs with a model of the world economy that computes, for each country and product, both what the opportunistic tariff would be and how any given import restriction affects real incomes across trading partners. They estimate a welfare weight for all 756 country pairs in their sample, and find the following:

  • There is significant international cooperation. In 2001, on average, countries value a dollar transferred to a foreign trading partner 25 cents less than a dollar transferred to their own residents. Not a single country in the sample placed zero weight on any trading partner’s welfare, which is what pure self-interest would predict.
  • Yet the system is not as cooperative as it could be. Even accounting for the substantial welfare weights countries place on one another, global tariffs in 2001 were not set in a way that exhausted all possible mutual gains, meaning the world trading system, cooperative as it used to be, has never reached its full efficiency potential. 
  • There is widespread reciprocity in the world trading system. Countries that put more weight on the welfare of foreigners also tend to receive higher welfare weights from foreigners. In 2001, the United States was a slight outlier: while not treated poorly by other countries, it did tend to place higher welfare weights on them than they placed on it in return.
  • All countries lose from exiting the global trading system. When the authors simulate what would happen if a country stopped placing any weight on its trading partners’ welfare, and others responded by withdrawing their cooperative treatment in return, every country in the sample ends up worse off. The median real income loss is 1.9%.
  • The authors then consider an extreme scenario in which the United States stops placing any weight on its trading partners’ welfare while managing, perhaps through threats of further punishment, to induce every other country to value US welfare as highly as its own. Even then, the United States would gain only 0.8% of real income. To put that in perspective, if the gambit failed and other countries responded by cutting off trade with the US entirely, the losses would stand at 17.8%.

This research reveals a web of mutual restraint, built up over decades, in which every country, including the United States, benefits from others’ cooperative behavior more than it would gain from defecting. The findings have implications extending well beyond the current moment, speaking to ongoing debates about trade wars, economic sanctions, geoeconomics, and the conditions under which international cooperation, once unraveled, can be restored.

Written by Abby Hiller Designed by Maia Rabenold