Do Tariffs Reduce Trade Deficits?

Tariffs can compress bilateral trade balances mechanically: taxing imports from a specific partner reduces purchases from that partner. But this does not resolve the aggregate trade deficit, which is determined by the gap between national saving and investment. A tariff that shrinks the bilateral deficit with one country tends to redirect imports through other suppliers or shift the deficit to other partners, leaving the aggregate balance largely unchanged.

The Liberation Day tariffs paper illustrates the problem concretely. The April 2025 tariff schedule set rates proportional to bilateral trade deficits, implicitly treating each bilateral imbalance as evidence of unfair trade. This design has no grounding in optimal tariff theory. The welfare-maximizing tariff for the United States is roughly 19 percent and uniform across partners; it depends on aggregate market power, not on which country happens to run a surplus with the U.S.

The deeper issue is that deficit reduction is the wrong objective. What matters for national welfare is the terms of trade, the efficiency of domestic production, and the revenue the government collects net of distortions. The revenue paper shows that even the fiscal case for tariffs is weak: at most 16 percent of revenue can be replaced, and retaliation halves that figure. A tariff that shrinks a bilateral deficit while provoking retaliation and distorting production networks can easily reduce welfare even as the measured deficit falls.

The right question is not whether tariffs move a trade balance, but whether they improve the allocation of resources once all general-equilibrium channels (terms-of-trade effects, retaliation, input-output linkages, and revenue feedbacks) are accounted for.