Who bears the burden? Tariff incidence in U.S. import markets, 2018-2025
This working paper by Gita Gopinath and Brent Neiman examines how U.S. import tariffs imposed in 2018-2019 and 2025 affect prices, sourcing patterns, manufacturing costs, and exchange rates. The core question is: who bears the economic burden of tariffs, U.S. importers or foreign exporters? The study combines multiple data sources and analytical approaches:
Data sources:
– U.S. Census Bureau trade data on 19,000 product categories
– Bureau of Labor Statistics import price indices adjusted for quality changes
– Input-output tables linking tariff impacts to domestic manufacturing costs
The authors distinguish between announced tariff rates and actual tariff rates, which is a critical distinction. By September 2025, the announced statutory rate was 27.4%, but the actual collected rate was only 14.1%. This gap arises from shipping delays (goods in transit when tariffs are announced), product exemptions, compliance with trade agreements like USMCA, and uneven enforcement.
To measure tariff pass-through, Gopinath and Neiman examine whether import prices rise by the full amount of the tariff (complete pass-through) or whether exporters cut prices to absorb some of the cost (partial pass-through). They analyze both aggregated price indices by country and sector, and detailed product-level unit values from trade data. To isolate tariff effects from underlying price trends, they detrend price data using historical pre-tariff patterns.
A novel contribution is calculating “production tariffs” which is the equivalent tax rate on U.S. manufacturing costs when imported inputs face tariffs. It combines tariff rates with input-output data showing which industries depend most heavily on tariffed imports.
Findings
1. Tariff pass-through is nearly complete
The central result: exporters have not reduced prices in response to tariffs. Instead, U.S. importers pay nearly the full tariff increase:
In 2025, for every 1 percentage point of tariff increase, import prices rose by approximately 0.94 percentage points. This means U.S. businesses and consumers bear the bulk of tariff costs, not foreign suppliers.
2. Exchange rate pass-through is minimal
Interestingly, when the dollar weakens (making imports more expensive naturally), exporters are much more willing to absorb costs by cutting prices. Exchange rate pass-through is only 6-8%, far below tariff pass-through. Gopinath and Neiman suggest exporters view currency fluctuations as temporary, while tariffs appear more permanent.
3. Dramatic shift in import sources
China’s share of U.S. imports plummeted from 22% (end-2017) to just 8% (September 2025). Beneficiaries of the “China+1” diversification strategy—India and Vietnam—captured substantial gains. However, whether these reflect genuine domestic production or Chinese goods being transshipped through other countries remains unclear.
4. U.S. manufacturing costs rise significantly
Because U.S. manufacturers depend heavily on imported inputs (machinery, components, materials), they face a “double hit”: they pay higher tariffs on inputs and cannot fully pass these costs to customers because of domestic competition.
The authors calculate that manufacturing production costs increased by over 1 percentage point sector-wide in 2025. High-tariff sectors saw even larger increases—some industries faced production cost increases exceeding 3-4 percentage points. This directly undermines the stated goal of making U.S. manufacturing more competitive.
5. Dollar depreciation contradicts tariff theory
Standard economics predicts that tariffs should strengthen the currency (importers need more foreign currency to pay tariffs). This happened in 2018-2019: the dollar appreciated as expected.
In 2025, however, the dollar has weakened significantly despite record tariff levels. Gopinath and Neiman speculate that other macroeconomic forces such as uncertainty, expected Federal Reserve rate cuts, and foreign investors reducing dollar holdings, have overwhelmed the tariff effect.
The Incidence of Tariffs: Rates and Reality
Authors: Gita Gopinath, Brent Neiman
From: Harvard University, University of Chicago
Production networks and tariff costs
This paper by Ṣebnem Kalemli-Özcan, Can Soylu, and Muhammed A. Yildirim develops a global New Keynesian Open Economy model to analyze how U.S. tariffs in 2018-2019 and 2025 affect macroeconomic outcomes across different countries. Unlike standard trade models, their framework incorporates global production networks, price rigidities varying by sector, incomplete financial markets, trade imbalances, and heterogeneous monetary policy responses across countries. The authors build a multi-country, multi-sector dynamic general equilibrium model with five key components:
- Consumption shares: How much each country prefers home vs. foreign goods
- Production networks: Full input-output linkages showing how industries depend on imported intermediates
- Elasticities of substitution: How easily countries can switch between domestic and foreign suppliers
- Nominal rigidities: Sector-specific price stickiness calibrated from Nakamura and Steinsson (2008)
- Monetary policy rules: Country-specific Taylor rules capturing how central banks respond to inflation
Kalemli-Özcan et al. derive a “New Keynesian Open Economy Leontief Inverse” matrices that trace how tariffs propagate through interconnected global production chains. Finally, they use data from the OECD Inter-Country Input-Output tables (2019), and WTO-IMF Tariff Tracker.
Findings:
The central result: 2025 U.S. tariffs cause stagflation globally i.e. rising inflation combined with falling output with the costs distributed unevenly.
2025 Unilateral U.S. Tariffs (Actual Implementation)
All-Out Trade War (Symmetric Retaliation)
Dynamic effects over time
- Short run (Year 1): All countries experience inflation spikes and output contractions on impact. U.S. trade balance improves temporarily.
- Medium run (Years 2-5): Under unilateral tariffs, U.S. output partially recovers due to high substitution elasticity allowing some production reshoring, but consumption remains depressed. Under retaliation, no recovery occurs as output stays -1% below baseline. Trade balances return to original levels as tariffs don’t change fundamental consumption-saving behavior.
- Long Run: Trade deficits return to pre-tariff patterns because households’ intertemporal consumption-saving decisions are unchanged by tariffs. This confirms theoretical results in Itskhoki & Mukhin (2025) that financial positions, not tariffs, determine long-run trade balances.
Global Networks, Monetary Policy and Trade
Authors: Ṣebnem Kalemli-Özcan, Can Soylu, Muhammed A. Yildirim
From: Brown University, Harvard University
