Pass-through beyond the Border: Oligopoly, Networks, and Tariffs 
Tariffs affect the domestic economy through multiple, correlated channels: as final goods purchased by consumers, as imported intermediate inputs that propagate through production networks, and as foreign competition that influences the pricing decisions of domestic firms. Existing empirical work on tariff pass-through produces a wide range of estimates, likely because studies typically isolate one channel at a time rather than modeling them jointly. This paper addresses that gap by building a unified multisector open-economy model with oligopolistic competition that simultaneously captures cost pass-through, network propagation, and competitive repricing, then applies it to trace the 2025 US tariffs from the legal tariff schedule through aggregate consumer price inflation.
Key Findings
- A cost-only pass-through model dramatically under-predicts observed 2025 producer price movements (actual prices moved three times as much as predicted). Adding network propagation reduces this gap, and allowing for competitive repricing of foreign varieties closes it almost fully—evidence that all three channels are needed to match the data.
- Once network amplification and competitive effects are included, the model shows essentially full pass-through into both producer prices and, ultimately, consumer prices, with no significant difference between pass-through modeled in logs versus levels.
- Generalized method of moments estimation confirms that oligopolistic competition is a quantitatively important and statistically significant channel.
- Over the February–October 2025 tariff escalation, the direct cost channel contributed 0.69 percentage points to PCE inflation, network propagation added 0.12 points, and competitive effects added 0.10 points, for a total of 0.91 points (rising to 1.14 points under a log pass-through assumption).
- Tariff effects were concentrated in March–May 2025. China accounted for roughly 40 percent of the cumulative inflation contribution, and price increases were concentrated in specific commodities (electrical machinery, vehicles) and consumption categories (new vehicles, garments, furniture)—often diverging from the industries where tariffs were nominally imposed due to network and distribution effects.
Implications
The findings demonstrate that credible estimates of tariff pass-through require jointly modeling cost exposure, network propagation, and competitive dynamics. Analyses that isolate a single channel will misattribute the sources of price changes and risk mismeasuring aggregate inflationary impact. The paper also shows that the ultimate incidence of tariffs on consumer prices does not map cleanly onto the products or countries targeted by the tariff schedule itself. Production linkages and distribution margins can relocate the inflationary burden. The paper offers a replicable, publicly implementable framework for mapping any future tariff schedule into predicted consumer price effects, providing policymakers with a tool to anticipate and decompose the inflationary consequences of trade policy before they fully materialize in the data.
Abstract
We develop a multisector open-economy model that maps border price shocks into US producer and consumer prices through imported inputs, domestic production networks, and competition with foreign suppliers. The framework, which we apply to the 2025 US tariffs, accommodates pass-through in either levels or logs and maps directly to public data on input–output linkages, trade, and prices. Allowing firms to respond to the prices of competing imported varieties substantially improves the model’s fit to producer prices, and the resulting model is also consistent with the consumer-price response. Aggregating the estimated price effects across the full consumption basket, the February–October 2025 tariff increases imply a first-order contribution of 0.91 percentage points to headline PCE inflation, including 0.10 percentage points from the foreign competition channel.