The latest evidence on tariff incidence

Introduction

Since early 2025, US trade policy has shifted sharply toward protectionism, marked by a broad set of tariffs imposed under emergency and trade-related authorities. These measures, justified by national security and industrial strategy, are wider in scope and more volatile than previous episodes, and have seen frequent reversals and exemptions. Our prior work documents the short-run macroeconomic effects of this policy turn, rising inflation and slower growth (Benigno et al., 2025)¹, consistent with models of permanent trade shocks (Schmitt-Grohe and Uribe, 2025)². This note summarises recent research (Wales, 2025)³ evaluating a central political claim: that the costs of tariffs are being absorbed by foreign exporters, rather than US consumers or firms.

We find this claim has little support in theory or in practice. Customs and fiscal data confirm that US tariffs are generating over $300 billion annually, roughly 1% of annual nominal US GDP. However, this revenue does not reveal who ultimately pays. Using detailed trade data, we show that exporters have diverted roughly 15% of affected trade flows away from the US, shifting sourcing from China to countries like Vietnam and South Korea. Meanwhile, there is little evidence of exporters reducing dollar prices, mirroring the 2018-2019 episode (Amiti et al., 2019)⁴. While domestic prices have so far responded only modestly, with core PCE rising around +0.25pp, model-based estimates suggest larger long-run pass-through is likely (Benigno et al., 2025; Cuba-Borda et al., 2025)⁵. The burden of these tariffs is therefore falling primarily on the US side of the border.

Measuring Tariffs and Diversion

To analyse tariff transmission, we construct a high-frequency dataset of announced tariffs by country and product, weighted by 2024 import shares to estimate an average US tariff rate. As Figure 1, Panel (a) shows, the tariff rate rose from around 2.5% in January to a 28% peak in April, before moderating due to suspensions and delays. The current effective tariff rate is about 16%, slightly lower than announced rates due to implementation lags, exemptions, and trade adjustments.

Figure 1, Panel (b) decomposes the tariff rate increase into changes in tariffs themselves and shifts in import shares. This is known as a shift-share analysis. Firms have rerouted supply chains to avoid high tariffs, largely substituting imports from China with those from Vietnam and South Korea. This trade diversion offsets roughly -1.4 percentage points of the +8.4pp tariff increase. The result is a significant dilution of the headline policy, as trade diversion, rather than exporter price cuts, explains some of the observed adjustment. In the underlying data we show that this largely reflects substitution away from China with corresponding increases in trade from Vietnam and South Korea.

Figure 1: US Tariffs

(a) Announced US Import Tariffs

(b) Change in US Tariff Rate

Sources and Notes: Fulcrum Asset Management LLP, US Census Bureau, and US White House. Panel (a) shows high frequency tariff tracker based on announced policies, using methodology similar to (Amiti et al., 2019)⁶. Panel (b) uses trade data to show the impact of trade diversion and substitution, partially offsetting announced tariff rates.

Producer Price Pass-Through

A key argument in support of the 2025 tariffs has been that foreign exporters would lower their prices to maintain competitiveness in the US market, thereby shielding American buyers from the full cost. But the data tells a different story. When we look directly at the prices paid by US importers at the border, before any tariffs are added, we find little evidence (so far) that exporters have adjusted their prices downward. These “producer prices” have remained remarkably stable, even as tariffs have risen sharply. This pattern mirrors what happened during the 2018 tariff increases under the Trump administration. Despite major shifts in trade policy, foreign firms largely held firm on their prices, leaving importers to absorb the extra cost.

Instead of adjusting prices, exporters have responded by reducing the quantity of goods shipped to the US. These volume declines have been remarkably universally distributed across products and regions. The burden has thus shifted to American importers, who either pay more, find new suppliers, or reduce purchases altogether. While some of the impact has been softened by trade diversion, sourcing from countries not targeted by the tariffs described above, there is little sign that foreign producers are bearing the cost through lower prices. Across two very different episodes of protectionism, the 2018 and 2025 episodes, the conclusion is consistent: US tariffs have raised the cost of trade, and the adjustment has come not through foreign price cuts, but through changes in volumes and trading partners.

Establishing a Benchmark

We introduce a simple economic model, in the spirit of Barbiero and Stein (2025)⁷ and Minton and Somale (2025)⁸, that helps us understand how tariff shocks, sudden increases in taxes on imported goods, affect domestic prices. This model, detailed fully in the accompanying paper (Wales, 2025)⁹, features four agents: households, domestic firms, foreign firms, and the government. It accounts for how consumers might switch between buying imported or domestic products and how companies rely on both domestic and imported materials in their supply chains. By mapping these connections, the model shows how a tariff shock spreads through the economy, influencing prices and production. This framework helps us understand how tariff shocks ripple through the economy and impact prices.

We find that, when this model is mapped to the US data, the effect of recent tariff announcements on consumer prices should be substantial. Initially, around 2 April, the tariff shock was expected to increase headline inflation by about +2 percentage points and core goods inflation by +6pp. However, after many tariff announcements were reversed, the net estimated effect dropped significantly to roughly +0.9pp for headline inflation and +2.6pp for core goods inflation. These effects are shown in Figure 2, panel (a). These findings suggest that while tariffs can raise prices, subsequent policy changes have substantially moderated their overall impact.

Looking more closely at specific product categories, the largest price increases are projected for goods like games, toys, and hobbies, primarily due to direct increases in import prices. Interestingly, between one quarter and one third of the overall price effect comes from indirect channels, as tariffs influence prices through complex supply chain connections involving both domestic and imported intermediate goods. This highlights that tariffs affect consumer prices not only directly but also through the broader production network.

Figure 2: Tariff Impact on Price Level

(a) Tariff Impact on Price Level

b) Pass-Through Regression Results

Sources and Notes: Fulcrum Asset Management LLP, US Census Bureau, and BLS. Panel (a) shows theoretical impact on inflation from announced tariff shocks split into the direct and indirect components. Panel (b) shows results from domestic price pass through regressions from tariff announcements. The y axis represents the proportion of tariff announcements which have passed into consumer prices each month in 2025.

Current Tariff Incidence

Finally, we compare these theoretical tariff impacts to those seen in the monthly data. The model estimates how much prices should rise based on tariff changes, while observed price data show what has actually happened so far. The degree to which expected price increases have been passed through to consumers has increased over time. Our results suggest that around 40% of the theoretical price increases have already passed-through to consumer prices by July. These effects are shown in Figure 2, panel (b), and have increased substantially with the June and July data. However, this still suggests that most of the tariff costs have not yet been fully reflected in consumer prices.

Given that neither foreign producers, nor domestic consumers have yet absorbed the majority of the tariff shock we conclude by supposing that domestic firms have absorbed the majority of the shock in their margins, so far. This finding is contrary to some views that suggest the cost is shared evenly between domestic producers and foreign exporters, the evidence indicates a very uneven split. Domestic producers have absorbed the vast majority, around 60% of the tariff shock so far, while foreign firms have absorbed almost none. Consumers have seen an increasing share of the cost passed on to them at this stage, meaning the tariffs have mainly squeezed company profit margins rather than significantly raising retail prices.

Looking ahead, our analysis suggests that if domestic firms eventually restore their usual profit margins, more of the tariff costs could be passed on to consumers over the coming months. This means that while consumers have not yet felt the full impact of tariffs, there is potential for prices to rise further this year.

 Conclusion

This work yields several important conclusions. First, and perhaps most notably, the foreign absorption of the recent tariff shock appears to be functioning as expected: it is minimal, as is consistent with existing literature for other episodes. In other words, foreign producers are not significantly bearing the cost of these tariffs. Instead, the burden of the tariffs largely falls on domestic producers and consumers. These tariffs are being paid by Americans.

However, a rapid inflationary impulse is not necessarily imminent. Our shift-share analysis suggests that roughly 15% of the tariff shock has already been mitigated by trade diversion, as imports shifted from high-tariff to lower-tariff countries. Furthermore, current estimates indicate that firm profit margins have absorbed a substantial portion of the remaining shock. While this absorption is unlikely to persist in the long run, it has helped cushion the short-term impact and smooth the impact over time.

To better understand these dynamics, we employ an explicit model that incorporates intermediate input structures, enabling us to trace the incidence of the US tariff shock across a range of consumer prices. Relative to the benchmark model, we find that only a limited portion of the shock has been transmitted to consumers.