NY Fed: Tariffs Added 2.9 Points to Goods Inflation

Tariffs added 2.9 percentage points to U.S. consumer goods inflation by February 2026, according to new research from the Federal Reserve Bank of New York. Had the Trump administration's 2025-26 tariffs never been imposed, prices across the 67 goods categories the researchers tracked would have fallen slightly instead of rising, the study found.

What the New York Fed's research found

The paper, published on the bank's Liberty Street Economics blog and authored by economists Mary Amiti, Sebastian Heise and David E. Weinstein, set out to measure how much of the increase in goods prices since 2025 is attributable to tariffs rather than other forces like demand or wages. The researchers compared goods with different levels of tariff exposure while holding broader economic conditions constant, then tracked how prices moved as tariff rates changed.

Their headline estimate: by February 2026, tariffs had contributed 2.9 percentage points to the 12-month change in consumer goods prices. Stripped of that effect, the Fed researchers concluded that prices for the goods in their sample would have declined by nearly 1% over the same period. CNBC, which first reported the findings, noted that the researchers did not disclose which 67 product categories they evaluated.

The study also quantifies the pass-through rate directly. For every one percentage point increase in the average U.S. tariff rate, the researchers found that consumer goods prices rose by roughly a quarter of a percentage point one year later. Applied to a hypothetical example, a 10% tariff imposed on all imports would raise consumer goods prices by about 2.6% after twelve months, in the authors' framework.

Why tariffs raise prices on goods that are never imported

About two-thirds of the price increase the researchers measured came from the direct channel: imported consumer goods simply cost more once the tariff is added. The remaining third is indirect, reflecting higher costs for U.S. producers that rely on imported parts and materials, plus what the authors call "strategic complementarity" — domestic companies raising their own prices when competing imports become more expensive. By February 2026, that indirect channel accounted for about 30% of the total price effect, the researchers found, and it builds more slowly than the direct effect because it depends on producer costs working through supply chains over six to twelve months.

The Fed researchers' forecast shows the tariff effect on the consumer goods price level peaking near 3% in February 2026, then easing to about 2% by August 2026. That moderation traces to a legal setback for the tariff program: a U.S. Supreme Court ruling earlier in 2026 struck down tariffs that had been imposed under emergency economic powers, and they were replaced with a lower 10% surcharge, according to the Liberty Street post. CNBC reported separately that the same February court action produced billions of dollars in refunds to retailers that had already paid the invalidated duties.

The key figures

A research team that has already drawn White House pushback

This is not the first time this group of New York Fed economists has put a number on who pays for the tariffs. An earlier Liberty Street Economics post in February 2026, co-authored by Amiti, Heise and Weinstein along with research analyst Chris Flanagan, examined import-price data and concluded that U.S. importers and consumers bore the large majority of the tariff cost rather than foreign exporters discounting their prices to absorb it. That paper drew a public rebuke from the White House: CNBC reported that National Economic Council director Kevin Hassett said the authors should be "disciplined" and called it the "worst paper I've ever seen." The October paper extends that earlier work by tracing the effect through to final consumer prices rather than just import prices, but it comes from the same research team and the same broader dispute over how much of 2025-26's inflation is tariff-driven.

What remains unknown

The New York Fed researchers did not identify the specific 67 product categories in their sample, so readers cannot check whether a particular grocery item, appliance or clothing line was included. The paper also does not isolate how much of the 2.9-point effect shows up in the official Consumer Price Index versus the narrower basket the researchers constructed, and it does not address services inflation, which is driven by different factors like wages and rents. The researchers' own forecast, which extends through September 2027, assumes tariffs stay at their end-of-September 2026 levels except for one scheduled change: a tariff increase on Canadian cars, trucks and auto parts set to take effect in January 2027. If trade policy shifts again before then, the researchers' projected easing of the tariff effect would need to be revised.