Tariffs Added 2.9 Points to Goods Inflation, New York Fed Finds

The finding: a price increase that tariffs fully account for

The question of who pays for tariffs now has one of its clearest answers so far. Economists at the Federal Reserve Bank of New York estimate that tariffs imposed in 2025 and early 2026 added 2.9 percentage points to inflation across a sample of consumer goods as of February. Without those tariffs, prices for the same goods would have fallen slightly.111 The research comes from New York Fed economists Mary Amiti and Sebastian Heise and Columbia University professor David E. Weinstein. They covered 67 categories of non-oil consumer goods and left out services.8

The counterfactual is what stands out. CNBC reported that prices in the tracked categories would have dropped by almost 1% if the tariffs had never been imposed.3 Put plainly, all of the inflation shoppers saw in those categories can be traced to trade policy. Prices would otherwise have been drifting down. The researchers state their main rule of thumb directly: each percentage point added to the average tariff raises consumer goods prices by about a quarter of a percent a year later.11

The work is not new to economists. It first appeared in August 2026 as a New York Fed staff report titled "The Anatomy of Tariff Pass-Through into Consumer Prices" and was revised in September.138 It also circulated as an NBER working paper and a CEPR discussion paper.1819 Wider coverage followed a summary on the bank's Liberty Street Economics blog dated October 6.11 Timing explains much of the attention. UPI pointed out that the findings land only weeks before the November 3 midterm elections.2

How a tax at the border shows up on a store shelf

The paper's real contribution is that it follows a tariff in stages, from the border to the checkout. It does more than produce a single headline number.

At the border, foreign exporters barely cut their prices to absorb the tax. About 90% of the tariffs showed up in U.S. import prices, and the increase arrived within roughly a month of a tariff going up. That held even after the researchers accounted for the weaker dollar.11

From wholesale to retail, cost increases get diluted. A 10% rise in import and producer prices caused by tariffs lifts retail prices by about 5.6%. Much of a product's shelf price is distribution margin, such as shipping, warehousing and retail labor, and the tariff does not touch those costs.11

Combined, about 26% of a tariff increase ends up in consumer goods prices.18 The researchers illustrate this with a uniform 10% tariff on all imports, which they estimate would raise consumer goods prices by about 2.6% within a year.1011

The 26% figure is easy to misread. Part of the coverage presented it as proof that companies absorbed most of the cost, and UPI described it as about a quarter of price increases being passed on to consumers.2 That framing is misleading. The figure is low mainly because imported goods make up only part of a retail price, not because foreign suppliers or U.S. retailers are quietly paying the tariff. With about 90% passed through at the border, the burden stays largely with Americans: importing firms first, then their customers. A Peterson Institute review of recent research reaches the same conclusion. It says several studies, this team's work among them, find that U.S. buyers of imports carried about 90% of the tariff burden in 2025.16

The overlooked channel: domestic goods

The paper's most important result for policy is that tariffs push up prices on goods made in the United States, not only on the imports being taxed. About a third of the total effect comes from domestic products.11 The authors identify two ways this happens. U.S. manufacturers pay more for imported parts and materials. And when foreign competitors raise their prices, domestic producers face less pressure and can widen their own margins.819

The researchers say standard pass-through estimates mostly miss this spillover.13 It also unfolds slowly. Retailers pass on about half of the direct import effect within three months and all of it within six. The indirect effect on domestic goods more than doubles between month six and month twelve.11 The authors summarize the point themselves: tariffs affect consumer prices more widely and for longer than the direct effect alone would suggest.3

This undercuts the argument that tariffs on foreign goods leave Americans who "buy American" untouched. If domestic producers raise prices because import competition has weakened, protection costs consumers money even when they avoid imports entirely.

Where the coverage agrees, and where it overreaches

The main numbers are consistent across outlets: 2.9 points, 67 categories, roughly a quarter passed through, two-thirds direct and one-third indirect.231720 The framing is less consistent. One outlet said the entire inflationary pressure in those categories could be attributed to trade policy rather than broader economic forces.6 The authors' own description is more careful. Their estimates compare goods with more and less tariff exposure, which helps strip out general price trends. They say this method cannot show how much of those general trends was itself caused by tariffs.11 The staff report adds that its estimates hold aggregate conditions fixed.13 The fair reading is that the 2.9-point figure is a well-identified relative estimate. It is not a complete accounting of everything tariffs did to the economy, and the true total could be higher or lower.

Some reports also blur two different measures: the price level and the inflation rate. The New York Fed separates them clearly. Tariffs' effect on the consumer goods price level peaked near 3% in February 2026. Their contribution to twelve-month goods inflation also peaked in February, at 2.9 percentage points, and is projected to fall to about zero by August.11 For households, the price level matters more. Prices are rising more slowly now, but they are not going back down.

Consistent with the wider research

The estimate sits within a growing body of evidence. Harvard's Alberto Cavallo and co-authors used daily retail prices and found pass-through of up to 24% on applied tariffs after seven months. That is very close to the New York Fed's figure. They also found that imported goods rose about twice as much as domestic ones.15 Yale's Budget Lab estimated that 46% to 86% of tariffs passed through to imported core goods prices, depending on method. Its estimate is higher because it measures prices of imported goods only, not the full consumer basket.12 The Peterson Institute describes a study by Minton, Ray and Somale that goes further. It finds complete pass-through after about seven months and attributes all of post-pandemic excess inflation to tariffs.16 The St. Louis Fed reported in August that the inflation effect of tariffs had leveled off in the first half of 2026 as effective rates fell.14

The New York Fed result therefore falls in the middle of a range of credible estimates. That makes it hard to wave away as an outlier.

The politics, the courts and the refunds

The administration rejects the conclusion. A White House spokeswoman told CNBC that the administration has always held that foreign exporters ultimately bear the cost of tariffs.3 The border data cut against that claim directly. One report noted that White House economic adviser Kevin Hassett criticized earlier New York Fed tariff research in February, saying it ignored product substitution by consumers and gains in real wages.4

The legal picture has changed considerably. In February the Supreme Court ruled 6–3 that the president had exceeded his authority by imposing emergency tariffs in peacetime.2 The emergency levies were replaced by a lower 10% surcharge, and the tariff effect on the price level eased to about 2% by August.11 The Yale Budget Lab estimated that about $165 billion in unlawfully collected duties could be refunded to importers.12 One analysis cited a finding that those refunds go to corporate importers, not to the shoppers who paid the higher prices.4 This is the companies' side of the story. Firms that passed the tax on to customers may now receive the refund as well.

What comes next

The authors do not expect relief soon. Their forecast shows the tariff effect on the price level rising again after August. That is driven mainly by August 2026 tariffs on Canadian goods that are still working through the system and by a scheduled January 2027 increase on Canadian cars, trucks and auto parts.11 They expect tariffs to add slightly to goods inflation again by mid-2027.8

The clearest conclusion from this research is that tariffs keep pushing prices up long after they are announced. Import prices adjust within weeks, retail prices within months, and domestic producers within a year. Because of that delay, even reversed tariffs leave prices permanently higher. The policy debate has also changed. Whether consumers pay is largely settled. What remains is how much they pay and for how long.