Working Families Pay the Price for Record Tariffs

In August 2025, U.S. Customs and Border Protection collected roughly $31 billion in duties for a single month, the highest monthly haul on record and more than four times what the agency collected in August 2024. The figure appears on line "Customs and Certain Excise Taxes" of the Daily Treasury Statement for August 29, 2025. The administration calls this revenue. It is revenue. The question worth asking is whose pocket it came out of, and whose pocket it landed in.

The answer is on every receipt at every register in the country, even when the line item does not say so.

Who actually writes the check

The legal incidence of a tariff falls on the importer of record. A retailer, wholesaler, or manufacturer buying goods from abroad pays the duty at the port. That is the formal bookkeeping. The economic incidence, meaning who absorbs the cost in the end, is a different question and an empirical one.

Two recent studies settle the matter for the current round. Cavallo, Llamas, and Vazquez, "Tracking the Short Run Price Impact of U.S. Tariffs" (Harvard Business School working paper, August 2025), using scanner data on retail prices, find that roughly 60 percent of the 2025 duty has passed through to consumer prices within four months of imposition, with pass through still rising. The Yale Budget Lab, in "State of U.S. Tariffs: September 4, 2025," estimates the average household will pay $2,400 more in 2025 because of the tariff schedule announced through that date.

These numbers are not projections from advocacy shops. The Cavallo paper uses point of sale data covering tens of thousands of products. The Yale Budget Lab is staffed by former CBO and Treasury economists and publishes its methodology in full. Both find the same pattern the 2018 tariff round produced: importers eat a sliver, foreign exporters eat almost nothing, and the household at the register eats the rest.

Why the household at the register, specifically

A flat consumption tax is regressive by construction. A family in the bottom quintile spends roughly 80 percent of its income on goods and services, much of it on tradable categories: clothing, footwear, small appliances, toys, electronics, packaged food. A household in the top quintile spends closer to 50 percent of income on consumption, with a much larger share going to services, real estate, and financial products that the tariff schedule does not touch.

The Yale Budget Lab estimate breaks this down by income decile in Table 3 of the September 4 release. The bottom decile loses 4.0 percent of after tax income to the 2025 tariffs. The top decile loses 1.6 percent. In dollar terms the top decile pays more, but as a share of what the family has to live on, the bottom decile pays roughly two and a half times the rate. That is the textbook signature of a regressive tax. The administration is welcome to dispute the framing. The arithmetic does not move.

Apparel and footwear are the cleanest illustration. The United States imports roughly 97 percent of the apparel sold domestically and roughly 99 percent of footwear, per the American Apparel and Footwear Association's "ApparelStats and ShoeStats 2024" report drawing on Commerce Department trade data. There is no domestic substitution to speak of, because the domestic capacity does not exist at price points the bottom three quintiles can afford. A 30 percent duty on a $40 pair of children's sneakers from Vietnam does not cause a Pennsylvania factory to spin up. It causes the sneakers to cost $48 to $52, and the family that buys them to absorb the difference or buy a worse pair.

Where the money goes

Tariff revenue flows to the General Fund of the Treasury. From there it is fungible. The administration has been explicit, in repeated statements from the President and the Treasury Secretary, that the tariff revenue is intended to offset the cost of the 2025 tax package, which extended and expanded the 2017 individual rate cuts, raised the estate tax exemption to roughly $15 million per individual, and made permanent the qualified business income deduction for pass through entities.

The Tax Policy Center's "Distributional Analysis of the One Big Beautiful Bill Act" (July 17, 2025) finds that 64 percent of the net tax cut accrues to the top quintile of households, with 38 percent going to the top one percent alone. The bottom quintile receives an average cut of about $130 per year. The top one percent receives an average cut of about $61,000 per year.

Set the two flows next to each other. Tariff revenue is collected disproportionately from lower income and middle income consumers at the cash register. General Fund outlays, including the tax cuts the revenue is explicitly meant to finance, accrue disproportionately to households in the top decile and top percentile. The transfer is not hypothetical. It is visible in two separate sets of Treasury and JCT tables that anyone with a browser can pull up.

This is what the Transfer Ratio measures: dollars extracted from the bottom four quintiles per dollar delivered to the top percentile. Run the numbers. The bottom four quintiles, roughly 104 million households, pay about $1,500 each in tariff driven prices and receive about $400 each in tax cuts, for a net loss of roughly $114 billion. The top one percent, roughly 1.3 million households, pay about $9,800 each in tariff driven prices and receive about $61,000 each in tax cuts, for a net gain of roughly $66 billion. The Transfer Ratio is $1.73 extracted from the bottom four quintiles for every $1.00 delivered to the top percentile. The direction of the flow is not in dispute. It is on the tables.

The political language versus the accounting

Tariffs are described in the current discourse as a tool against foreign producers, as leverage in trade negotiations, as a revival of domestic manufacturing. Those claims are testable on a decade horizon. The register bill is due now. None of the long run claims changes who pays the bill in the meantime. The bill is paid at the register, by the household whose grocery cart and back to school list are full of imported inputs, and the proceeds flow into a Treasury account that funds, among other things, a tax schedule tilted toward households that do not shop the way the bottom three quintiles shop.

This is not a claim about motive. Motive is unknowable and beside the point. The system is visible: a regressive consumption levy on the bottom four quintiles funds a tax cut concentrated in the top one. Calling the levy a tariff and the cut a growth measure does not change the direction of the dollars. Treasury's own daily statement and JCT's own distributional tables describe the same transfer in different vocabularies.

What would change the picture

Three things would. A demonstration that foreign exporters, not American consumers, are absorbing the duty. The Cavallo scanner data does not show this. A demonstration that the tax cut financed by the revenue is broadly distributed. The TPC tables do not show this. A demonstration that domestic capacity is replacing imports at comparable price points. The Census Bureau's manufacturing surveys through Q3 do not show this.

Until one of those three appears, the accounting holds. The tariff is a tax. It is paid at the register. The receipts run uphill.

What say you?

Originally published at https://henrygoodstone.com/commentary/working-families-pay-the-price-for-record-tariffs/