Washington refunds $122 billion of tariffs, and companies are keeping the cash

The United States is refunding $122 billion in tariffs collected under President Trump, according to 24/7 Wall St., which noted that consumer tariff costs were larger than the refund pool and cited an estimate that originated with Democrats on a congressional committee¹.

CFO Dive reported on a Federal Reserve Bank of Atlanta survey finding that most companies will hold onto tariff refunds as cash². Nine percent of companies surveyed had already received refunds and fourteen percent believed they were eligible for reimbursement².

Why it matters to investors

The refunds are landing inside reported earnings, and that is the distortion. Bloomberg Economics’ Anna Wong argues that tariff refunds and shortages define the latest earnings season³, which is a warning that the quarter’s numbers are carrying a one-off item that will not repeat.

H&M is the clearest worked example. The retailer’s profit beat estimates on tariff refunds and higher margins, and analysts had flagged the tariff payments in advance as a potential boost to reported profit that would not reflect an improvement in the underlying business⁴. Reuters coverage of the same results noted that refunds from United States tariff payments boosted operating profit, while slow sales undermined the profit boost, and quoted Chief Financial Officer Adam Karlsson pointing to better sourcing and cost control⁵.

For anyone reading through a results season, the practical instruction is to strip the refund out before comparing operating performance year on year. The Atlanta Fed data suggest that most of this money is not being deployed, so it should not be read as a signal about capital expenditure or hiring intentions either².

What to watch

The first marker is the pace of disbursement. Only nine percent of surveyed companies had received refunds at the time of the survey, against fourteen percent that believed they were eligible², so a large share of the $122 billion has yet to arrive¹.

The second is the distributional question 24/7 Wall St. raises: consumer tariff costs were larger than the refund pool, which means the refunds do not make the affected parties whole and do not go to the same parties who bore the cost¹.

The third is disclosure quality. H&M’s results show both the boost and the underlying softness in the same statement^4,5, and Wong’s framing implies that this pattern is general rather than specific to one retailer³.