The Impact of Tariff Refunds on Business Activity

July 08, 2026

In 2025, the United States sharply raised tariffs, the duties on goods imported from abroad. In February 2026 the Supreme Court invalidated some of these tariffs, and the Court of International Trade ordered the duties to be refunded. In total, this refund will amount to approximately $166 billion, representing roughly 63 percent of 2025 customs-duty receipts (York and Kraschel, 2026).

With refunds now flowing to firms, attention has turned to how and how much the $166 billion will affect future business activity. Some press coverage has highlighted examples of firms' plans to use the rebates to invest and expand.¹ However, the headline figure might overstate the stimulus effect of refunds. These payments compensate firms for duties paid in the past and are paid out regardless of what firms do next. Because of this backward-looking nature of refunds, many recipients might leave hiring, investment, and pricing decisions unchanged.

The extent to which a refund stimulates firm-level investment and hiring critically depends on whether the firm is financially constrained. We attribute 34 percent of the total refunds, about $56 billion, to firms we classify as most financially constrained-and therefore might be more likely to use the money to expand investment, increase hiring, or reduce prices. The remainder goes to firms that are more likely to save the refund, use it to repay debt, or distribute it to shareholders, with little or no immediate impact on their operations.

To obtain our results, we draw on data from Penciakova, Nguyen, Minoiu, and Taylor (2025) to impute tariff exposure and refunds and construct a proxy of firm-level financial constraints. These constraints are key to understanding how firms respond to refunds. As we discuss in more detail below, a financially constrained firm that receives an unexpected refund is more likely to put it to immediate use-for example, by investing, hiring, or adjusting prices.

Why financial constraints matter

To provide goods and services, a business must pay for equipment, inputs, and workers, often well before receiving payments from its customers. The necessary funds might come from owners' capital, loans, credit lines, and/or internal cash. Even a firm with strong demand might lack sufficient liquidity to operate at its desired scale. When this condition exists, we consider the firm to be financially constrained.

Whether a cash inflow, such as a refund, stimulates investment and hiring depends on whether the firm is financially constrained. If a firm is unconstrained, it is operating at its desired level. By definition, no constraints prevent it from carrying out its activities to maximize profits. Therefore, a payment unrelated to fundamentals that affect demand or costs is unlikely to fuel more investment or expansion (Modigliani and Miller, 1958). Instead, the firm is likely to save the cash, use it to repay debt, or pay shareholders (Dharmapala, Foley, and Forbes, 2011; Albertus, Glover, and Levine, 2025). By contrast, a financially constrained firm could be holding back on investment or hiring owing to a lack of funds. In this case, a cash windfall relaxes its constraint, allowing the firm to fund investment or hire workers it otherwise could not (Fazzari, Hubbard, and Petersen, 1988).

Financial constraints can also affect firms' pricing decisions. When a firm has limited access to external finance, it has an incentive to keep prices high to generate cash, even at the risk of losing existing customers or failing to attract new ones. Additional cash, such as refunds, could ease this pressure, allowing the firm to lower prices or hold them steady when costs rise (Gilchrist, Schoenle, Sim, and Zakrajšek, 2017).

Measurement

To calculate how much of the $166 billion in refunds is likely to be spent rather than saved, we classify importers by how likely they are to face tight financial constraints. To do so, we use data from Penciakova et al. (2025), which links S&P Global's Panjiva Supply Chain Intelligence data on maritime shipments with Dun & Bradstreet's National Establishment Time Series (NETS). The linked dataset tracks 33 million shipments from 2022 to 2024 across nearly 175,000 importers. For each importer, the data include employment, industry, and access to finance. For each shipment, we observe the origin country, product code, and volume. Because the data are limited to maritime trade and to importers matched to NETS, they cover only a subset of all US importers, and we therefore calculate refunds for the firms in our sample and then scale the results to reflect the overall US economy. The resulting dollar figures are therefore imputations that assume our sample is representative of the full importer population.

We identify which importers are likely to face tight financial constraints using two indicators. The first is firm size, measured by employment, which research has found to be a strong empirical predictor of financial constraints.² The second is the Dun & Bradstreet PAYDEX score, which rates on a scale of 1 to 100 how reliably a firm pays its vendors, suppliers, and creditors. We consider a missing or weak PAYDEX score to be a signal of limited credit visibility and therefore a reasonable proxy for limited access to external finance. We should note, however, that these groupings are rough proxies rather than direct measures of binding financial constraints: a missing PAYDEX score could reflect factors such as limited use of trade credit, rather than lack of access to external finance. But taken together, the two indicators are helpful in identifying the firms most likely to face meaningful financing frictions.

We sort firms into three groups. Firms most likely to be financially constrained are small and medium-sized enterprises (SMEs, those with fewer than 250 employees) that have no business credit score. Firms that are somewhat constrained are SMEs with a below-median business credit score. All other firms (SMEs with above-median scores and large firms) are classified as least constrained because these firms typically have access to various sources of external finance.

We calculate each firm's refund by computing its imputed shipment value at the product and origin-country level-where we classify products using HS-2 codes-multiplying that value by the change in the tariff rate between 2024 and 2025 for the same product and origin country and summing the resulting figure across all products and origin countries.³

Results

Our results show that about a third of refunds flow to firms we classify as most constrained, those most likely to use the funds to adjust investment, hiring, or pricing. Figure 1 summarizes the distribution of importers, employment, and refunds across our three groups.

Figure 1(a) shows that most constrained importers (SMEs with no business credit score) account for about 66 percent of importing firms but only 10 percent of importer employment, reflecting their small size. Somewhat constrained importers (SMEs with below-median PAYDEX scores) account for another 14 percent of firms and 8 percent of employment. The remaining firms, which we classify as least constrained, account for a minority of importing firms but represent the vast majority of importer employment.

Figure 1(b) shows how refunds are distributed across these groups. Most constrained importers account for 34 percent of imputed refunds (roughly $56 billion when scaled to the aggregate). Somewhat constrained importers account for an additional 20 percent, about $33 billion. The remaining 46 percent, roughly $77 billion, is attributed to least constrained firms.

In figure 2, we dig into the types of firms to which we attribute these imputed refunds, and how large the refunds are relative to firm size. Figure 2(a) shows the sector composition of refunds for each group, which largely reflects the industry composition of each group's imports. One interesting pattern is that most constrained firms have a noticeably larger services share (about 21 percent, compared to just 5 to 6 percent for the other two groups) and a smaller manufacturing share. To the extent that services firms tend to have fewer tangible assets to pledge as collateral, this sectoral tilt is broadly consistent with the tighter financing frictions we associate with this group.

We find that for a meaningful share of financially constrained firms, the refund is large relative to their size and could affect their operating decisions. Figure 2(b) reports imputed refunds per employee for each group, calculated both as a simple average across firms and as the midpoint of the distribution within each group.⁴ Mean refunds per employee are highest for most constrained firms, at $55,103, compared to $49,203 for somewhat constrained firms and $37,463 for least constrained importers. However, because a handful of large refunds can pull the average up, a better guide to the experience of a typical firm is the midpoint of the distribution (that is, the median). By this measure, the refund for most constrained firms is about $5,091 per employee, more than twice the $2,361 for a typical least constrained firm, with somewhat constrained firms falling in between at $3,500.

Conclusion

Although $166 billion in refunds is substantial, not all of it will promote business activity. Only 34 percent of total refunds flow to firms we classify as most constrained-those most likely to put the money to immediate use. Even within this group, the refund's effect on business activity will vary widely: though some firms have outsized refunds relative to their size, the typical most constrained firm receives a more modest sum.

Ultimately, the economic impact of refunds is likely to be smaller than what one would infer from the $166 billion figure, and uneven even within the group of firms most likely to respond. How the refunds are distributed across firms-and how concentrated that distribution is-matters as much as the total amount.

References

Albertus, James F., Brent Glover, and Oliver Levine. 2025. The real and financial effects of internal liquidity: Evidence from the Tax Cuts and Jobs Act. Journal of Financial Economics 166: 104006.

Dharmapala, Dhammika, C. Fritz Foley, and Kristin J. Forbes. 2011. Watch what I do, not what I say: The unintended consequences of the Homeland Investment Act. The Journal of Finance 66.3: 753-87.

Fazzari, Steven M., R. Glenn Hubbard, and Bruce C. Petersen. 1988. Financing constraints and corporate investment. Brookings Papers on Economic Activity 1988.1: 141-95.

Gilchrist, Simon, Raphael Schoenle, Jae Sim, and Egon Zakrajšek. 2017. Inflation dynamics during the financial crisis. American Economic Review 107.3: 785-823.

Hadlock, Charles J., and Joshua R. Pierce. 2010. New evidence on measuring financial constraints: Moving beyond the KZ index. Review of Financial Studies 23.5: 1909-40.

Modigliani, Franco, and Merton H. Miller. 1958. The cost of capital, corporation finance and the theory of investment. American Economic Review 48.3: 261-97.

Penciakova, Veronika, Valerie Nguyen, Camelia Minoiu, and Lauren Taylor. 2025. Are US importers ready for the new tariff landscape? Federal Reserve Bank of Atlanta Policy Hub: Macroblog: atlantafed.org/research-and-data/publications/policy-hub-macroblog/2025/08/26/are-us-importers-ready-for-new-tariff-landscape (August 26).

Skadden, Arps, Slate, Meagher & Flom LLP. 2026. Tariff refund mechanism takes shape after the Supreme Court's IEEPA ruling. skadden.com/insights/publications/2026/03/tariff-refund-mechanism-takes-shape.

York, Erica, and Emily Kraschel. 2026. Liberation Day was one year ago: Did the president's tariff promises happen? Tax Foundation blog post, March 30. taxfoundation.org/blog/liberation-day-trump-tariffs.

Zwick, Eric, and James Mahon. 2017. Tax policy and heterogeneous investment behavior. American Economic Review 107.1: 217-48.