Macroblog

On July 8, a 90-day pause on reciprocal tariffs ended. Only to be extended, for the most part, until August 1. We think. Maybe. Don't worry if you're confused; you're not alone. (By the way, here's a quick primer  on the current state of flux.)

Tariff-related uncertainty has skyrocketed since February 2025. One often-cited measure  that captures the frequency of articles about trade policy in U.S. newspapers reached all-time highs in its 40 year history and remains several times its pre-2025 average.

Given the sizeable effect policy-related uncertainty can have on firms' willingness to move forward on capital investment projects and hire new workers, we posed a set of questions to firms in our Survey of Business Uncertainty in June (our fielding period was June 9–20) to better understand the breadth of tariff-related uncertainty and to gauge how long business execs expect it to last.

Across the more than 900 responses to the question of whether respondents' firms were facing "...any uncertainty related to tariffs or trade policy?”, seven out of every 10 executives said yes. While that figure may appear to be high, it touches on the importance of global trade in the US economy and how heavily the issue is weighing on the minds of business decision-makers.

What is, perhaps, unsurprising is that the presence of tariff-related uncertainty is highest for manufacturing firms, a heavily trade-sensitive sector, where nearly 90 percent of execs reported facing tariff-related uncertainty in June. As figure 1 shows, the pervasiveness of tariff-related uncertainty was nearly as elevated in the retail and wholesale trade sector (roughly 85 percent) and in the construction, real estate, and mining sectors (nearly 70 percent). Interestingly, more than half of services firms reported facing tariff-related uncertainty as well. Potential explanations range from concerns over the impact of tariffs on consumer demand for services to worries over retaliatory tariffs . (This explanation especially begs us to field a follow-up question on the specific effect that uncertainty is exerting on these firms!)

While trade policy uncertainty seems to be part of the zeitgeist now, a key question that may likely determine the propensity for this uncertainty to disrupt firms' planning processes and fundamentally change business activity, is whether or not it is fleeting. And, here, business execs are of two minds.

The typical firm (as measured by the median respondent) anticipates tariff-related uncertainty to be largely resolved by November 2025. But measures of central tendency like the mean and median are not very useful in characterizing diffuse (and multi-modal) distributions such as the one shown in figure 2. What we can say is that business executives disagree to a large extent on how long it will take for trade-policy uncertainty to appear in the rearview mirror rather than on the windshield.

Even using simple "eyeball econometrics,” it appears that there are roughly two camps. A narrow majority of firms see trade policy uncertainty settling down by the end of the year. Another, albeit smaller, camp sees uncertainty over the direction of trade policy continuing to be a salient feature of the economic landscape for the better part of the next three years or so.

As events of the past few weeks have made clear, trade policy continues to evolve. For executives in our Survey of Business Uncertainty, uncertainty over tariffs and trade policy remains pervasive. Moreover, beliefs over when this uncertainty will fade vary widely. As of mid-June, a slight majority of firms see a relatively quick end to trade policy uncertainty, but a large cohort expect the fog of uncertainty to persist awhile. We will continue to explore these questions in the Survey of Business Uncertainty, so stay tuned.

By Jose Maria Barrero , assistant professor of finance at Instituto Tecnológico Autónomo de México Business School,

Nick Bloom , the William D. Eberle Professor of Economics at Stanford University,

Steven J. Davis , the Thomas W. and Susan B. Ford Senior Fellow and Director of Research at the Hoover Institution and Senior Fellow at Stanford Institute for Economic Policy Research, and

Kevin Foster, survey director in the Atlanta Fed's Research Department,

Aaron Jalca, economic research analyst in the Atlanta Fed's Research Department, and

Brent Meyer, assistant vice president and economist in the Atlanta Fed's Research Department

The breadth and magnitude of tariff imposition continue to evolve . Similarly, the effect of tariffs on CFOs' outlooks has changed since we first examined the topic prior to the events of early April. Concern over tariffs expressed by financial decisionmakers in The CFO Survey  has risen sharply during the past two quarters. In the most recent survey, which closed on June 6, 40 percent of respondents cited tariffs or trade policy as their firm's most pressing concern. Prior to mid-2024, these concerns were essentially nonexistent (see figure 1).

To understand the evolution of firms' expectations amid a fast-changing trade policy landscape, we analyze a matched sample of firms responding in each of the last three quarters (during which time tariff concerns increased dramatically) and reexamine how CFOs' expectations evolved for firms that sourced their inputs/supplies from abroad versus those that did not.

Firms exposed to tariffs see much dimmer economic prospects
Relative to firms' views at the beginning of the year, the economic outlook of most firms has deteriorated. Even so, we continue to observe a dichotomy between firms sourcing their inputs/supplies from the United States and those importing them from abroad. The latter category of firms expects a much slower growth trajectory for the overall US economy and assigns double the probability to negative real GDP growth as those not importing from abroad (see figures 2 and 3). Interestingly, the delineation appears to be centered on whether a firm sources inputs from abroad, as we do not observe any meaningful difference based on the share of inputs/supplies sourced internationally.

Firms ratchet up price and cost expectations, further downgrade expected revenue
Compared to last quarter, firms importing supplies/inputs from abroad expect even worse performance than their peers across most key metrics. Figure 4 demonstrates that expected unit cost growth has increased for all importing firms, with firms more heavily reliant on imports expecting the highest unit cost increases. The expected unit costs are more than a percentage point higher than what firms expected last quarter, and in most cases they are more than double what firms expected in the fourth quarter of last year. This pattern is quite striking for firms that source more than 30 percent of their inputs from abroad. From the fourth quarter of 2024 to the second quarter of 2025, these firms have revised up their unit cost growth expectations for 2025 from 3.3 percent to 6.8 percent and have ratcheted up their anticipated price growth for 2025 from 3.4 percent to 5.6 percent. That tariff-impacted firms expect to increase prices by 70 to 80 percent of their unit cost growth suggests that margins will likely shrink as these firms absorb some portion of this cost rather than pass it through to customers.

Interestingly, figure 5 reveals that all firms—including those sourcing all supplies/inputs domestically and expecting more subdued unit cost growth—have revised higher their expected price growth from last quarter. Firms that exclusively source inputs/supplies domestically see their price growth as outpacing unit cost growth in 2025 (4.6 percent versus 4.1 percent, respectively), suggesting some growth in margins and, at least in aggregate, consistent with the notion that anticipated price growth from tariff-impacted firms will allow nonimpacted competitors to raise prices (albeit more modestly) without sacrificing market share.

Firms sourcing all supplies/inputs domestically expect significantly higher revenue and employment growth than their peers (see figures 6 and 7). The sizeable increase in expected employment growth among nonimporting firms might result from their much stronger revenue growth expectations, consistent with the potential for these firms to capture market share.

Firms that sourced inputs/supplies from tariff-hit countries have downgraded their expected revenue growth relative to last quarter, though their employment growth has curiously remained only little changed since the fourth quarter of last year.

Looking across all three quarters, we observe a meaningful deterioration in firms' expected performance between the first and second quarter surveys, which corresponds to the period when most of the additional tariffs were announced and when uncertainty surrounding trade policy peaked (see figure 8). To the extent that greater certainty exists surrounding trade policy, or that progress in negotiating policies more favorable to respondents occurs, we might expect their expected performance to improve somewhat moving forward.

Conclusion
Firms' outlooks continued to darken in the second quarter. Respondents to The CFO Survey have revised downward their expectations for real GDP growth and sales revenue growth, and at the same time they anticipate much higher cost and price growth during 2025. Tariff-exposed firms appear to be driving much of the overall deterioration in own-firm expectations. That said, at least to date, firms not importing supplies/inputs from abroad anticipate raising prices too, suggesting some ability to grow margins as their competitors face sharply higher cost growth due to tariffs. Trade policy remains fluid, and further developments are likely to have an impact on firms' expectations for the economy and their own performance. We will continue to closely monitor these developments and provide additional updates later this year.

By Zach Edwards , research analyst on the Regional and Community Analysis team at the Richmond Fed,

Brent Meyer, assistant vice president and economist in the Atlanta Fed's Research Department, and

Daniel Weitz, survey director at the Atlanta Fed

The US Treasury recently announced that it started the process of phasing out production of the penny and will soon stop putting new one-cent coins into circulation, (These CNN  and WSJ  articles, dated May 22, 2025, discuss the announcement.) This decision should not come as a surprise considering the fact that, since the late 1980s, several bills have been submitted to Congress regarding the issue: H.R.3761  (1989), H.R.2528  (2001), H.R.5818  (2006), S.759  (2017), and H.R.1270  (2025). Whereas the media focused on the cost of producing a penny coin (3.69 cents in 2024) and a nickel coin (13.78 cents in 2024), this essay analyzes an equally important issue, which is related to what economists call the "optimal currency and coin denominations." Table 1 compares coin denominations among three similar economies.

Two remarks about table 1: First, although the 1-cent coin is still minted in the euro area, Belgium, Estonia, Finland, Ireland, Italy, Lithuania, the Netherlands, and Slovakia permit rounding of cash payments to the nearest 5 cents. Second, although the 1-dollar coin is in circulation in the United States, it is rarely used, probably because of its size and the weight relative to that of the 1-dollar paper bill, which is also in circulation.

The burden of paying cash: The principle of least effort
Cash payments are two-way exchanges of currency notes and coins. The payer (consumer, buyer) hands in currency notes and coins. Then, if needed, the payer receives change from the payee (merchant, seller) in the form of currency notes and coins. To determine the optimal denominations for each economy, economists and mathematicians have tried to define a metric that would allow them to determine whether the ongoing currency and coin denominations are efficient according to the specific metric. This literature dates to the 1980s and 1990s and is surveyed in this review article . For this blog post, following some of the literature, we define the following metric: The "burden of a cash payment" is the total number of tokens exchanged between the payer and the payee in both directions. A token is defined as one unit of an existing denomination (for example, one penny, one quarter, one dollar, or a 50-dollar banknote).

Following some of the literature, we also assume that both the payer and the payee (for example, a buyer and a seller) adhere to the "principle of least effort." According to this principle, once the payment dollar value is determined, the payer and the payee coordinate the payment so that they minimize the total number of tokens they exchange. That is, they attempt to minimize the sum of tokens that the payer initially hands in to the payee plus the number of tokens handed back to the payer as change.

Figure 1 illustrates the principle of least effort for transaction values from 1 to 25 cents.

Figure 1 shows that when the penny is in circulation (blue dots), payments in the amounts of 1 cent, 5 cents, 10 cents, and 25 cents require the payer to hand in only one coin and therefore no change is needed; a very easy transaction. However, a payment for 13 cents requires a minimum effort of exchanging four tokens. Either the payer hands in a combination of one dime and three pennies—a transaction that requires no change—or the payer hands in one dime and one nickel and receives two pennies back as change. Either way, the minimum effort is four tokens. As another example, figure 1 shows that the burden of paying 17 cents is also four tokens (the payer hands in one dime, one nickel, and two pennies). The burden of paying 18 cents is also four tokens (the payer hands in two dimes and receives two pennies as change).

Now, what would happen if hypothetically the penny is removed from circulation (see the orange dots in figure 1). Using the same rounding guidelines as in Canada , all payments valued between 3 and 7 cents are rounded to 5 cents (nickel), which reduces the burden of paying cash to one token. All payments between 8 and 12 cents are rounded to 10 cents (dime), which also reduce the burden to one token. All payment values from 13 to 17 cents are rounded to 15 cents, for which the burden is two tokens. All payments between 18 and 22 cents are rounded to 20 cents, for which the burden is two tokens. Finally, payments in the amount of 23 or 24 cents are rounded to 25 cents, for which the burden is only one token.

Calculating the burden of paying cash
Cash is the third-most-used payment instrument in the United States (see this report ). The United States has 12 denominations of currency notes and coins: $0.01, $0.05, $0.10, $0.25, $0.50, $1, $2, $5, $10, $20, $50, and $100. (The rarely used $2 bill will be removed from the computations below.) Table 2 displays computations of the burden of cash payments that were derived in this research paper .

The data used in the construction of table 2 are from the 2015 to 2019 editions of the Survey and Diary of Consumer Payment Choice, for which consumers record how much they paid in cash (and other payment methods). However, survey respondents do not record the precise denominations used for their cash payments. For that, we assume that the payee (merchant, seller) has all denominations. However, for the payer (consumer, buyer) we compute two opposite extreme scenarios: one, that payers have all available denominations (column two in the table), and two, that payers have only $20 bills that they withdrew from an ATM (column three). The first scenario underestimates the burden of paying cash because the payer and the payee exchange notes and coins according to the principle of least effort. The second scenario may overestimate the burden because paying with $20 bills may require a substantial amount of change. Nevertheless, it may provide a better approximation of cash transactions.

The first row in table 2 computes the average number of tokens (coins and banknotes) exchanged in all the cash transactions recorded in our consumer survey data. Column (1) shows that the average number of tokens exchanged according to the principle of least effort is 3.14 coins and notes. However, if buyers can pay only with $20 bills, they receive numerous tokens as change and the average burden increases to 4.96 (almost five tokens), as column (2) shows.

The second row applies our data to the same computations assuming that the penny is no longer in use and merchants round the pennies to their nearest 5-cent transaction value according to the rounding rule used in Canada  (which we call symmetric rounding). Under the symmetric rounding rule, payments that end with 1 or 2 cents are rounded down to 0 pennies. Payments that end with 3, 4, 6, and 7 cents are rounded to 5 cents. Payments that end with 8 and 9 cents are rounded up to 10 cents. If buyers have all denominations, the average burden of paying cash falls from 3.14 to 2.729 tokens because pennies are no longer exchanged. If buyers pay only with $20 bills, the average burden falls from 4.96 to 4.55 tokens after the penny is eliminated.

The third row applies when payees (merchants or sellers) deviate from the symmetric rounding rule after the penny is eliminated and round only upwards to the nearest 5 cents. Under this rule, payments that end with 1, 2, 3, and 4 pennies are rounded up to 5 cents. Payments that end with 6, 7, 8, and 9 cents are rounded up to 10 cents. This shows that upward rounding to the nearest 5 cents does not much change the burden of paying cash relative to symmetric rounding. We use this scenario to verify that the penny elimination does not have any significant inflationary consequences (that is, consumer spending does not increase).

Two final remarks
One possible limitation of the results presented in table 2 is the assumption that the payee (seller) can always provide change with the least number of tokens. As happens to all of us, some merchants may run out of quarters, which may increase the number of dimes and nickels that they hand back as change to the payer (customer). However, we compensate for this deficiency by analyzing the scenario in column (2), in which payers do not have any denomination except for $20 bills that they obtained from an ATM.

Finally, the reader may wonder why the elimination of the penny resulted in a relatively small reduction in the burden of paying cash: from 3.14 to 2.729 tokens or from 4.96 to 4.55 tokens, as table 2 shows. This reduction contrasts with the theoretical simulations displayed in figure 1, which show a larger reduction in the range of one to two tokens. The reason for this difference could be that some sellers (and buyers) have already been rounding their cash payments to the nearest 5 cents to avoid dealing with pennies—even now, when the penny is still in circulation. In other words, it is possible that, for some transactions, both the payer and the payee agree to ignore the penny part of the payment. In addition, we cannot rule out the possibility that some survey respondents rounded the penny part of their reported cash payments.

In summary, economic theory suggests that removing the penny is likely to reduce the burden of cash payments in the economy, although the effect appears relatively small in our research, perhaps because some cash payments are already rounded to their nearest 5-cents value either at the point-of-sale or by our survey respondents.

Author's note: I would like to thank Tom Heintjes for most valuable comments and suggestions on earlier drafts and Whitney Strifler for the interactive charts.

By Oz Shy, senior policy adviser and economist in the Atlanta Fed's Research Department

Over the past couple of months, newswires  have focused on the potential for elevated tariff rates to feed through into higher inflation and potentially affect output growth as well. Indeed, Chair Powell, in his last post-FOMC meeting press conference  said, "What looks likely, given the scope and scale of the tariffs, is that…the risks to higher inflation, higher unemployment have increased."

Recent research from economists at the Atlanta Fed suggests that if firms are able to pass through all the costs of tariffs, retail prices would increase significantly―as much as 1.6 percent (depending on how effective tariff rates evolve from here). And even at a 50 percent passthrough rate, the impact on prices would be large enough to be felt in the aggregate (0.8 percent increase in retail prices). How plausible is full passthrough? Going back to the last episode with rising tariffs in 2018, research  showed that the cost of the tariffs was almost entirely passed through onto domestic prices.

Rising tariff rates pose threats to the real side of the economy as well. Another highly cited academic paper  on tariffs points out that in a simple supply and demand model, "higher prices also reduce demand by domestic consumers."

In this environment, where policy changes lead to sharp increases in costs for many firms, we were curious about how firms would respond, especially in light of a potential reduction in demand that typically accompanies a price hike. So, we turned to the Atlanta Fed's Business Inflation Expectations survey (BIE), a monthly survey of Sixth District firms that is well positioned to ask timely questions on economic conditions facing firms. In gathering information for the April 2025 BIE survey, we asked firms about their ability to pass through increased costs caused by a new economic policy without a resulting reduction in demand.

To that end, we posed two questions to firms about their ability to pass through a hypothetical cost increase to consumers. One of the questions was, "Suppose a new economic policy causes unit costs for your main good or service to increase by [10 or 25] percent, effective immediately. Based on current levels of demand for your main good or service, how much of that cost increase would your firm be able to pass through to customers?" The survey was in fielded from April 7 through April 18, at a time when10 percent was the baseline tariff for almost all countries, and 25 percent was the tariff on all foreign-made autos and auto parts. Some countries or regions, including China and the European Union, had much higher tariffs at the time.

The interesting twist in this line of questioning is the inclusion of the phrase "Based on current levels of demand." The interpretation here is that firms are telling us how much of the cost increase they would be able to pass through to customers before it had a negative impact on demand for that good or service.

Although a diversity of views is apparent, on average firms tell us they expect to be able to pass through 51.1 percent of a 10 percent cost increase, and 47.3 percent of a 25 percent cost increase, without reducing current levels of demand. A closer look at the responses reveals a wide range of expected passthrough rates. Compared to the 2018 episode, where research suggests nearly full passthrough of costs into prices, our results suggest many firms believe their customers are price-sensitive enough this time around (perhaps owing to the recent inflationary surge that isn't too far in the rearview mirror) that they cannot pass through the entire cost increase without reducing demand. A natural next step, then, is to evaluate these responses in the context of firms' perceptions of their current demand.

Figure 2 plots the distribution of firms' passthrough percentages by their strength of sales revenue growth relative to "normal." Viewing this distribution reveals that firms with about normal or above-normal sales levels expect to pass through a much higher percentage of the cost increase compared to those firms with lower-than-normal sales levels. For context, half of firms reported having below-normal sales levels, 38 percent reported about-normal sales levels, and 12 percent reported greater-than-normal sales levels. On average, firms that reported "much less than normal" or "somewhat less than normal" sales levels expect to pass through 45.6 percent of the 10 percent cost increase, firms that report "about normal" sales levels expect to pass through a little more than half of the cost increase, and firms reporting "somewhat" or "much" greater than normal sales expect to pass through nearly two-thirds of a 10 percent cost increase. Results were similar for the 25 percent policy-related cost increase question.

To add further context to these results, we can look at the timeseries behavior of firms' quantitative sales gaps. Figure 3 shows that firms, on average, see sales levels flagging relative to normal, on the order of roughly 8 percentage points in the red. Importantly, at the outset of the trade tensions and tariffs episode of 2018–19, firms perceived their sales levels to be about "normal" (in other words, their quantitative average sales gap was near 0 percent). Given the relationship firms' passthrough rates have to their current levels of demand, we can infer that because perceived demand is weaker now that in the prior period of trade and tariff tensions, firms will be more hesitant to fully pass through tariff-related cost increases.

In sum, firms with about normal or greater-than-normal sales expect to be able to pass through more of the cost increases while maintaining the same levels of demand for their goods or services, as figure 2 shows. And figure 3 shows us that those firms are more likely to be larger firms, due to their smaller sales gap compared to "normal." In the aggregate, business executives see their current sales levels as about 8 percentage points below "normal," which is much weaker than firms' relative position entering 2018. In this environment, firms on average anticipate passing through a little more than half of a 10 percent cost increase without damaging demand. It's not yet clear where the average tariff rate will ultimately settle, or how firms' passthrough rates will evolve from here. However, it does appear that most firms anticipate sacrificing demand should they choose to fully pass a tariff-related cost increase on to customers.

By Kevin Foster, survey director,

Aaron Jalca, economic research analyst, and

Brent Meyer, assistant vice president and economist, all in the Atlanta Fed's Research Department

Finding new international suppliers and reconfiguring supply chains take time and money. As Grossman, Helpman, and Redding (2024)  and Baldwin and Freeman (2022)  emphasize, finding new suppliers abroad involves substantial search and matching costs, such as identifying reliable partners, ensuring compliance with local rules and regulations, and establishing new logistics networks. Due to these costs, relationships along the supply chain tend to be sticky, with even large importers often relying heavily on foreign suppliers from a single country (Antràs, Fort, and Tintelnot 2017 ). Multinationals such as Apple might have a geographically diversified supplier base, but they are not representative.

In recent years, global supply chains have come under intense pressure. Rising tariffs, geopolitical risks, and the COVID-19 pandemic have upended the flow of goods around the world (Aiyar, Presbitero, and Ruta 2024 ). Once relying heavily on Chinese suppliers, US importers have started to rethink their sourcing strategies—a shift that economists have dubbed the "great reallocation" of trade (Antràs 2020 , Alfaro and Chor 2023 , Goldberg and Reed 2023, and Fajgelbaum et al. 2024 ). Figure 1 depicts the shift in maritime trade shares for US firms, showing that their import share from China has fallen by close to 30 percent between 2017 and 2023.

Although previous research in international trade has shown that tariffs are significantly changing global trade patterns (see Freund et al. 2023  and Gopinath et al. 2025  for product-level analyses), we know less about how individual firms manage this transition and even less about the role of financial intermediaries in enabling these costly adjustments.

In a new Atlanta Fed working paper (Alfaro et al. 2025), we shed light on this critical question. In it, we explore whether, and how, commercial banks helped US firms navigate supply chain disruptions after the 2018–19 waves of tariffs imposed by the United States on imported goods from China (henceforth, "China tariffs").

Using newly linked data on US firms' trade activities and their banking relationships, we document the vital role of banks—especially that of banks with expertise in Asian-trade finance—in helping firms find new suppliers and thus adjust to the new tariff regime more successfully.

Our analysis draws on two rich administrative datasets that provide a unique window into the micro-dynamics of supply chain relationships and financial intermediation. First, S&P Global's Panjiva Supply Chain Intelligence , a shipment-level database that tracks US importers and their foreign suppliers, allows us to track firms' supplier relationships over time at a very granular level—product by product, supplier by supplier, and country by country. Second, the Federal Reserve's Y-14 dataset  provides data on individual loan contracts between large US banks and corporate borrowers, many of whom are manufacturing firms reliant on Chinese imports for their production. The trade-credit linked data enable us to analyze changes in firms' financing patterns—in particular, their demand for bank credit—in response to trade shocks, and how those changes vary with the business model of banks. The median firm in our data is private, dependent on bank credit, and eight times smaller than the median publicly traded firm.

The challenge of changing suppliers
The 2018–19 China tariffs were a significant shock to global supply chains. Companies that depended on Chinese suppliers for final or intermediate goods faced higher costs and had to decide whether to keep their partners, absorb the higher prices, or seek alternatives elsewhere. Many firms chose to reorient their sourcing strategies.

In our empirical analysis, we find that tariff-affected firms were significantly more likely to cut ties with Chinese suppliers and establish new trade relationships elsewhere in Asia and the rest of the world, extending the country-level findings in the literature (Handley, Kamal, and Monarch 2024 ) to the firm level.

But this shift was far from seamless. On average, firms took more than 2.5 years to establish new supplier relationships. We estimate that the search costs associated with switching suppliers amounted to around $1.9 million per firm, or roughly 5 percent of their annual sales revenue.

The critical role of banks
So how did firms manage this complex and costly adjustment? Our study finds that commercial banks played a central role, not only by meeting importers' demand for more credit, but also by offering valuable information about potential suppliers.

Using the trade-credit linked dataset, we show that tariff-hit importers sharply increased their demand for bank credit. Firms drew more heavily on their existing credit lines and took out new bank loans, often at higher interest rates, reflecting the financial strain of rebuilding supply chains.

Importantly, not all banks were equally helpful. Tariff-hit firms in a relationship with banks that specialize in financing trade with Asia performed significantly better. These firms secured significantly cheaper credit compared to firms borrowing from other banks and were 15 percentage points more likely to find new suppliers outside China (see figure 2). They also reestablished supplier relationships nearly three months faster and grew their Asian import shares by 5.6 percentage points more than comparable firms with non-specialized banks.

In short: having the right bank relationship significantly eased the burden of supply chain reallocation.

Why specialized banks matter
What gave specialized banks their edge? Two factors seem key: better financing terms and better information (Blickle, Parlatore, and Saunders 2023  and Paravisini, Rapopport, and Schnabl 2023 ).

First, firms borrowing from banks with trade finance expertise in Asian markets secured new loans at lower interest rates—close to 19 basis points cheaper—than firms working with other banks. Specialized banks offer lower-cost loans than other banks by leveraging knowledge of their trade-oriented borrowers' creditworthiness and business opportunities in Asia. In turn, improved loan terms made it easier for firms to absorb the costs of searching for and onboarding new suppliers.

Second, specialized banks emerged as information hubs. They leveraged their informational advantage and local knowledge to assist clients in identifying new suppliers abroad. We show that US importers were more likely to match with new suppliers in those countries where their specialized bank also has a local office, allowing it to pursue direct or syndicated lending to local borrowers, or has ties with correspondent banks. Moreover, after the China tariffs, specialized US banks' fee income from advisory services increased significantly at their Asian subsidiaries compared to non-Asian subsidiaries, indicating that the Asian advisory and consulting business grew faster as they helped clients navigate the transition.

Interestingly, these benefits accrued to firms in a relationship with banks specialized in Asian trade finance. By contrast, relationships with banks specializing in European markets had no impact on firms' ability to diversify away from China to other Asian suppliers.

No signs of increased risk or substitution to trade credit
One natural concern is whether this surge in the demand for bank credit from firms affected by the China tariffs led to riskier lending practices. We find no evidence that specialized banks suffered from higher loan defaults or charge-offs during this period. Supporting firms' supply chain diversification did not come at the cost of weaker balance sheets.

We also find no evidence that firms replaced bank credit with increased trade credit from their Chinese suppliers—a reassuring sign that support from specialized banks was genuinely easing financial and informational bottlenecks rather than merely substituting one form of financing for another.

Conclusion
The global economy is increasingly vulnerable to trade shocks, geopolitical shifts, and other disruptions that rattle supply chains. Our findings highlight an underappreciated but vital part of supply chain resilience: the role of financial intermediaries.

Commercial banks can act as both financiers and information brokers, helping firms weather shocks and reconfigure their supply chains more swiftly and successfully. As policymakers and business leaders grapple with the challenges of building more resilient supply networks, recognizing the importance of financial relationships—particularly that of banks' specialized expertise in particular industries and geographies—is crucial.

Disclaimer: The views expressed in this blog post are those of the authors and do not necessarily represent those of the Federal Reserve Bank of Atlanta; the Federal Reserve System; or the International Monetary Fund, its executive board, or its management. A similar version of this post was originally published on VoxEU .

By Laura Alfaro , professor of business administration, Harvard Business School,

Mariya Brussevich , professor of economics, Ibmec-RJ and economist, International Monetary Fund,

Camelia Minoiu, research economist and adviser in the Atlanta Fed's Research Department, and

By Andrea Presbitero , deputy division chief, International Monetary Fund

Chair Powell, in his post-FOMC meeting press conference  last week, noted that "Surveys of households and businesses ...report a sharp decline in sentiment and elevated uncertainty about the economic outlook, largely reflecting trade policy concerns. It remains to be seen how these developments might affect future spending and investment."

Indeed, uncertainty over trade policy has spiked to historic levels in recent months across a variety of measures . Going back to that press conference to hear from Chair Powell again, "There's so much uncertainty. If you talk to businesses, or market participants, or forecasters, everyone is just waiting to see how developments play out..." It's precisely this waiting to see how "developments play out" that can lead businesses, en masse, to pull back on capital investment and hiring.

To assess this concern, we put special questions to nearly 1,000 business executives in the latest Survey of Business Uncertainty, which was fielded from April 14 to 25. We first asked, "How has uncertainty about tariffs, taxes, government spending, monetary policy, or regulation affected your firm's plans for hiring [investment] over the next 6 months?" Forty percent of the executives plan to scale back hiring at their firms due to policy uncertainty, while 45 percent anticipate pulling back on capital investment for the same reason. Only a small percentage plan to ramp up hiring and investment in response to policy uncertainty (see the chart below).

Manufacturing firms, as well as those operating in the trade and transportation sectors, reported a somewhat more broad-based pullback of investment in particular. Just over 50 percent of durable and nondurable manufacturers and slightly less than half of all retail trade, wholesale trade, and transportation firms plan to scale back investment over the near term.

We also investigated which sources of policy uncertainties are weighing most heavily on the minds of execs. If a respondent indicated either pulling back or ramping up hiring or investment due to policy uncertainty, we asked, "What is your firm's top concern with respect to uncertainty affecting your firm's hiring [investment] plans over the next 6 months?" Here, tariffs are overwhelmingly the top source of uncertainty (see the chart below). Monetary policy ranks a distant second, followed by government spending and regulation. Perhaps unsurprisingly, manufacturing firms and those in the trade and transportation industries also more frequently cited tariffs as their top source of policy uncertainty.

Finally, we asked business executives how much policy uncertainty would lead them to alter investment and hiring plans at their firms. Averaging over all survey responses, businesses plan to slow hiring by 13 percent and scale back investment by 16 percent during the next six months due to policy uncertainty.

For executives anticipating hiring cuts, a large group of firms sits at minus10 percent, with a long left tail. More than 5 percent of firms plan to cut hiring by at least 90 percent—essentially imposing a hiring freeze—until key uncertainties resolve. The picture is very similar for near-term investment plans (see the chart below).

Across the firm-size dimension, the survey results noted modest variation in hiring and investment plans by firm size and industry ranging from minus 10 percent to minus 17 percent. Across industries, the mean causal impact of policy uncertainty on hiring was minus 11 percent to minus 17 percent, with the largest impact again observed on manufacturing and trade and transportation industries. For investment, the pattern is similar with a mean response range of minus 13 percent to minus 18 percent across the firm-size dimension and minus 12 percent to minus 24 percent across broad industries.

To add some context, the magnitude of these declines (at least in capital investment) are similar to what executives told us about the restraint induced by key uncertainties back in October 2020 , during the first year of the pandemic. And are much more severe than they anticipated at the onset of the war between Russia and Ukraine. These earlier results were borne out in the pattern of real business fixed investment during the late 2020–22 period, which was marked by tepid (at best) equipment and software spending and a sustained decline in structures investment.

To sum up what the survey results tell us, heightened policy uncertainty (particularly over tariffs) is leading a large share of firms to scale back near-term hiring and investment. On average, the impact of policy uncertainty is lowering hiring and investment by 13 percent and 16 percent, respectively. To paraphrase Chair Powell, businesses are in wait-and-see mode at the moment. Should uncertainty over trade policy continue to go unresolved, its dampening effect will continue to weigh on the near-term growth trajectory of the US economy.

By Jose Maria Barrero , assistant professor of finance at Instituto Tecnológico Autónomo de México Business School,

Nick Bloom , the William D. Eberle Professor of Economics at Stanford University,

Steven J. Davis , the Thomas W. and Susan B. Ford Senior Fellow and Director of Research at the Hoover Institution and Senior Fellow at Stanford Institute for Economic Policy Research,

By Kevin Foster, survey director in the Atlanta Fed's Research Department,

Aaron Jalca, economic research analyst in the Atlanta Fed's Research Department, and

Brent Meyer, assistant vice president and economist in the Atlanta Fed's Research Department

The US Bureau of Economic Analysis (BEA) estimated  that real GDP contracted at an annualized rate of 0.3 percent in the first quarter of 2025, the first negative reading since the first quarter of 2022. As figure 1 shows, this reading was still well above the final nowcast of −2.7 percent growth for what was then the standard GDPNow model and meaningfully above, but closer to, the final "gold-adjusted" model forecast of −1.5 percent.

As detailed here, the gold-adjusted model—unlike the former version—subtracts international trade in gold when estimating net exports of goods. Otherwise, the models are essentially identical, and both of their final projections have the change in private inventories (CIPI) adding 0.30 percentage points to first-quarter real GDP growth.

The April 29 GDPNow nowcasts were well below the BEA's estimate  that CIPI added 2.25 percentage points, accounting for the largest subcomponent contribution forecast error for either of the models. The excerpt below from the GDP release  gives a clue as to why the models were so wrong on this CIPI subcomponent (boldface added):

The largest contributor to the increase in investment was private inventory investment, led by an increase in wholesale trade (notably, drugs and sundries). The estimates of private inventory investment were based primarily on Census Bureau inventory book value data and a BEA adjustment in March to account for a notable increase in imports. For more information on the source data and BEA assumptions for inventories, refer to the key source data and assumptions table (available at 10 a.m.).

It's unclear from the release what the size of BEA's March adjustment was. However, if we take the March values for merchant and nonmerchant wholesale inventories (provided in the BEA's key source data and assumptions  and investment related underlying detail  files) and simply plug them in place of the GDPNow forecasts for those values in the Inventories tab of the April 29 GDPNow spreadsheet , the model's CIPI contribution forecast goes up by 0.84 percentage points. Using the model's monthly inventories data-based forecast rather than averaging it with the quarterly-based model forecast, as described on page 17 here , further increases the CIPI contribution to 1.02 percentage points. The estimate cited in the Wall Street Journal  assesses the impact of the adjustment to the March inventories data on GDP growth to be 1.2 percentage points.

Real GDP growth was slightly negative despite the large CIPI contribution and a solid 3.0 percent real growth rate for aggregate personal consumption expenditures (PCE) and private fixed investment primarily because net exports subtracted 4.83 percentage points off first-quarter real GDP growth, as shown in the third bar in figure 2. BEA data show that nearly all of the reduction (4.79 percentage points) was accounted for by a 50.9 percent annualized surge in real goods imports that, apart from the pandemic related rebound in the third quarter of 2020, was the largest increase since 1972. In theory, this gain should be offset in the CIPI and/or other spending subcomponents of GDP. But the aforementioned BEA inventories adjustment shows that this cancellation may not be evident in the published monthly GDP source data.

It's unclear whether the reverse phenomenon—spending on goods drawn from inventories that are not accounted for in the published Census Bureau inventories data—can or will occur. But we can anticipate that it is likely that either the BEA's estimate of inventories contribution to first-quarter GDP growth will be revised down or GDPNow's projected contribution of it to second-quarter GDP growth will be revised down on June 27. Until June 27, GDPNow will make its own calculation of first-quarter CIPI in GDP using Census Bureau data on the book-value of inventories and BEA data for the remainder of the CIPI related data. This is because using Census Bureau book-value data usually generate virtually the same CIPI estimate for the prior quarter as what one would get using only the BEA data immediately after the GDP release. This allowed the model to anticipate BEA revisions of CIPI for the prior quarter in second and third release GDP estimates after Census Bureau revisions to monthly inventory book values. However, GDPNow currently calculates a first-quarter annualized CIPI of $94 billion in 2017 dollars, while the BEA calculated it as $140 billion. The published first-quarter CIPI will be revised two more times by the BEA on May 29 and June 26. After the latter revision, the BEA's first-quarter CIPI will be "final" until the annual revision that will take place on September 26 if the BEA follows the same annual revision scheduling pattern as it did in 2022–24. With respect to the July 30 2025:Q2 GDP release, 2025:Q1 CIPI will be "frozen" at the level published in the June 26 GDP (third) release estimate. So GDPNow would switch to the temporarily "frozen" BEA estimate on June 27 (see also the third-to-last FAQ here and/or more formal discussion on pages 13–17 here ). If both GDPNow and the BEA estimates for CIPI remain at their current, but different, estimates through June 26, the GDPNow switch to the higher 2025:Q1 value for CIPI would reduce its topline nowcast by 0.8 percentage points on June 27.

Figure 1 also shows that when we extend the gold-adjusted model forecasts from when it went "live" on March 6 back to February 26—the last forecast date before January international trade data with the surge in gold imports were released—the standard and gold-adjusted model forecasts are nearly identical in this final pre–gold import surge forecast. In the historical forecast accuracy analysis below, we use what was then the GDPNow growth forecasts for the first quarter of 2025 prior to February 26 and the gold-adjusted growth forecasts for the first quarter of 2025 beginning on this date.

Figure 2 shows Blue Chip and GDPNow forecasts for the composition of second quarter real GDP growth. The Blue Chip forecast includes all subcomponents shown in the chart except for "government plus residential investment." We essentially derive this as a residual using the topline GDP and other subcomponent forecasts along with some chain-weighting calculations. The current and now defunct GDPNow forecasts are also shown in the figure, the latter of which doesn't include the gold adjustments to the international trade data described here. The latter being stronger than the former implies that the model sees (net) gold imports falling this quarter relative to the first, which is plausible given the sharp decline in gold imports from January ($32.6 billion) to March ($17.5 billion).

Figure 2 also shows the consensus Blue Chip Economic Indicators (BCEI) forecast  for the second quarter of 2025 is weaker than the most recent GDPNow forecast as well as a typical quarter, proxied by the left-most two bars showing average annualized growth between the fourth quarter of 2007 and the fourth quarter of 2019 National Bureau of Economic Research business cycle peaks , and average growth since the latter peak. The subcomponent composition of the Blue Chip consensus forecast is also quite different than any of the others shown. It has both net exports and inventories partially reversing their first-quarter contributions but netting out to a positive, rather than a negative, total contribution. Its remaining subcomponent contributions are negative on balance and all somewhat weaker than their corresponding, "typical," first-quarter, and GDPNow-projected values. GDPNow was about as weak on May 1 as the Blue Chip forecast after it first incorporated April "soft data" in the ISM Manufacturing  and consumer attitudes reports. But it was revised up the next day after incorporating better April data on employment  and motor vehicle sales .

How might we assess the uncertainty around the Blue Chip and GDPNow forecasts? Prior to the pandemic, GDPNow projections two-and-a-half-months ahead were generally moderately less accurate than professional forecasts, but they were about on par with them within a month of the release. And since 2020, GDPNow has been less accurate on average than it was prior to the pandemic. In particular, using the gold-adjusted model for the first quarter of 2025, from the first quarter of 2021 to the first quarter of 2025, GDPNow's average absolute error for its final forecasts (0.65 percentage points, using a seasonally adjusted annual rate) was greater than it was in its prepandemic history (from the second quarter of 2014 to the fourth quarter of 2019) since it was first published (0.51 percentage points). Moreover, 80-day ahead prepandemic forecasts over this span were about as accurate, on average, as 20-day ahead forecasts starting in the first quarter of 2021.

The pandemic posed a number of forecasting challenges, including fiscal stimulus and supply chain disruptions, that made forecasting more difficult and probably less accurate than it was in the 2010s. The difficulty of forecasting also increases as the probability of a recession rises. The subdued Blue Chip forecast of GDP growth in the second quarter of 2025 is very similar to the consensus in the April Wall Street Journal Economic Forecasting Survey , and the respondents in that WSJ survey put the 12-month recession probability at 45 percent on average. Although lower than the 48 percent to 63 percent recession probability range prevailing in the survey during the second half of 2022 and all of 2023, this latest reading is well above its 17 percent average seen during 2014–19. All of this suggests that, in a probabilistic sense, we should expect forecasts in the second quarter of 2025 to be less accurate than they were in 2014–2019. For some additional (and more formal) statistical evidence, as seen in the chart in GDPNow's home page and here , the difference—or so-called "forecast disagreement"—between the average of the 10 highest Blue Chip forecasts and the average of the 10 lowest Blue Chip forecasts is much larger in the most recent BCEI survey than it was in late March. Research  by Constantin Bürgi and Tara Sinclair (you can also see it here ) shows that increased forecaster disagreement in two-and-a-half-month ahead GDP growth expectations are significantly associated with higher recession odds in a probit model. Further, a regression analysis I did with the historical BCEI data (excluding the first three quarters of 2020) implies that the current "Top 10-Bottom 10" BCEI forecast disagreement is associated with an 80-day ahead GDP growth nowcast that is 0.92 percentage points less accurate than it would be if that disagreement was instead at its 2014–19 average. With these differences in mind, figure 3 shows the average absolute forecast errors for the 80-day ahead GDPNow and BCEI real GDP growth forecasts and its associated subcomponent contributions beginning in the first quarter of 2021.

The Blue Chip forecasts for topline growth are a bit less accurate, on average, than GDPNow. On the other hand, they are a bit more accurate for their subcomponent contribution forecasts. Evidently, GDPNow had a more fortuitous cancellation of subcomponent errors than the BCEI did. Nevertheless, these accuracy metrics are of similar magnitude. Since the average absolute forecast error is equal to (√(2/π)) ≈ 0.8 standard deviations in a Gaussian distribution , the post-2020 period accuracy standard suggests that a negative GDP growth rate in the second quarter of 2025 is within the 70 percent confidence interval. But a fairly solid-to-strong growth rate is also within it. Figure 3 also shows that real PCE, CIPI, and net exports account for the largest, and similarly sized, forecast errors.

Figure 4 is comparable to figure 3 but uses shorter-term forecasts. The 20-day ahead Blue Chip forecast errors of real GDP growth are larger than those for GDPNow, but—as we noted before—the subcomponent contribution forecast errors are comparable. I should note that some luck is also evident when we compare GDPNow's 20-day ahead nowcast with its final forecasts, as the latter topline nowcasts are slightly less accurate than the former despite the improved accuracy of the final nowcasts for real PCE, CIPI, and net exports. For these shortest-horizon forecasts, the PCE contribution error metric is smaller than the same metrics for CIPI and net exports. Much of the final PCE forecast improvement relative to its 20-day ahead forecast is the result of incorporating a final retail sales release for the quarter 10 to 12 days before the GDP release.

It's an open question whether another judgmental BEA adjustment to CIPI, or perhaps one of the other subcomponents, could again factor into the forecast errors for GDP growth in the second quarter of 2025. But the above analysis and figures suggest a reasonable likelihood of being able to rule out either negative or relatively strong GDP growth this quarter as we approach the release date. The analysis also shows that net exports and CIPI are among the more difficult subcomponents to forecast—and that difficulty is present even without the added uncertainty surrounding tariffs.

By Pat Higgins, policy adviser and economist in the Atlanta Fed's Research Department

The Atlanta Fed's GDPNow model, which provides a running estimate of quarterly real GDP growth prior to official reports by the US Bureau of Economic Analysis (BEA) , is sometimes an outlier relative to professional forecasters. This has been true in spades since the last week of February 2025, as the chart on our GDPNow Home page shows. There are two reasons for this: the first is related to accounting for gold imports (which prior to March 6, 2025, we did not do); the second is due to the accounting of a handful of GDP subcomponents, for which GDPNow has a weaker outlook.

The BEA provides background on how they handle gold imports and exports in the GDP measures of imports and exports (here  and here ). And we have discussed what adjustments we have made to our model to better account for gold imports and exports (here  and here ). But it is worth looking more closely at some of these data, plotted in figure 1 below, since I (and perhaps others) had some early difficulties in spotting the surge in gold imports in the official data, a surge that has been reported in the press (here ).

The measure of nonmonetary gold imports, tracked by the orange line in figure 1 and whose monthly value has never exceeded $9 billion in its 35-year history and has never exceeded $4 billion since 2020, is reported in Exhibit 7 of each month's international trade report released jointly by the BEA  and the US Census Bureau  at 8:30 a.m. (ET) in a regular pattern of release dates . Not included in that report, but released at 10 a.m. by the BEA on these trade release dates , is a second "balance of payments" (BOP) measure of nonmonetary gold imports. That measure spiked to a monthly average of $29.1 billion for January and February of 2025 (the yellow line in figure 1).

Why the disparity in these measures of gold imports? As the BEA notes in an FAQ , the BOP measure of gold includes a portion of "finished metal shapes" imports (the green line in figure 1) that are reclassified by the BEA as gold. Recently, the most important of these commodities by far is identified as harmonized system code  7115900530: "Articles of precious metal, in rectangular shapes, 99.5 percent or more by weight of precious metal, not otherwise marked or decorated, of gold." In other words, gold bars. According to data from USA Trade Online , these imports surged from $2.08 billion in November of 2024 to $28.69 billion in January 2025 before falling back a bit to $22.96 billion in February 2025.

In the 8:30 a.m. (ET) international trade report, the finished metal shapes imports that include these gold bar imports are all classified on a Census basis within "industrial supplies and materials." These also spiked recently, as can be seen in figure 1 (the blue line). Industrial supplies and materials imports are of interest because they, along with five other sub-aggregates of goods exports and imports, are included in the Census Bureau's Advance Economic Indicators (AEI)  report often released about a week before the full international trade report.

Unfortunately, the AEI report does not separate gold bars or finished metal shapes from the remainder of the industrial supplies and materials imports aggregate, and this affects our methodology. Is there a way we can utilize the international trade data in the AEI until the full report is released? A number of analysts have noted Swiss gold exports to the United States have surged. According to Swiss-Impex  data and Federal Reserve Board exchange rates , Swiss gold exports to the United States spiked from under $400 million in each of the first two months of 2024 to $17.2 billion in January 2025 and $14.8 billion in February 2025. The reason  for the spike is likely due to the move of smaller gold bars, stored in London, to Switzerland, where they are refined into the larger gold bars acceptable in the New York market. These Swiss data are often released about a week to ten days before the AEI, and Comex data  on gold inventories are available even earlier. Nonetheless, because this data is not available from US government sources and the industrial supplies and materials trade data in the AEI is not further disaggregated, we discard that trade data when using the report to forecast net exports. We do, however, use the remaining trade data in that report as described here .

In terms of the quantitative significance for GDP growth, the January and February 2025 average (BOP) gold imports of $29.1 billion would, if maintained in March, put it $22.2 billion above its fourth quarter average. Annualizing this difference (by multiplying by 12) implies that a "straight" gold adjustment ($265.9 billion ≈ 12*$22.2 billion) would increase the GDP growth forecast relative to the standard GDPNow model by 3.6 percentage points. The difference between the gold-adjusted and standard model nowcasts is a little more than half this large, partly because the model is (implicitly) forecasting less of a gold impact in March and also because the model forecast for goods net exports doesn't put all of its weight on the portion of the model using the monthly foreign trade data (see page 11 of Higgins, 2014 ). With this (smaller) gold adjustment, the model is forecasting smaller (in magnitude), but still slightly negative, first-quarter real GDP growth (reflected in the right most bar and marked by the orange circle in figure 2 below).

In figure 2, the second bar from the right represents the GDP subcomponent contributions to the consensus April Blue Chip Economic Indicators  forecast, which is less than one percent but still in positive territory.

Compared to either the latest gold-adjusted GDPNow nowcast, or the one made at the same time as the Blue Chip survey, the largest subcomponent contribution difference between our model and Blue Chip forecasts is for net exports and the change in private inventories. A number of GDPNow users have asked whether the model is underestimating the inventories contribution, given the recent surge in (non-gold) goods imports. While the model does have at least advance Census Bureau inventory estimates  for many industries through February 2025, there are other industries (farm, nonmerchant wholesalers, construction, and utilities) not covered in these reports. They account for a bit more than 20 percent of the total inventory stock.

We don't have direct insight into what is happening with these "other industry" inventories, but using figure 3, we can look at whether the historical relationship between the quarterly growth rates of real goods GDP and manufacturing industrial production growth is consistent with their first quarter estimates. Although goods share of GDP is three times as large as manufacturing—value added is 29.9 percent versus 9.9 percent in 2024:Q4—we can see that there is a strong relationship between the two quarterly real growth rates of the two measures of output (with a correlation of 0.76 for 1972–2024). The region between the two dashed lines is the 95 percent confidence region where we would expect that proportion of the points to fall given the strength of the positive relationship of the two growth rates apparent in the plot.

The combination of the first-quarter growth rates of manufacturing IP and goods GDP is well below the 95 percent confidence band for the standard model and nearly on, but just above, the lower boundary for the gold adjusted one. One could interpret this as suggesting that even the gold adjusted model forecast is a little too pessimistic. But we'll have to wait until the official BEA report on April 30  to see if this is indeed the case.

We plan on releasing the gold-adjusted model's nowcast through April 29. Starting on April 30, what is now the gold-adjusted model will replace the standard model, as noted here . If gold imports were to fall and remain at $0 in April and thereafter, the standard model nowcast would likely understate second-quarter growth of imported goods and thereby overstate the GDP growth contribution of net exports. While this isn't likely to happen, it's also plausible that gold imports could meaningfully distort the (current) standard model's nowcast in the second quarter—hence, the switch.

By Pat Higgins, policy adviser and economist in the Atlanta Fed's Research Department

Many recent policy actions have aimed at a reduction in the size of the federal workforce, including a reduction-in-force, a hiring freeze, buyouts, and early retirement incentives. As the federal government is the largest single US employer—an estimated over 8 million employees are either employed directly by the federal government or are supported by federal contracts and grants—these policies could have important implications for the labor market.¹ In this blog post, I first document the size and characteristics of the federal workforce in September 2024, before these policies were enacted, and then explore actual and potential reductions in the full federally supported workforce. I estimate that the potential reduction in the full federal workforce, including contract and grant employees, could be as high as 1.2 million.

Federal employment
Over the last 25 years, the number of employees directly employed by the federal government has ranged from a low of 1.8 million during the Clinton administration to more than 2.4 million at the end of 2024 (see figure 1). The federal share of total payrolls has hovered around 1.5 percent of payrolls since 1999, increasing during downturns and falling during expansionary periods.

These federal employees are not just located in the capital area. There are federal employees in every state, as shown in figure 2. On average, federal employees accounted for 1.6 percent of state payrolls in September 2024. The largest share of state employment was in the District of Columbia, where more than 20 percent of employees were federal employees. This is followed by over 5 percent of payrolls in Maryland and almost 4 percent in Hawaii. The presence of federal employment was also strong in the Sixth District, with federal employees accounting for 1.9 percent of state payrolls in Alabama,1.7 percent in Mississippi, and 1.63 percent in Georgia (the latter largely due to the Centers for Disease Control and Prevention in Atlanta).

Contract and grant employees
However, the numbers presented in figures 1 and 2 represent only a fraction of the total number of employees whose work the federal government supports, as federal government contracts and grants (C/G) support a substantial workforce. Although precise numbers don't exist, Light (2019) provides estimates on the total "Federal Industrial Workforce" (FIW), which includes federal employees as well as employees supported by federal grants and contracts, over time.² Light reviewed all federal contracts and grants to estimate the total number of full-time equivalent employees (FTEs) directly employed using federal funds. In addition, Light also calculated the number of FTEs who were required to provide inputs into these grants and contracts (indirect employees).³

Figure 3 depicts the total FIW, reflecting the estimates provided by Light (2019). The blue line shows the number of federal employees reported by the Office of Personnel Management  (OPM) over the period 1999–2024, along with estimates through February 2025 based on the growth in federal employment from the payroll data. Light's estimates of the contract FTEs (green line) and grant FTEs (yellow line) for 1999, 2002, 2005, 2010, 2015, and 2017 are also included, with a linear interpolation between the years. In 2017, for every federal employee, there were 2.6 C/G FTEs. Unfortunately, there is no new information on this ratio since 2017. The dashed portions of the C/G lines estimate C/G FTEs holding constant the ratio of contract or grant workers to federal workers at the 2017 level. Thus, any movement in the C/G lines after 2017 are due to the changing size of the federal workforce. Under this assumption, the size of the FIW in February 2025 is estimated to be approximately 8.3 million FTEs.⁴

Estimated employment loss in context
Given the size of the federal workforce, the reduction in employment levels could have a measurable impact on labor market outcomes. The website layoffs.fyi  estimates that, as April 1, 2025, there have been 136,621 total exits from the federal workforce along with an additional 170,214 planned reductions. This represents more than 12 percent of the federal workforce. Importantly, this estimate of exits does not include C/G employees who have lost their jobs due to the numerous cancellations and modifications of grants and contracts. For example, the Environmental Protection Agency recently announced the cancellation of $1.7 billion in contracts , and the Department of Education announced the cancellation of more than $1 billion in contracts and grants.⁵ Furthermore, the National Institute of Health (NIH) announced plans to cut indirect costs in grants and contracts from 30 percent to 15 percent, leading to hiring freezes at many of the nation's research universities. It seems reasonable to expect a decline in the level of C/G FTEs as well.

If the impact of these C/G reductions is proportional to the impact on federal employment—that is, if the ratio of C/G workers to federal employees remains constant—then we would expect an additional loss of approximately 900,000 FTEs, for a total of 1.2 million separations from the FIW. To put this into perspective, this total equals 2 percent of the total separations from employment reported in the Job Opening and Labor Turnover Survey  for 2024.

Will these separations from the FIW show up immediately in labor force statistics? Not likely. Although the current hiring freeze could affect employment levels, many federal employees are currently on administrative leave, and many others also selected deferred retirements. It is expected that these reductions will begin trickling in this summer.

There are no prior examples to assist in estimating the impact of such a large-scale reduction on the labor market. The Clinton administration also oversaw large cuts in federal employment, with reductions of more than 400,000 between 1993 and 2001, as part of the "Reinventing Government" initiative (as seen in figure 1). However, those cuts took place over seven years, which dampened the impact on the labor market for any given point in time. Furthermore, the largest single mass layoff event in the private sector occurred in 1993 when IBM laid off 60,000 employees .

Given the uncertainty on the full size and the timing of the reduction, predicting the impact on the unemployment rate is difficult. It is unclear how many of those separated from the FIW will exit the labor market and what share of those who remain will be able to find new employment. A confounding issue is that the federal workforce is older and more educated than the workforce as a whole. In general, according to a 2024 Department of Labor survey of workers  displaced between 2021 to 2023, older workers are more likely to exit the labor force. This could suggest there will be a higher rate of exit from the labor force than during other mass layoffs, thus minimizing the impact on the unemployment rate. However, the report also indicates that displaced government workers  had the lowest exit rate from the labor force at 7.4 percent. Another complication is that, for those who remain in the labor force, older  and more educated  workers generally take longer to find new employment. Fortunately, states in the Washington metro area, which account for a large share of federal employees, have already implemented increased efforts to support separated federal workers, which might help dampen the impact.

By M. Melinda Pitts, research center director, Center for Human Capital Studies in the Atlanta Fed's Research Department

In the most recent CFO Survey , we explored the extent to which firms' inputs/supplies are imported from countries facing recently imposed tariffs and how this exposure affects firms' expectations. (We should note that our analysis exclusively focuses on firms' exposure to China, Canada, and Mexico because tariffs had been announced and implemented for these specific countries in the period leading up to and during the survey fielding period. As a result, although the survey question also asks for the percent of inputs/supplies that firms sourced from "Other" countries, we leave this category aside for the present analysis.) Figure 1 shows that 53 percent of survey respondents reported sourcing some or most inputs/supplies for their firm's US operations from China, Canada, and/or Mexico (that is, tariffed countries). In the remainder of this post, we will examine how the expectations of firms that sourced their inputs/supplies from these three countries differ from those that did not, and whether the degree of reliance on these countries coincides with differences in firms' economic outlook and performance expectations.

Dimmer economic outlook among firms exposed to tariff-hit countries
After firms broadly increased their economic and own-firm optimism and their GDP growth expectations last quarter , in the first quarter of 2025 we observed a stark dichotomy between firms that relied on the countries hit hardest by tariffs for their inputs/supplies and firms that did not. Firms that sourced inputs/supplies from China, Canada, or Mexico were 6 to 8 points less optimistic about the US economy and 8 points less optimistic about their own firm's prospects relative to their peers that did not import from these countries (see figure 2). Similarly, firms that sourced inputs/supplies from these countries expected real GDP growth in the next year to be 0.6 to 0.8 percentage points lower than their nonexposed peers (see figure 3). In both cases, whether or not firms sourced inputs/supplies from these countries—rather than the degree of reliance on these countries—appeared to account for much of the divergence.

Firms exposed to tariff-hit countries expect meaningful impact on performance
In addition to downgrading their economic outlooks, firms that sourced inputs/supplies from tariff-hit countries expected worse performance than their peers across most key metrics. Figures 4 and 5 demonstrate that the more firms imported inputs/supplies from tariff-affected countries, the higher their expected price and unit cost growth for 2025.

Looking ahead to 2026, firms that imported inputs/supplies from tariff-hit countries expected their unit cost expectations to remain elevated, but price growth is expected to be roughly similar across all firms. Respondents offered several potential explanations for this, including that they may attempt to offset higher unit costs by margin compression. As one firm noted, "tariffs will impact customers and expected inflation will create margin pressures."

Firms that sourced inputs/supplies from tariff-hit countries also expected lower revenue and employment growth this year, with expected revenue growth declining as firms' reliance on tariff-hit countries increased (see figures 6 and 7). The divergence is expected to continue next year, though to a lesser degree. One respondent explained that with margins already compressed in their industry, revenue and employment might suffer because "potential tariffs [have] my industry worried about price escalation leading to cancelled projects."

Impact of tariffs similar on a matched-sample basis
Finally, we examine changes in optimism; expectations for real GDP growth; and expected revenue, employment, cost, and price growth for the 282 firms that responded to our surveys in both the fourth quarter of 2024 and the first quarter of 2025. This analysis allows us to minimize the impacts of differences in sample composition between the two surveys. Figures 8 and 9 illustrate that compared to their peers, firms that relied on tariff-hit countries to source their inputs lowered their optimism (economy-wide and own firm) and their expected GDP growth.

Furthermore, figure 10 reveals sizeable differences in expected firm performance depending on whether firms sourced inputs/supplies from tariff-hit countries. Expectations for price, unit cost, and revenue growth in 2025 were magnified for those firms sourcing higher shares of their inputs from tariff-hit countries. These results suggest that tariff-related concerns are likely the cause of the divergence in firms' performance expectations.

Conclusion
Just over half of the firms in our sample sourced inputs/supplies from China, Canada, and/or Mexico and are thus directly exposed to the tariffs recently imposed on these countries. Relative to their nonexposed peers, these firms became less optimistic, expected lower revenue and employment growth, and expected higher price and unit cost growth—with only price growth expected to stabilize by 2026. In addition to the sizeable impact on firms' prospects, financial decision-makers have also highlighted considerable uncertainty surrounding the breadth and duration of tariffs and the attendant consequences for their firms. We will look to future surveys to inform how financial decision-makers' views evolve on the expected impact of tariffs on their firms and the US economy.

By Brent Meyer, assistant vice president and economist,

Lauren Taylor, economic research analyst, and

Daniel Weitz, survey director, all of the Atlanta Fed's Research Department