�۶���Ƶ

The US government does not have unlimited capacity for debt. The recent rise in 30-year Treasury bond yields above 5%, despite expectations for slower economic growth, highlights growing investor concern about the persistently deteriorating fiscal situation. Rating agency downgrades merely confirm what is already widely recognized, yet these concerns have not prompted legislative changes to improve the fiscal outlook. The current reconciliation package will further expand deficits, and tariff revenues will not be sufficient to put the US government back on a sounder fiscal footing. Lawmakers are assuming that lower taxes for businesses and individuals will generate enough economic growth to improve the deficit-to-GDP ratio, but a recession would severely undermine long-term fiscal sustainability.

Reconciliation package

In our most recent fiscal update in December 2024, we projected a two-bill approach: an initial, smaller spending package to fund border security and defense, followed later in the year by a separate bill to extend the 2017 Tax Cuts and Jobs Act (TCJA). Instead, Congress and the administration have moved rapidly to advance the “One, Big, Beautiful Bill” that combines both priorities, a debt limit increase, and much more. The bill, which passed the House 215-214, includes a wide range of major spending and revenue changes for 2025-34 that would increase the deficit by more than USD 3 trillion over the next decade, according to an by the Committee for a Responsible Federal Budget (see Fig. 2). This bill remains subject to further changes as it progresses to the Senate.

The bill has advanced rapidly, but many of the long-anticipated and delayed debates within the Republican Party over its size and structure are now coming to a head. Republicans will need to reach a compromise, as passing the bill requires near-unanimous GOP support given razor-thin majorities in both the Senate and the House. Lawmakers are aiming to have the legislation on the president’s desk by 4 July, but this deadline could slip to late July or early August when Congress must raise the debt ceiling before the Treasury exhausts its extraordinary measures to fund the government. If the bill is not passed in time, Republicans will likely be forced to abandon the one-bill strategy in favor of a stop-gap measure to separately lift the debt limit.

Fiscal conservatives—encouraged by Moody’s mid-May downgrade of the US debt rating to Aa1, stable—are demanding deeper spending cuts, while blue-state House Republicans have pushed to increase the cap on the State and Local Tax (SALT) deduction. When the Senate takes up the bill, it will likely undergo further changes, requiring yet another vote in the House before it is finally sent to the president. Senate Republicans have expressed concern that the cuts to social programs are too deep, as are some additional new business and personal income tax cuts.

The tax cuts are front-loaded, as many of the new individual and business tax reductions in the House bill are set to expire midway through the 10-year window to limit the plan’s cost. Meanwhile, the savings are back-end loaded, building toward the middle of the window as cuts to social programs are phased in. If the new tax cuts are extended or made permanent in future legislation, as is often the case given the political unpopularity of tax increases, the total cost over the next decade would rise by an additional USD 1.5tr, excluding interest costs (see Fig. 3). Overall, the House bill is mildly expansionary. If it passes largely intact, federal government deficits would increase from an already elevated 6.3% of GDP in 2024 to more than 7% in both 2026 and 2027.

Tariff revenues

Higher tariff revenues will not materially reduce the deficit when other factors are driving it higher. While tariff income serves as an offset to the increase in deficits from the reconciliation package, the latest round of "reciprocal" and baseline tariffs is not a reliable long-term revenue source, given their questioned legal basis and the discretionary nature of executive orders. Legislated tariffs would provide more predictability and could be incorporated into the budget process, but they remain politically unpopular. That said, a 15% effective tariff rate could generate an additional USD 300-450 billion in annual revenue over the next decade, depending on the impact on import volumes and the need to support sectors affected by retaliation. This would represent 1-1.5% of 2024 US GDP. However, higher tariffs do not necessarily result in higher revenue. The Peterson Institute for International Economics that the economic damage caused by higher effective tariff rates can actually reduce potential revenue.

2026 budget

Bipartisan cooperation will soon be required to pass either the 2026 federal budget or a series of continuing resolutions to fund the government and prevent a shutdown. Unlike reconciliation bills, budgets need 60 votes in the Senate to overcome a filibuster. In 2025, the government relied on a series of continuing resolutions to maintain funding. The most recent resolution, approved on 14 March, expires on 30 September, by which time Congress must complete and pass the annual appropriations bills. President Trump’s budget request, submitted to Congress in early May, set the stage for significant spending cuts across a range of agencies. Non-defense discretionary spending would be reduced by 23%—to levels last seen in 2017—to fund increases in defense and border security. However, we see little likelihood that Congress will approve cuts of this magnitude.

When Moody's downgraded the US to Aa1, stable, from Aaa, negative on 16 May, it chose to act based on its expectations for the upcoming budget, rather than wait for its finalization. The agency projects that mandatory spending will rise to 78% of total spending by 2035, up from about 73% in 2024. Moody’s also expects a full extension of the 2017 Tax Cuts and Jobs Act to add USD 4tr to the primary deficit over the next decade. As a result, the federal debt burden is forecast to increase to 134% of GDP by 2035, from 98% in 2024. For context, the general government debt ratio—which includes state and municipal debt and is the metric most other countries use as their official debt ratio—stood at 121% in 2024 (see Fig. 4).

Higher funding costs are a key challenge

When assessing debt affordability, the net cost of debt relative to government revenues is more important than the debt ratio. We estimate this ratio at 12.2% for 2024 at the general government level (see Fig. 5). For the federal government alone, it was 18% last year, and Moody’s projects that, if current fiscal policies persist, it could rise to 30% by 2035.

When thinking about the outlook for funding costs, some forces are pulling them lower, but most are pushing them higher. On net, we think debt service costs will rise further as a share of government revenues.

The average coupon of US Treasuries stands at 2.99%, while new Treasury notes and bonds issued this year carry a weighted average coupon of 4.18% (it was 4.20% in 2024). The average monthly Treasury issuance this year is USD 2.5tr, with the 10-year+ bucket representing 18% of total monthly issuance (see Fig. 6). The weighted average tenor at issuance this year was 7.5 years, up slightly from 7.4 years in 2024. Further debt refinancing will likely continue to add to interest expenses as bonds mature. Excluding bills, the US will need to refinance USD 4.4tr of bonds with a weighted average coupon of 2.74% through the end of 2026 (see Fig. 7). Applying the current weighted average yield of Treasuries of 4.30%, the debt rollover would add USD 69bn in interest costs.

Another factor pushing up debt service costs is the compensation investors received for holding Treasury Inflation Protected Securities (TIPS) when inflation was high. Depending on the future path of inflation, the USD 2tr in TIPS could also keep adding to debt service costs. After a decline recently, this cost may rise again depending on the ultimate impact of tariffs on consumer price inflation. For example, should the tariffs add one percentage point to the US CPI, the immediate additional cost of compensating holders of TIPS would be USD 20bn.

The USD 6.5tr of bills that are being rolled over frequently should incur a lower cost during this year compared to 2024, given the decline in short-end yields. As an indication, average three-month yields were 5.1% during last year and have averaged 4.3% year-to-date—assuming an average for the year of 4.1% would translate into a reduction in debt cost of USD 65bn.

Altogether, we expect the rise of the net cost of debt-to-government-revenues ratio to slow in 2025, in particular as short-term yields have declined, somewhat narrowing the yield gap between maturing and new bonds. A higher net cost of debt will result from financing of the budget deficit and from lower Treasury holdings by the Fed, which pays out interest earned as a dividend to the Treasury (i.e., there is no net interest cost for government bonds held by the central bank).

We estimate the net cost of debt-to-general-government revenues will be 12.7% this year, a doubling from 2020 levels (see Fig. 5). In our base case, we expect the Fed to cut interest rates toward 3.00-3.25% by mid-2026 and yields across the curve to decline moderately, which should help to contain the debt service cost to 13.5% of general government revenues by 2026. If instead Treasury yields were to rise across the curve toward a weighted average of around 5% and stay there, the net interest cost would rise by almost one percentage point of revenues every year, reaching about 17.2% by 2030.

On this important debt affordability measure, the US already has the highest level of debt service among developed countries, posing a meaningful fiscal burden that policy cannot immediately influence.

How high can US government debt rise?

There is no strict limit on the amount of debt a government can carry, but the true “referee” on debt sustainability is the collective group of creditors who determine the government’s effective cost of debt. As a country's debt burden grows, creditors may reduce their exposure or demand higher compensation. In turn, rising debt costs can consume a larger share of future revenues, which can be offset by spending cuts or increased revenues (also known as “fiscal austerity”) or by incurring more debt to pay interest (what Hyman Minsky called “Ponzi finance”). This latter strategy often leads to a debt crisis, where debt levels and costs both spiral upward.

US exceptionalism, which stems partly from its status as the world’s leading economy and provider of the world's principal reserve currency, has afforded the US government some added scope to run high budget deficits and levels of debt at a highly affordable cost. But at the same time, no country has an unlimited capacity to carry debt, and past rating downgrades to Aa1/AA+ indicate that the US has exhausted its extra debt capacity compared to other AAA-rated countries, which carry far lower debt ratios. For example, France, with a debt-to-GDP ratio 10 percentage points below the US and a net cost of debt of only 3.5% of revenues, saw Standard & Poor’s and Moody’s downgrade its debt one notch to AA- and Aa3, respectively, in 2024 (see Fig. 9). Further fiscal deterioration in the US could lead to more negative rating actions, albeit slowly, with one-notch cuts over several years.

The reconciliation bill misses an opportunity to get the US on firmer fiscal footing and instead offers a range of temporary new tax exemptions that could grow the debt even further if extended. At some point, the US will need to reduce its deficit to a more sustainable level—whether by crisis, by choice, or by good fortune. There are several paths to reducing high government debt burdens:

Mechanisms of financial repression

Financial repression can be achieved through legislation and regulation, monetary policy, and soft pressure on large bond investors (such as foreign central banks). The line between standard, prudent policy, and financial repression is often blurred.

What would financial repression mean for investors?

The primary effect of successful financial repression is that government bond interest rates are kept lower than they would otherwise be. To reduce the government debt-to-GDP ratio, the growth in the debt stock must be slower than nominal GDP growth. The higher the fiscal deficit, the more interest costs need to be suppressed relative to nominal growth rates to achieve this. Typically, as long as debt grows faster than nominal GDP, negative real yields (nominal bond yields less inflation) are required to facilitate a reduction in the debt ratio (see Fig. 10). Of course, the debt ratio could still rise under financial repression, but it would do so more slowly than without such measures.

Crucially, meaningful reductions in deficits and the debt ratio through financial repression require the presence of inflation. By contrast, while financial repression was widespread during the Great Depression, deflation actually led to a rising debt-to-GDP ratio.

From an investment perspective, we distinguish two stages of financial repression:

1. The introduction of measures to lower the cost of debt.
2. The period during which these measures are maintained.

Conclusions

For a country as large and wealthy as the US, widespread financial repression seems feasible and could enable the government to continue financing a growing debt burden without materially increasing its risk of default.

Financial repression policies could be deployed temporarily to provide fiscal breathing room, allowing for budget consolidation and improvement, followed by a phase-out and a return to more conventional policy settings. In such a scenario, economic distortions should remain temporary and manageable.

However, sustained financial repression facilitated by the central bank would heighten the risk of de-anchored inflationary expectations and currency depreciation. Over time, this could necessitate aggressive interest rate hikes—potentially triggering an economic contraction—and the maintenance of restrictive monetary policy until inflation expectations are re-anchored. There are numerous examples from emerging markets of efforts to rein in inflationary spirals, and the US itself experienced a high-inflation episode in the 1970s, which was ultimately brought under control by Fed Chair Paul Volcker in the early 1980s.

Assets that may offer some protection against a prolonged period of financial repression and thus can help investors to preserve the purchasing power of their wealth include gold, inflation-linked bonds (only initially, until real yields turn negative), real estate, and segments of the equity market that are less exposed to currency swings and possess the pricing power to pass elevated inflation rates on to customers.