U.S. Treasury Yields Under a Simultaneous Federal Reserve and ECB Tightening Shock

1. The Hypothetical Monetary Shock

Assumptions. The Fed raises the upper bound of its range by 525bp from 3.75% to 9.00%; the ECB raises its refinancing rate by 450bp from 2.15% (deposit rate 2.00%) to 6.65% (deposit rate 6.50%). Both remain restrictive together, so the entire developed-market risk-free complex reprices. Since May the Fed has raised its range by 25bp (to 3.75–4.00%) and the ECB its deposit rate by 50bp (to 2.50%, refinancing rate 2.65%), so the remaining distance from today is 500bp for the Fed and 400bp for the ECB. The scenario assumes a restrictive regime lasting long enough to be priced, and that inflation falls only after a major contraction in demand and credit.

2. Why May 2026 Is the Correct Reference Point

The ECB shock starts from its May rate, so the Treasury curve is measured from the same moment: 29 May 2026, the last trading day of the month. The data are the Treasury’s official daily par yield curve. The 20-year point is a par yield read from the Treasury’s curve model, which is fitted to on-the-run securities including the 20-year bond; it is not a separately traded constant-maturity series, and we do not construct any additional data. Table 1 shows the official curve in May and on 7 October.

Two observations. The long end was already near 5% in May, so the scenario does not start from the low-yield era; the 29 May close sits inside the month’s range (10-year 4.36–4.67%, 20- and 30-year about 4.9–5.2%). And since May the curve has flattened at the long end (30Y–10Y from +54bp to +36bp on 9 October), consistent with a policy-led repricing. The 20-year and 30-year are almost equal (−1bp in May, +5bp on 9 October), a structural feature we use below. The 10-year closed at 5.31% on 5 October, its highest close of 2026 in the Treasury series, and touched about 5.36% intraday on 7–8 October, the highest since 2002 (CNBC).

3. The US Macroeconomic Starting Point

Official data. Real GDP rose at a 2.2% annualised rate in the second quarter (final estimate, revised up from 1.5%). Unemployment was 4.2% in September, with payrolls up only 29,000 (BLS). CPI inflation was 3.4% in August, with core at 2.4% and energy up 16.3%; the Fed’s preferred measure, PCE, rose 3.4%, with core PCE at 3.0% (BEA). The Fed’s September projections have PCE inflation at 3.7% in 2026 and 2.3% in 2027, core PCE at 3.4% and 2.5%, growth of 2.3% and a funds rate of 4.1% at end-2026.

Market pricing. The 10-year breakeven was 2.36% on 2 October (FRED), so the implied real yield is about 2.9% (LBMR calculation). Consumers expect 3.9% inflation over a year and 3.0% over five years (University of Michigan, via Trading Economics). The market prices a further Fed hike around December (as reported).

LBMR interpretation. The economy is strong, so it can absorb higher rates for longer, and a 9% policy rate would probably be the response to persistent inflation, not to a recession. For long Treasuries inflation matters more than growth: a funds rate of 9% reached because of persistent inflation would push up real yields and the term premium, while the same rate reached with rapid disinflation would let the long end peak below it.

4. Fiscal Position and Debt Dynamics

Official data. CBO projects a fiscal 2026 deficit of $1.9 trillion (5.8% of GDP), debt held by the public of 101% of GDP rising to 120% in 2036, and net interest rising from 3.3% of GDP (about $1 trillion) to 4.6% and more than $2 trillion by 2036. Its monthly review put the deficit at $2.0 trillion after eleven months, and gross federal debt has passed $40 trillion. The primary deficit is therefore about 2.5% of GDP (5.8% less 3.3%, our calculation). Fitch estimates the general-government deficit at 7.4% of GDP in 2026.

Maturity structure. The average interest rate on marketable debt was 3.475% in August, the average maturity is 71 months and about 33% of marketable debt matures within twelve months (Treasury). Bills were $7.25 trillion, or 22% of public debt. The effective rate therefore lags market yields, and only maturing debt and new issuance reprice.

The r−g mechanism. The change in the debt ratio is roughly (r−g) times the debt ratio minus the primary balance. Today r (about 3.5%) is below nominal growth (real growth of about 2% plus inflation of about 3.4%), which is why the ratio is near stable despite a primary deficit. In the scenario new debt is priced at 8–9% while recession cuts nominal growth, so the marginal r−g could become positive and large. The debt-stabilising primary balance is about the debt ratio times (r−g): roughly +1.0% of GDP at r−g of 1pp, +2.0% at 2pp and +4.0% at 4pp, against a deficit of about 2.5% today, requiring swings of 3.5pp to 6.5pp of GDP in the primary balance.

Mechanical illustrations (not forecasts). If the one-third of marketable debt that matures within a year repriced 400bp higher, the average rate would rise about 1.3pp and annual interest about 1.3% of GDP in the first year. Bills alone, $7.25 trillion at a funds rate of 9% against about 3.7% today, would add roughly $380 billion a year, or 1.2% of GDP (assuming GDP near $33 trillion). Full pass-through of +400bp on a debt ratio of 101% would add about 4% of GDP, taking net interest above 7% of GDP. The United States does not become insolvent: it issues in its own currency. The issue is fiscal sustainability, inflation credibility, the term premium and financing conditions.

5. External Position, Ratings and Reserve-Currency Status

Official data. The current-account deficit was $246.0 billion in the second quarter (3.0% of GDP) and the net international investment position −$22.42 trillion, with assets of $46.97 trillion and liabilities of $69.39 trillion (BEA, 24 September); on our estimate that is about two-thirds of GDP. BEA’s June annual update revised the end-2025 position from −$27.54 trillion to −$21.87 trillion. In 2025 the goods and services deficit was $901.5 billion, with a goods deficit of $1,240.9 billion and a services surplus of $339.5 billion (BEA and Census). Foreign holdings of Treasuries were $9.25 trillion in July, of which official $3.77 trillion; Japan held $1.10 trillion and China $618 billion (TIC). The dollar is 56.7% of allocated reserves (IMF COFER, Q2). Ratings are AA+ stable (S&P, affirmed 26 June; Fitch, affirmed August), Aa1 stable (Moody’s, since May 2025) and AA− stable (Scope).

LBMR interpretation. A large negative investment position does not mean for the United States what it means for a smaller emerging economy: the country issues the dominant reserve currency, the largest sovereign market and the main global collateral asset, which acts as a buffer in global stress. But reserve status does not make Treasury yields immune to inflation, fiscal deterioration, duration risk, supply or falling foreign demand; it lowers the probability of a currency-crisis path, not of a term-premium shock. A downgrade does not mechanically add a number of basis points; the channel runs through fiscal credibility, investor risk perception, the term premium and yields.

6. Decomposing the Treasury Yield

Conceptually, a Treasury yield is the average expected short rate over its life, plus inflation compensation and a real-rate component, plus a term premium and fiscal and issuance effects. The San Francisco Fed’s model puts the 10-year term premium at 1.49% on 5 October against 1.28% at the 16 September FOMC and 1.16% a year earlier, with an average expected short rate of 3.82% on a zero-coupon 10-year yield of 5.31%. Kim–Wright was 1.07% on 9 October (FRED) and the NY Fed’s ACM model 0.76% in August (monthly average), so current estimates range from about 0.8% to 1.5%, a model-dependent spread similar to that of 2023. The 10-year is anchored by the expected policy path over the next decade. The 20-year adds term premium and fiscal and duration risk. The 30-year is the most exposed to long-run inflation, fiscal credibility, supply and term premium.

7. Transmission: The Fed, the ECB, the Dollar and the Balance Sheet

The Fed channel. A funds rate of 9% passes first to money-market rates and Treasury yields, then with lags to mortgages, corporate borrowing, bank lending and financial conditions, and finally to housing, investment, consumption, unemployment and inflation. Market rates move at once; the effective cost of federal debt follows over a period of years. The 30-year mortgage rate was 7.40% on 8 October, the highest since November 2023 and up from 6.30% a year earlier (Freddie Mac).

The ECB channel. An ECB at 6.65% would lift Bund yields (the 10-year stood at about 3.5% on 7 October against 2.97% in May), euro-area sovereign yields and the global term premium. Two effects pull in opposite directions. Higher European yields make European bonds relatively more attractive and could reduce marginal demand for Treasuries, while a global risk-off shock could increase demand for Treasuries as the ultimate safe asset. The net effect is ambiguous and scenario-dependent: Scenario A favours the safe-haven effect, Scenario C the relative-yield effect.

The dollar. Two configurations are possible. In the safe-haven configuration, global tightening raises risk aversion, the dollar appreciates and Treasury demand increases; this fits Scenario A. In the fiscal–inflation configuration, high inflation, fiscal deterioration and falling confidence weaken the dollar, foreign demand drops and the term premium rises; this fits Scenario C. Today the dollar is strengthening with yields (DXY 101.9, up about 4% in a year), consistent with the first.

The Fed balance sheet. Quantitative tightening ended on 1 December 2025 and the Fed has since bought Treasury bills for reserve management; its Treasury holdings were $4.46 trillion on 29 May 2026. A restart of runoff in the scenario would shift duration to private investors, who would require more compensation. We do not assume QT produces a specific yield rise; its channel is private-sector absorption and the term premium.

8. Treasury Supply, Auctions and Foreign Demand

Treasury kept coupon auction sizes unchanged at its August refunding and at least doubled long-end buyback limits to $4 billion from 9 September through 4 November; the next refunding statement on 4 November is a key signal. The 30-year auction on 8 October cleared $22 billion at 5.618% with a bid-to-cover of 2.54 (as reported). Foreign holdings of Treasuries are stable at about $9.25 trillion, but official holdings fell from $4.01 trillion in February to $3.77 trillion (Japan −$135 billion, China −$76 billion), partly from valuation. Foreign demand matters, but the market is deep and domestic institutions are essential. The long end could reprice more if investors require extra compensation to absorb duration.

9. The 10-, 20- and 30-Year Treasury

10-year (May 4.45%). We triangulate three methods. A ratio method: at the 2023 peak the 10-year yield was about 0.93–0.95 times the effective funds rate (about 5.0% against 5.33%), which gives about 8.3–8.4% at an effective rate near 8.85%. A decomposition method: scaling the Federal Reserve Board’s 2023 model and survey splits to a 9% funds rate gives 7.1–8.1% (Table 4). And a curve method: in May the 10-year was about 82bp above the funds-rate midpoint, but at a 9% peak the curve would more likely be inverted, with the 10-year below the funds rate. We use 8.25%, a rise of 380bp or 0.72 of the Fed shock, at the top of the decomposition range because the fiscal and supply backdrop is weaker than in 2023. Lower-stress: 7.0%; severe: 10.0%; the full severe range is 9.0–11.0%. The estimate would be invalidated by rapid disinflation or a deep recession bringing early cuts.

20-year (May 4.98%). More exposed to term premium, fiscal and duration risk. Its slope over the 10-year was +53bp in May and +43bp now. We assume a central slope of about +50bp in Scenario B, giving about 8.75% (+377bp). The 20-year trades near the 30-year (a quirk of the Treasury curve and supply), so we keep the two within 15bp.

30-year (May 4.99%). The most exposed to long-run inflation, fiscal credibility and supply. We lower the working estimate of 9.00% to 8.75% because a 30-year yield above the terminal funds rate would require a steeper 10s30s curve (about +100bp) than May’s +54bp, and at the peak of policy tightening the curve tends to flatten, as in 2022–23. The terminal rate is not a ceiling: in Scenario C, with persistent inflation and falling foreign demand, the 30-year could trade above 9%, and the 30-year could trade below it if recession dominates.

10. Three Scenarios and Central Estimates

Scenario A: successful disinflation. The Fed reaches 9%, inflation falls, markets price cuts, fiscal deterioration is manageable and Treasury demand stays strong; the 10-year peaks well below the funds rate.

Scenario B: tightening plus fiscal deterioration (central stress case). The Fed reaches 9%, inflation stays elevated, deficits remain large, issuance is heavy, the term premium rises and foreign demand weakens at the margin.

Scenario C: fiscal–inflation–duration feedback loop (extreme tail). Persistent inflation, weak growth, large deficits, rapidly rising interest costs, heavy issuance, falling foreign demand and credibility concerns reinforce one another. Even here the stress is not default but a term-premium regime.

Calibration of the decomposition. The split of a long yield between the expected policy path and the term premium is not observable, and models disagree. The Federal Reserve Board’s own decomposition of the 10-year yield at its peak on 19 October 2023 (zero-coupon yield about 5.05%, effective funds rate about 5.33%) gives an expected average policy rate of 3.1–4.6% and a term premium of 0.4–2.0%, depending on the model or survey. Scaling each decomposition to an effective funds rate of about 8.85% (a 9.00% upper bound) gives a 10-year yield of 7.1–8.1%, with a mean near 7.6%. The observable ratio of the 10-year yield to the funds rate at that peak, about 0.93–0.95, gives 8.3–8.4%. Our central case of 8.25% therefore sits at the top of the 2023-based range with a modest uplift for fiscal and supply risk; Scenario A (7.0%) sits at its bottom; and Scenario C (10.0%) lies about 2pp above anything the calibration produces, which is why it is a tail.

Testing the working estimates. We raise the 10-year from 8.00% to 8.25% because the 2023-based calibration (Table 4) gives 7.1–8.1% on decompositions and 8.3–8.4% on the ratio method, and because today’s term premium is already above 2023’s model estimates. We keep the 20-year at 8.75% and lower the 30-year from 9.00% to 8.75% because the 20- and 30-year are near-equal and a steeper curve than May’s would require a credibility shock, which is Scenario C. Changes are +380bp, +377bp and +376bp. The curve stays about as steep as in May (+50bp): the scenario is a parallel-ish bear shift, not an extreme bear-steepening.

11. Default Risk versus Duration Risk

The Treasury market can suffer a severe sell-off without any investor believing the government will default. The stress arises from inflation risk, higher real yields, a higher term premium, fiscal risk and supply, which together raise nominal yields. This is why reserve-currency status is a buffer against crisis but not against losses.

12. Financial Stability, Housing and Corporate Funding

Banks and funds. US banks had $326.7 billion of unrealised securities losses at end-June (5.5% of amortised cost, down from 6.8% a year earlier; FDIC), with return on assets of 1.37%. Another 380bp on the 10-year would add large losses and reduce collateral values for banks, insurers, pension funds, money-market funds, asset managers and leveraged investors. A Treasury shock could become a financial-stability event through duration losses, margin calls and collateral stress.

Housing. At 7.40% the 30-year mortgage is already at its highest since November 2023, and a 10-year Treasury at 8–9% would push mortgage costs higher. We do not give precise projections; the likely effects are lower affordability, fewer transactions, weaker residential investment and stress for commercial real estate refinancing.

Equities and credit. The S&P 500 forward price-earnings ratio was 19.0x on 30 September, a forward earnings yield of about 5.3% against a 10-year yield of 5.28%, so the equity premium over the long bond is already close to zero. A Treasury yield at 8% would raise discount rates through the whole cost-of-capital structure: investment-grade and high-yield credit, private credit, leveraged buyouts and technology and growth valuations. Credit spreads have so far held (US aggregate spread 32bp), which leaves little buffer.

13. Global Transmission and the Sovereign-Duration Series

Treasury yields at 8–10% would lift European sovereign yields, JGBs, emerging-market financing costs, global corporate credit and equity discount rates, tighten dollar funding and affect currencies and commodities. Table 6 connects this paper to our BTP and OAT stress tests, which use the same ECB and Fed assumptions and May reference date.

Consistency check. The central 10-year Treasury of 8.25% against a central Bund near 5.5% implies a transatlantic gap of about 275bp, against about 148bp in May (4.45% against 2.97%) and about 175bp now (5.24% against 3.49%). The gap needs to be read against the policy-rate gap, which is about 250bp in the stress (Fed upper bound 9.0% against an ECB deposit rate of 6.5%), so the 10-year gap would be about 1.1 times the policy gap. That ratio was about 0.85 in May (148bp against 175bp) but about 1.2 now (175bp against 150bp) and about 1.3 at the 2023 peak (about 200bp against 150bp, approximate), so 275bp is not an outlier by recent standards. If the ratio reverted to May’s, the gap would be about 210bp and the Bund about 6.1%, at the top of our range. The Bund assumption remains the weakest link in the series.

14. What Could Make the Estimates Too High or Too Low?

Too high if: rapid disinflation, a deep recession, early Fed easing, a falling term premium, stronger foreign Treasury demand, fiscal consolidation or stronger productivity growth.

Too low if: persistent inflation, wider deficits, heavier issuance, weaker foreign demand, a rising term premium, a weaker dollar, or fiscal–monetary credibility concerns.

15. Monitoring Dashboard

The earliest signals that the scenario is becoming reality would be a rising term premium with stable policy expectations, a dollar that weakens as yields rise, weaker auction coverage and rising breakevens, ahead of the fiscal and external ratios.

Conclusion

Under a simultaneous 525bp Fed and 450bp ECB shock, the 10-year Treasury could plausibly reach about 8.25%, and the 20- and 30-year about 8.75%, from official May levels of 4.45%, 4.98% and 4.99%, with a range from about 7% (successful disinflation) to about 10% (a fiscal–inflation–duration loop). The yield would not reach the policy rate mechanically, because long yields price the expected path, inflation and the term premium. The key risk is a repricing of duration, not default: the question is whether the world’s reserve-asset market can absorb a higher price of duration without a loss of fiscal and inflation credibility.