Board of Governors of the Federal Reserve System

Proposed 2026 Stress Test Scenarios

Preface

The Federal Reserve promotes a safe, sound, and efficient banking system that supports the U.S. economy through its supervision and regulation of domestic and foreign banks.

As part of its supervision efforts and as required by the Dodd Frank Act, the Federal Reserve annually conducts a stress test.¹ The stress test assesses how large banks are likely to perform under hypothetical economic conditions.²

The Federal Reserve conducts stress tests to help ensure that large banks are sufficiently capitalized and able to lend to households and businesses even in a severe recession. They evaluate the financial resilience of banks by estimating losses, revenues, expenses, and resulting capital levels under hypothetical economic conditions.

As part of the annual supervisory stress test cycle, the Federal Reserve publishes four documents:

• Stress Test Scenarios describes the hypothetical economic conditions used in the supervisory stress test. The Stress Test Scenarios document is typically published by mid-February.
• Stress Test Methodology provides details about the models and methodologies used in the supervisory stress test. The Supervisory Stress Test Methodology document is typically published at the end of the first quarter.
• Federal Reserve Stress Test Results reports the aggregate and individual bank results of the supervisory stress test, which assesses whether banks are sufficiently capitalized to absorb losses during a hypothetical severe recession. The Federal Reserve Stress Test Results document is typically published at the end of the second quarter.
• Large Bank Capital Requirements announces the individual capital requirement for all large banks, which are partially informed by the results of a supervisory stress test. The Large Bank Capital Requirements document is typically published during the third quarter.

These publications can be found on the Stress Test Publications page (https://www.federalreserve.gov/publications/dodd-frank-act-stress-test-publications.htm).

For information on the Federal Reserve's supervision of large financial institutions, see https://www.federalreserve.gov/supervisionreg/large-financial-institutions.htm.

For information on the Federal Reserve's supervision of capital-planning processes of banks, see https://www.federalreserve.gov/supervisionreg/stress-tests-capital-planning.htm.

For more information on how the Board promotes the safety and soundness of the banking system, see https://www.federalreserve.gov/supervisionreg.htm.

This proposed 2026 scenario includes a request for comment from the public on all aspects of the proposed scenario, including the substantive components of the proposed scenario, as well as the format and contents of this proposal, and the timing of the public comment period. The Board requests that interested parties submit comments to the Board by December 1, 2025, as described in the "Request for Comment" section of this document. The effective date of the 2026 stress test for firms' balance sheet and income statement data will be December 31.

Request for Comment

To enhance the transparency of the annual stress tests and to seek public feedback, the Board is inviting comment on the proposed stress test scenarios for the 2026 supervisory stress test as described in this document. These proposed scenarios, as well as supporting models and data, are available on the Board's website at https://www.federalreserve.gov/supervisionreg/dfa-stress-tests-2026.htm.

Comment Submission

You may submit comments by any of the following methods:

• Agency website: https://www.federalreserve.gov/apps/proposals/. Follow the instructions for submitting comments, including attachments. Preferred method.
• Mail: Ann E. Misback, Secretary, Board of Governors of the Federal Reserve System, 20th Street and Constitution Avenue NW, Washington, DC 20551.
• Hand delivery/courier: Same as mailing address.
• Other means: [email protected]. You must include the docket number in the subject line of the message.

Comments received are subject to public disclosure. In general, comments received will be made available on the Board's website at https://www.federalreserve.gov/apps/proposals/ without change and will not be modified to remove personal or business information, including confidential, contact, or other identifying information. Comments should not include any information such as confidential information that would be not appropriate for public disclosure. Public comments may also be viewed electronically or in person in Room M–4365A, 2001 C St. NW, Washington, DC 20551, between 9 a.m. and 5 p.m. eastern time during federal business weekdays.

Comment Deadline

The Board requests that interested parties submit comments on or before December 1, 2025.

Contact for Further Information

Division of Supervision and Regulation: Doriana Ruffino, Assistant Director, (202) 452-5235; Hillel Kipnis, Assistant Director, (202) 452-2924; and John Simone, Lead Financial Institution Policy Analyst, (202) 245-4256. Division of Financial Stability: William Bassett, Senior Associate Director, (202) 736-5644; Bora Durdu, Deputy Associate Director, (202) 452-3755; Elena Afanasyeva, Principal Economist, (202) 736-1971; Levent Altinoglu, Principal Economist, (202) 721-4503; and Sam Jerow, Financial Analyst, (202) 245-4299. Legal Division: Asad Kudiya, Deputy Associate General Counsel, (202) 360-6887; Julie Anthony, Senior Special Counsel, (202) 658-9400; Brian Kesten, Senior Counsel (202) 843-4079; and Tara Hofbauer, Senior Attorney (202) 680-2503. For users of TDD-TYY, please call 711 from any telephone, anywhere in the United States. Board of Governors of the Federal Reserve System, 20th Street and Constitution Avenue NW, Washington, DC 20551.

Executive Summary

The Federal Reserve's supervisory stress tests help ensure that large banks are able to lend to households and businesses even in a severe recession. The stress tests evaluate the financial resilience of large banks by estimating bank losses, revenues, expenses, and resulting capital levels—which provide a cushion against losses—under hypothetical recession scenarios into the future.³ The Federal Reserve uses the results of a stress test, in part, to set large bank capital requirements.

The proposed 2026 severely adverse scenario is informed by the analysis supporting the Board's proposed update (the Proposed 2025 Scenario Design Policy Statement) to its 2019 Policy Statement on the Scenario Design Framework for Stress Testing (the 2019 Scenario Design Policy Statement).⁴ The Proposed 2025 Scenario Design Policy Statement makes the design of the supervisory severely adverse scenario more transparent by narrowing the ranges of potential values for certain variables through proposed additional guides. The 2026 scenario reflects three enhancements to scenario design, as set forth in the Proposed 2025 Scenario Design Policy Statement, that are intended to describe more clearly to the public how the Board determines the trajectory of variables within the macroeconomic scenario.

First, the Proposed 2025 Scenario Design Policy Statement expands the number and scope of guides used for certain scenario variables. The Board proposes to enhance the two existing guides, already outlined in the 2019 Scenario Design Policy Statement, for the unemployment rate and house price scenario variables, and the proposal describes newly developed guides for additional domestic and international scenario variables. The proposed 2026 severely adverse scenario is consistent with the guides in the Proposed 2025 Scenario Design Policy Statement.

Second, the Board is publishing a new, small-scale macroeconomic model of the U.S. economy that will be used for Stress Testing (hereafter, the "macro model for Stress Testing"), which is described in detail on the Board's website.⁵ The model will be used to determine the values of gross domestic product (GDP), per capita disposable personal income (DPI), inflation, and certain parts of the trajectory of interest rates that are internally consistent with the guide-based change in unemployment and other factors described in the model documentation. This new model is set forth in the Proposed 2025 Scenario Design Policy Statement and has been specifically structured and calibrated to fulfill the needs of the stress testing program. As such, the variable paths prescribed by it should not be interpreted as economic forecasts of the Board or the Federal Open Market Committee (FOMC). Prior scenario paths for these variables were informed by a similar modeling structure with the outputs adjusted using the Board's expert judgment and experience to align with the scenario narrative.

Third, the Board is publishing the model used to generate the global market shock (the GMS model). The Board invites comment on both the macro model for Stress Testing and the GMS model as part of this notice.

The 2026 severely adverse scenario is characterized by a hypothetical severe global recession triggered by an abrupt decline in risk appetite that causes substantial declines in the prices of risky assets, declines in risk-free interest rates and high levels of financial market volatility.⁶ Equity prices fall about 54 percent in the first three quarters of the scenario while the U.S. Market Volatility Index (VIX) spikes and reaches a peak of 72 percent in the second quarter of the scenario. Those conditions also lead to a widening in corporate bond spreads to a level of 5.7 percentage points. The ensuing disruptions depress demand for goods and services from households and prompt businesses to dramatically reduce employment and investment, conditions from which the economy and asset prices are slow to recover. The U.S. unemployment rate rises 5.5 percentage points from the scenario's jump-off point of 4.5 percent in the fourth quarter of 2025 to its peak of 10 percent in the third quarter of 2027. The sharp decline in economic activity leads to a collapse in real estate prices, including a 29 percent decline in nominal house prices and a 40 percent decline in commercial real estate prices. The international portion of the scenario features recessions in three countries or country blocs and a sharp slowdown in developing Asia, and declines in inflation, with all countries or country blocs experiencing deflation. The value of the U.S. dollar appreciates against all countries and country blocs' currencies, except for the Japanese yen, as discussed below. For more information, please see "Details of the Severely Adverse Scenario."

As provided under the Board's stress test rule, banks with large trading operations are tested against a global market shock component that stresses their trading and certain other fair-valued positions.⁷ Furthermore, banks with substantial trading or custodial operations are tested against the default of their largest counterparty.⁸ As noted before, these hypothetical scenarios are not economic forecasts. Their components are described in additional detail in this publication.

Supervisory Stress Test Scenarios

The severely adverse scenario describes a hypothetical set of conditions designed to assess the strength and resilience of banks in an adverse economic environment. Meanwhile, the baseline scenario follows a profile similar to that of average projections from a survey of economic forecasters. These scenarios are not Federal Reserve forecasts.

The scenarios start in the first quarter of 2026 and extend through the first quarter of 2029. Each scenario includes 28 variables; the set of variables for the 2026 supervisory stress test is the same as the set provided in last year's supervisory stress test scenarios. The variables describing economic developments within the United States include:

• Six measures of economic activity and prices: quarterly percent changes (at an annualized rate) in real and nominal GDP, real and nominal disposable personal income, the Consumer Price Index for All Urban Consumers (CPI), and the unemployment rate of the civilian non-institutional population aged 16 years and over.
• Four aggregate measures of asset prices or financial conditions: indexes of house prices, commercial real estate prices, equity prices, and stock market volatility.
• Six measures of interest rates: the rate on 3-month Treasury securities; the yield on 5-year Treasury securities; the yield on 10-year Treasury securities; the yield on 10-year BBB-rated corporate securities; the interest rate associated with conforming, conventional, 30-year fixed-rate mortgages; and the prime rate.

The variables describing international economic conditions in each scenario include three variables in four countries or country blocs:

• The three variables for each country or country bloc: quarterly percent changes (at an annual rate) in real GDP and in consumer price indexes or local equivalent, and the level of the U.S. dollar exchange rate.
• Four countries or country blocs: the euro area (the 20 European Union member states that have adopted the euro as their common currency); the United Kingdom; developing Asia (the nominal GDP-weighted aggregate of China, India, South Korea, Hong Kong Special Administrative Region, and Taiwan); and Japan.

Question 1: The Board invites comment on all aspects of the scenario proposal, including the substantive components of the proposed scenario, the utility of the proposed scenario's format and contents to the public, and whether the length of the public comment period is sufficient to allow the public to comment within the annual stress test cycle.

Baseline and Severely Adverse Scenarios

The following sections describe this year's proposed baseline and severely adverse scenarios. The variables included in these scenarios are provided in tables at the end of this document.⁹ Historical data for the domestic and the international variables are reported in tables 1.A and 1.B, respectively.

The scenario jump-off quarter is set as the fourth quarter of 2025, and the scenario begins in the first quarter of 2026. Given that many of the relevant data points are not yet available for the second, third, or fourth quarters of 2025, estimates for scenario variables for those quarters are constructed using the same methodology used to generate the baseline scenario. The estimates for those quarters will be replaced in the final 2026 scenario with published values after they become available. Estimates will similarly be constructed for any data for the fourth quarter of 2025 that will not be available at publication in February 2026, as has been the case in past scenarios. With new data, the jump-off values for the scenarios will change. Accordingly, the scenario paths will change based on these new jump-off values. For more details, see "Methodology to Update the Scenarios to Incorporate Additional Data Releases."

Baseline Scenario

The proposed 2026 baseline scenario for U.S. real activity, inflation, and interest rates (see table 2.A) is similar to the consensus projections from the September 2025 Blue Chip Financial Forecasts released on August 29, 2025, and the August 2025 Blue Chip Economic Indicators released on August 11, 2025.¹⁰ The long-term components of the baseline scenario for U.S. real activity, inflation, and interest rates are similar to the March 2025 Blue Chip release. The proposed baseline scenario paths for the other scenario variables are constructed according to the macro model for Stress Testing discussed in the Board's Proposed 2025 Scenario Design Policy Statement. This scenario is not a forecast of the Federal Reserve.

The baseline scenario for the United States features moderate economic growth. The unemployment rate moves up to 4.6 percent in the first quarter of 2026, and stays at that level until the third quarter of 2026, before gradually declining to 4.2 percent in the third quarter of 2028, where it remains through the end of the scenario. Real GDP growth rises from 0.8 percent in the fourth quarter of 2025 to 2 percent by the first quarter of 2027 and hovers around that rate for the rest of the scenario. Inflation, measured as the quarterly change in the CPI and reported as an annualized rate, gradually declines from 3.4 percent at the end of 2025 to 2.2 percent in the first quarter of 2027, where it remains through the end of the scenario. The 3-month Treasury rate decreases from 4.0 percent at the end of 2025 to 3.8 percent in the first quarter of 2026, after which it declines to 3.3 percent in the fourth quarter of 2026. It remains there through the first quarter of 2028, after which it gradually declines to 3.1 percent through the end of the scenario. The 10-year Treasury yield ticks down from 4.3 percent in the fourth quarter of 2025 to 4.2 percent in the first quarter of 2026 and then declines gradually to 3.9 percent by the end of the scenario. The prime rate follows a path similar to short-term interest rates, but sits at a level 3 percentage points higher, reflecting the typical spread between the prime rate and the top of the federal funds target range.¹¹ Mortgage rates decline gradually from 6.4 percent at the end of 2025 to 5.5 percent by the fourth quarter of 2028 where they remain for the rest of the scenario. Yields on BBB-rated corporate bonds hover around their level in the fourth quarter of 2025 throughout the scenario, while the spread between yields on BBB-rated bonds and yields on 10-year Treasury securities increases gradually to a level of 1.6 percentage points by the third quarter of 2027 and remains around that level through the rest of the scenario.

Equity prices increase about 4.3 percent per year throughout the scenario. Equity market volatility, as measured by the VIX, increases gradually from 23 percent in the fourth quarter of 2025 to 25 percent in the third quarter of 2027, where it remains through the end of the scenario. Nominal house prices decline somewhat through the first quarter of 2027 before gradually increasing through the remainder of the scenario, while commercial real estate prices increase about 4.3 percent per year.

The baseline paths for the international variables (see table 2.B) are similar to the trajectories reported in the August 2025 Blue Chip Economic Indicatorsand the International Monetary Fund's April 2025 World Economic Outlook.¹² In the baseline scenario, real GDP growth in developing Asia increases from 2.8 percent to 5.6 percent in the third quarter of 2026, after which it gradually declines to 4.1 percent in the third quarter of 2027. It then fluctuates between 4.8 percent and 4.2 percent through the end of the scenario. Real GDP growth in the euro area increases from 0.3 percent at the end of 2025 to 2 percent by the third quarter of 2026. It then declines gradually to 1.1 percent in the second quarter of 2027, after which it fluctuates between 1.2 percent and 1.5 percent through the end of the scenario. Real GDP growth in the United Kingdom increases from 0.4 percent at the end of 2025 to 1.8 percent by the third quarter of 2026, after which it gradually declines to 1.1 percent in the second quarter of 2027. It then fluctuates between 1.2 and 1.4 through the end of the scenario. GDP growth in Japan begins at negative 0.4 percent in the fourth quarter of 2025 and increases to 1.8 percent in the third quarter of 2026. It then gradually declines to 0.2 percent through the third quarter of 2027, after which it hovers between 0.4 percent and 1.1 percent through the end of the scenario.

Consumer price inflation in the euro area increases from 1.8 percent to 2 percent in the second quarter of 2027 and hovers around that level for the rest of the scenario. Consumer price inflation in the United Kingdom declines from 2.7 percent to 2.1 percent in the fourth quarter of 2026 and hovers around that level through the end of the scenario. Inflation in Japan increases from 1.8 percent in the fourth quarter of 2025 to 2.4 percent in the second quarter of 2028, where it remains through the end of the scenario. The inflation rate in developing Asia increases gradually from 1.3 percent to 2.2 percent by the third quarter of 2027 and hovers around there for the rest of the scenario.

Question 2: The Board invites comment on all aspects of the proposed baseline scenario.

Severely Adverse Scenario

As noted, the proposed 2026 severely adverse scenario is informed by the Board's Proposed 2025 Scenario Design Policy Statement.¹³ Additionally, the guides described in the Proposed 2025 Scenario Design Policy Statement, the macro model for Stress Testing, and the proposed 2026 scenario that the Board has constructed based on that framework, are consistent with the Board's 2019 Scenario Design Policy Statement.¹⁴ While this proposal is designed to provide more transparency around the Board's process and decisions, the Board expects that the resulting scenarios will, on balance, be similar to those that it has used in prior annual stress test exercises.

In accordance with the Board's Proposed 2025 Scenario Design Policy Statement, the Board's scenario design approach is informed by current macroeconomic and financial conditions, which are summarized here for the purposes of the design of the proposed severely adverse scenario. The current unemployment rate is near the consensus forecast of the long-run natural unemployment rate in the March 2025 Blue Chip Economic Indicators. As discussed in the previous section, the proposed 2026 baseline scenario paths for GDP growth and inflation see GDP growth rising and inflation remaining somewhat above the FOMC's 2 percent longer-run inflation goal.¹⁵ Equity prices in the proposed baseline scenario are projected to have increased 7 percent from the end of last year through the fourth quarter of 2025 and about 30 percent over the past two years. While house prices have remained largely flat over the past two years, they are about 34 percent higher than their level five years ago. Commercial real estate prices have changed little since early 2024 after significant declines in 2023, and the BBB spread remains low relative to its historical series. This summary of current macroeconomic and financial conditions should not be interpreted as an assessment of likely future developments but informs the design of the proposed severely adverse scenario, consistent with the Board's expectation that scenario severity generally will be higher during economic expansions or periods when asset values have increased or to avoid adding sources of procyclicality through the stress test.

In the Proposed 2025 Scenario Design Policy Statement, the guides for the scenario variables include ranges for the values controlling certain features of the scenario paths, within which the Board has discretion to choose values based on current conditions and other relevant considerations. For this year's scenario, the Board proposes to calibrate the values of scenario characteristics for which the guides provide flexibility, such as the peak values of the BBB spread and VIX, near or in the upper one-third of their ranges of severity. The Board chose this calibration to generate an appropriate level of overall severity given the paths of the unemployment rate and house prices, and given the prevailing macroeconomic and financial conditions described above. This calibration is consistent with the Board's goal of improving predictability and transparency of the annual scenarios. The calibration also reflects both the Board's principle of conservatism and its goal that the annual stress tests should not add to other sources of procyclicality in the financial system.

Details of the Severely Adverse Scenario

The proposed 2026 severely adverse scenario is characterized by a severe global recession triggered by an abrupt decline in risk appetite that causes substantial declines in risky asset prices and declines in risk-free interest rates. At times during the first months of this scenario, financial market functioning is impaired, leading to substantial additional volatility. Those disruptions spill over into large reductions in household demand for goods and services and significantly reduce employment and business investment. The low levels of risk appetite and the declines in income and wealth persist for some time and lead to a protracted recession in the United States and abroad. This is a hypothetical scenario designed to assess the strength and resilience of banks and does not represent a forecast of the Federal Reserve.

Consistent with both the 2019 Scenario Design Policy Statement and the Proposed 2025 Scenario Design Policy Statement, under the proposed 2026 severely adverse scenario, the U.S. unemployment rate climbs to a peak of 10 percent in the third quarter of 2027 (see table 3.A), a 5.5 percentage point increase relative to its fourth-quarter 2025 level. The unemployment rate reaches its peak in the seventh quarter of the scenario, which is the midpoint of the range of six to eight quarters generally established by the guide for the unemployment rate to reach its peak. The Board chose this middle value for the timing of the peak of the unemployment rate to balance the marginal effect of a slightly more-severe or less-severe path for the unemployment rate on overall scenario severity.

House prices fall steadily through the fourth quarter of 2027, reaching a trough that is about 29 percent below their level in the fourth quarter of 2025. The level of this trough is determined by the relevant component of the guide for the house prices in the Proposed 2025 Scenario Design Policy Statement, which is the same as the guide in the Board's 2019 Scenario Design Policy Statement. The eighth-quarter trough timing is at the lower end of the range of 8 to 10 quarters prescribed by the Proposed 2025 Scenario Design Policy Statement's guide for house prices, in line with the Board's proposal to calibrate most of the scenario variables for which the Board retains some discretion near the upper one-third of their ranges of severity.

The spread between mortgage rates and 10-year Treasury yields widens 1.3 percentage points to reach a level of 3.4 percentage points by the third quarter of 2026 before narrowing to a level of about 2.4 percentage points at the end of the severely adverse scenario. The jump-off-to-peak increase of 1.3 percentage points is at the upper one-third of the range of 0.7 to 1.6 percentage points specified by the proposed guide for the mortgage spread. The mortgage spread reaches its peak in the third quarter of the scenario, which is the early end of the range of three to four quarters allowed by the guide for the spread to reach its peak. In determining the trajectory of the mortgage spread from its jump-off point to its trough, the guide for the mortgage spread in the Board's Proposed 2025 Scenario Design Policy Statement specifies a range for the proportion of declines in each quarter along the trajectory to the trough. In the proposed severely adverse scenario, 62 percent of the jump-off-to-peak increase in the mortgage spread occurs in the first quarter of the scenario, which is near the upper one-third of the range of 50 to 70 percent prescribed by the proposed guide, in line with the Board's proposal to calibrate most of the scenario variables for which the Board retains some discretion near the upper one-third of their ranges of severity.

Equity prices fall about 54 percent from the fourth quarter of 2025 through the third quarter of 2026. The jump-off-to-trough decline of about 54 percent is prescribed by the guide for equity prices in the Board's Proposed 2025 Scenario Design Policy Statement, and reflects that equity prices, as proxied by the U.S. Dow Jones Total Stock Market Index, rise 7 percent in 2025 in the baseline scenario. The guide for equity prices in the Proposed 2025 Scenario Design Policy Statement reflects the principle that equity prices in the scenario fall more after periods when they have risen more. In the proposed scenario, 67 percent of the decline in equity prices occurs in the first quarter of the scenario while 17 percent occurs in the second quarter. These values are at the upper one-third of the respective ranges specified by the guide, in line with the Board's proposal to calibrate most of the scenario variables for which the Board retains some discretion near the upper one-third of their ranges of severity.

The VIX, measured as the highest daily closing value per quarter, reaches a peak of 72 percent in the second quarter of 2026. Seventy-three percent of the increase in the VIX occurs in the first quarter of the scenario. This value, and the peak value of the VIX of 72 percent, are in the top one-third of their respective ranges in the proposed guide for the VIX.

Conditions in corporate bond markets deteriorate markedly, as the scenario specifies a sudden decline in risk appetite and worsening business conditions. The spread between yields on BBB-rated bonds and yields on 10-year Treasury securities increases 4.4 percentage points by the third quarter of 2026, reaching a level of 5.7 percentage points. Seventy percent of the jump-off-to-peak increase in the BBB spread occurs in the first quarter of the scenario. The peak level of 5.7 percentage points and the pace of the increase in the BBB spread are both near the upper one-third of their respective ranges prescribed by the guide for the BBB spread in the Board's Proposed 2025 Scenario Design Policy Statement.

Commercial real estate prices reach a trough in the fourth quarter of 2027 that is 40 percent below their level at the end of 2025. The jump-off-to-trough decline in commercial real estate prices is at the top one-third of the range of 30 to 45 percent prescribed by the guide for commercial real estate prices in the Proposed 2025 Scenario Design Policy Statement. Commercial real estate prices reach their trough in the eighth quarter of the scenario, which is the bottom of the range of eight to ten quarters prescribed by the guide for commercial real estate prices to reach its trough, in line with the Board's proposal to calibrate most of the scenario variables for which the Board retains some discretion near the upper one-third of their ranges of severity.

The proposed scenario paths for real GDP, inflation, and the 3-month Treasury rate are generated by the macro model for Stress Testing discussed in the Proposed 2025 Scenario Design Policy Statement, when given the path for the unemployment rate.¹⁶ Real GDP declines 4.8 percent from the fourth quarter of 2025 to its trough in the second quarter of 2027, before recovering to the level at the jump-off. This path for real GDP is based on the path for the unemployment rate given the version of Okun's law in the macroeconomic model.¹⁷ Real disposable income, which depends in part on real GDP, declines about 1 percent in the proposed scenario from the fourth quarter of 2025 to its trough in the fourth quarter of 2026, before recovering and gradually surpassing its level at the jump-off. The Phillips curve component of the model projects a significant reduction in inflation given the rising unemployment rate and the rapid decline in aggregate demand for goods and services. Inflation, measured as the quarterly change in the CPI and reported as an annualized rate, falls from 3.4 percent in the fourth quarter of 2025 to 1.1 percent in the second quarter of 2027 and then gradually increases only to 1.3 percent by the end of the scenario. The paths of inflation and a measure of the output gap, which depends on the unemployment rate, are inputs to the Taylor rule used in the macroeconomic model, which in turn determines the 3-month Treasury rate. The 3-month Treasury rate falls significantly from 4 percent in the fourth quarter of 2025 to 0.1 percent by the second quarter of 2026 and remains there for the remainder of the scenario.¹⁸

The paths for long-term interest rates, as measured by the 5-year and 10-year Treasury yields, are determined by two components. The initial paths to the trough are determined by the proposed guides. The paths from the trough to the end of the scenario are determined by the macro model for Stress Testing. These paths are therefore informed by the scenario paths of short-term interest rates and estimates of likely term premiums in an economic environment consistent with the narrative for the severely adverse scenario. Overall, the jump-off-to-trough declines in the 5-year and 10-year Treasury yields are consistent with key features of the scenario, including severe declines in aggregate demand for goods and services and in risk appetite at the start of the scenario. In general, a decline in long-term interest rates may have a positive or negative effect on the severity of the scenario for a given firm depending on the firm's exposure to interest rate risk—which may vary from year to year depending on the firm's portfolio—due to the opposing effects that changes in interest rates have on net interest margins and on the market-adjusted valuations of long-term, fixed-rate securities.

In proposing the paths for long-term interest rates, the Board took into consideration the current level of long-term interest rates and their changes in recent recessions. In the proposed scenario, the 5-year and 10-year Treasury yields fall 2.5 percentage points and 2 percentage points to 1.3 percent and 2.3 percent, respectively, by the fourth quarter of 2026. These declines are roughly in line with the declines of 2.6 percentage points and 1.6 percentage points, respectively, experienced between the third quarter of 2007 and the first quarter of 2009 during the 2007–2009 financial crisis. That recession was comparable in magnitude and length to the hypothetical recession in the scenario, and significantly deeper and longer lasting than the early 2000s recession or the 2020 pandemic recession, during which rates declined by less. Both the 5-year and the 10-year yields reach their troughs in the fourth quarter of the proposed scenario. Consistent with their guides, the declines in the 5-year and 10-year Treasury yields are frontloaded, with about 55 percent of the declines occurring in the first quarter of the scenario, in line with the severe contraction in economic activity and heightened uncertainty featured in the scenario. The sizable declines in the 5-year and 10-year Treasury yields, the speed of those declines, and the timing of their troughs are appropriate given the modeled decline of the 3-month Treasury bill yield and its persistence at a near-zero level in this scenario.¹⁹ In addition, the timing of the troughs of the long-term interest rates in the fourth quarter of the proposed scenario reflects the Board's willingness to balance the effects that changes in interest rates have on net interest margins and on the market-adjusted values of long-term, fixed-rate securities.

The international component of the proposed severely adverse scenario involves a sharp deterioration in foreign economic activity, in line with the experience of the 2007–2009 financial crisis (see table 3.B). In the euro area, the United Kingdom, and Japan, real GDP declines about 7.5 percent relative to its value in the baseline scenario by the end of 2026, which is consistent with the deviation observed in the level of real GDP in the first quarter of 2009 from a baseline path derived from the April 2008 IMF World Economic Outlook (WEO) forecast as described in the international guides. As a result, these advanced economies experience recessions with real GDP declining from jump-off to trough by 5.9 percent in the euro area, 6.1 percent in the United Kingdom, and 6.2 percent in Japan. In developing Asia, real GDP growth slows down and runs about 3 percent below baseline by the end of 2026. Over the same period, inflation declines about 3 percentage points below baseline in the advanced economies, and 5 percentage points below baseline in developing Asia. The U.S. dollar appreciates about 15 percent against the euro and the British pound, while it depreciates mildly against the Japanese yen by 1 percent, consistent with its historical behavior between the first quarter of 2008 and the first quarter of 2009. Consistent with the guides for the international variables in the Board's Proposed 2025 Scenario Design Policy Statement, the deviation of each international scenario variable from its baseline path is similar to that observed during the 2007–2009 financial crisis.

Question 3: The Board invites comment on all aspects of the proposed severely adverse scenario. What are the advantages and disadvantages of the proposed scenario? Are there risks or conditions that the Board should consider incorporating into the scenario that are not described in this scenario?

Question 4: Is the Board's proposal to calibrate similarly the severity of several of those scenario variables for which the Board retains some discretion in setting their values appropriate? Is the Board's proposed calibration of those variables appropriate in terms of severity? Should the Board adjust any of the variable paths? If so, what information should the Board consider in assessing these variable paths?

In prior years, the scenario disclosure document featured a section called "Additional Key Features of the Severely Adverse Scenario," which briefly described assumptions regarding the distribution of stress not fully captured by the BBB spread, indexes of U.S. real estate prices, and the four country blocs in the international scenario. For more information on the distribution of stress not fully captured by the BBB spread, indexes of U.S. real estate prices, and the four country blocs in the international scenario, please refer to the detailed documentation released separately. Specifically, see section III.B of the October 2025 notice.

Comparison of the 2026 Severely Adverse Scenario to the 2025 Severely Adverse Scenario

In general, changes in the paths of scenario variables from year to year reflect a combination of changes arising from different jump-off levels and, for those variables where the Board retains some discretion, choices by the Board about the appropriate level of scenario severity. In the proposed severely adverse scenario, the unemployment rate in the United States rises to the same level as in the 2025 severely adverse scenario. The increase in the unemployment rate currently is smaller in this year's scenario, reflecting the higher jump-off level implied by the August 2025 Blue Chip Economic Indicators. This is consistent with the guide for the unemployment rate in both the Proposed 2025 Scenario Design Policy Statement and the 2019 Scenario Design Policy Statement, which both call for an increase in the unemployment rate to at least a level of 10 percent given current conditions.

In this year's scenario, the paths for real GDP, real disposable income, inflation, and the 3-month Treasury rate are generated by the macro model for stress testing, given the path for the unemployment rate and other modeled factors. Real GDP declines by less and reaches its trough one quarter later compared to last year's scenario. The decline in inflation is larger compared to last year's scenario. These differences reflect both the new model and this year's jump-off conditions. The path for real disposable income also features a lower decline compared to last year, as it depends on the path for real GDP.

The 3-month Treasury rate reaches the same trough level as in last year's scenario but declines somewhat less, owing to its slightly lower jump-off level in this year's proposed scenario. As described above, the 5-year and 10-year Treasury yields decline somewhat less in response to the hypothetical drop in economic activity and inflation and reach their troughs somewhat later than in last year's scenario. The proposed severely adverse scenario also features a somewhat smaller decline in house prices as compared to the previous year's severely adverse scenario, consistent with the guide for house prices in both the Proposed 2025 Scenario Design Policy Statement and the 2019 Scenario Design Policy Statement, both of which call for a smaller decline in house prices when the ratio of nominal house prices to per capita disposable income is lower. Mortgage spreads reach higher levels than in last year's scenario. Commercial real estate prices decline more than in last year's scenario, but the decline is similar to past scenarios.

The decline in equity prices is somewhat more severe than in last year's scenario, reflecting the effect of the moderate increase in equity prices since last year in the proposed guide for equity prices. In this year's scenario, the VIX and the BBB spread both reach levels that are higher than in last year's scenario but similar to those in prior years. The international component of the proposed severely adverse scenario is consistent with the design of international scenarios described in both the Proposed 2025 Scenario Design Policy Statement and the Board's 2019 Scenario Design Policy Statement. The proposed scenario shows a recessionary episode that, relative to last year's severely adverse scenario, is somewhat less severe for Japan and developing Asia and is somewhat more severe for the euro area and the United Kingdom. This scenario is consistent with the level of stress in foreign economies that manifested during the 2007–2009 financial crisis.

Methodology to Update the Scenarios to Incorporate Additional Data Releases

After disclosing a proposed severely adverse scenario for public comment, the Board expects to make two types of revisions to the scenario. First, the Board may make revisions to the scenario to address comments received from the public. Second, the Board will make revisions to incorporate additional data releases that occur after the publication of this proposal and affect the historical values of specific scenario variables. This section of notice addresses the second: data-based revisions to the severely adverse scenario.

Generally, the historical dataset that accompanies the proposed severely adverse scenario will be based on data released through August 29. For the proposed 2026 scenarios, the final dataset will incorporate data released through mid-January, as well as updated external forecasts. In particular, the final dataset will reflect new data for the jump-off quarter of the scenarios. As the scenario paths for all variables are dependent on the conditions during the jump-off quarter, the severely adverse scenario will be revised to reflect the updated jump-off conditions.

The process for updating the scenario paths given an updated historical dataset will comprise two steps. First, variables that are determined by formula-based guides will be updated by referencing the scenario guides in the Proposed 2025 Scenario Design Policy Statement to the new jump-off points in the historical dataset.²⁰ Variables that are informed by guides that allow for some discretion by the Board are likely to be adjusted to match the same level of the peak or trough in the released scenario, unless doing so violated some other aspect of that particular guide. These updates likely will require a change in the trajectory from the new jump-off points that will follow the process described in that portion of the guides. Next, the updated historical dataset and scenario paths for guide-based variables will be used as inputs to rerun the scenario model to update the scenario paths for all model-determined variables.²¹

Question 6: The Board invites comment on how it plans to update the scenarios to incorporate additional data releases.

Guide-based variables

The scenario guides described in the Proposed 2025 Scenario Design Policy Statement allow, in some cases, for judgment within a range. As such, updating the paths for the guide-based variables will follow the general principle that revisions to the severely adverse scenario that arise because of changes in macroeconomic and financial conditions should avoid making the scenario more severe, should financial and macroeconomic conditions worsen. Conversely, the revisions should also avoid making the scenario less severe should financial and macroeconomic conditions improve. An additional consideration is that the overall scenario severity is a function of both levels and changes in the scenario variables.

In line with this principle, the Board expects that its standard approach to revising the scenario in response to new data is likely to keep constant the level reached by each variable at its peak or trough in the scenario released for comment for the guide-based variables, so long as this level implies a change relative to the jump-off value that is aligned with that variable's scenario guide. The Board expects to deviate from this default updating method for the peak or trough value of guide-based variables in cases where, for example, maintaining the peak or trough level achieved in the scenario would result in a violation of any element of the scenario guide for a variable. This could be the case when data released between the disclosure of the proposed and final scenario indicate significant changes in the jump-off conditions. In such instances, the Board expects to adjust to target an appropriate level of change, as described in the Proposed 2025 Scenario Design Policy Statement with respect to that variable's scenario guide and informed by the Board's supervisory experience and expertise.²²

Under either approach, the revisions are expected to keep constant the quarters in which the peak or trough values are attained, barring substantial changes in the economic or financial environment. For each variable, the revisions will be phased in by adjusting the scenario path through the peak or trough, thus altering the changes relative to the jump-off point.

Furthermore, the values of variables at the end of the scenario and the trajectory from the trough to the end point will update to reflect the new jump-off data. The Board expects this process will occur as outlined in the scenario guide in the Proposed 2025 Scenario Design Policy Statement.

Question 7: The Board invites comment on how it plans to update the scenarios with respect to guide-based variables.

Model-based variables

To update the scenario paths for model-based variables, the Board expects to recompute their full scenario paths using the macro model for Stress Testing outlined in the policy statement taking as inputs the updated historical dataset and the updated scenario paths for the guide-based variables. This approach may result in revisions to the scenario paths for these variables relative to the proposed scenario disclosed for public comment but should maintain the appropriate level of scenario severity and ensure overall consistency across all the variables in the stress test scenario.

Question 8: The Board invites comment on how it plans to update the scenarios with respect to model-based variables.

Global Market Shock Component for the Supervisory Severely Adverse Scenario

The global market shock component for the severely adverse scenario (global market shock) is a set of hypothetical shocks to a large set of risk factors reflecting general market distress and heightened uncertainty. Banks with significant trading activity must consider the global market shock as part of the supervisory severely adverse scenario in their company-run stress test.²³ The losses associated with the global market shock are recognized in the first quarter of the scenario and are carried through all subsequent quarters. In addition, certain large and highly interconnected firms must apply the same global market shock to project losses under the counterparty default scenario component. The global market shock is applied to positions held by the banks on a given as-of date, which must fall between October 1 and March 1.²⁴ These shocks do not represent a forecast of the Federal Reserve.

The design and specification of the global market shock differ from the macroeconomic scenarios in several ways. First, profits and losses from trading and counterparty credit are measured in mark-to-market terms, while revenues and losses from traditional banking are generally measured using the accrual method. Another key difference is the timing of loss recognition. The global market shock affects the mark-to-market value of trading positions and counterparty credit losses in the first quarter of the severely adverse scenario. This timing is based on an observation that market dislocations can happen rapidly and unpredictably at any time under stressed conditions. Applying the global market shock in the first quarter ensures that potential losses from trading and counterparty exposures are incorporated into banks' capital ratios in each quarter of the severely adverse scenario.

The global market shock is specified by a large set of risk factors that include, but are not limited to

• public equity returns from key advanced economies and from developing and emerging market economies, along with selected points along term structures of equity option-implied volatilities;
• exchange rates of foreign currencies, along with selected points along term structures of foreign exchange option-implied volatilities;
• government yields at selected maturities (e.g., 10-year U.S. Treasuries), swap rates, and other types of interest rates for key advanced economies and from developing and emerging market economies;
• implied volatilities on interest rate options for selected maturities and expiration dates, which are key inputs to the pricing of interest rate derivatives;
• futures prices at various expiration dates for commodity products such as energy, oil, metals, and agricultural products; and
• credit spreads or prices for selected credit-sensitive products, including corporate bonds, credit default swaps (CDS), securitized products, sovereign debt, and municipal bonds.

The Board considers emerging and ongoing areas of financial market vulnerabilities in the development of the global market shock. This assessment of potential vulnerabilities is informed by financial stability reports, supervisory information, and internal and external assessments of potential sources of distress such as geopolitical, economic, and financial market events.

The global market shock includes a set of risk factor shocks to financial market variables that apply to all banks with significant trading activity. Depending on the type of financial market vulnerability that the global market shock is intended to assess, the risk factor shocks could be based on a single historical episode, multiple historical periods, hypothetical events that are based on relevant economic indicators of economic and financial conditions, or a hybrid approach comprising some combination of historical episodes and hypothetical events. A market shock based on hypothetical events may result in changes in risk factors that have not been observed historically.²⁵

Risk factor shocks are calibrated using assumed time horizons. The calibration horizons reflect several considerations related to the scenario being modeled. One important consideration is the liquidity characteristics of different risk factors. These characteristics may vary depending on the specified market shock narrative. More specifically, the calibration horizons reflect the variation in the speed at which banks could reasonably close out, or effectively hedge, risk exposures in the event of market stress. The calibration horizons are generally longer than the typical times needed to liquidate exposures under normal conditions because they are designed to capture the unpredictable liquidity conditions that prevail in times of stress.²⁶ In addition, shocks to risk factors in more liquid markets, such as those for government securities, foreign exchange, or public equities, are calibrated to shorter horizons (1 month), while shocks to risk factors in less liquid markets, such as those for non-agency securitized products, have longer calibration horizons (3 months).

2026 Global Market Shock Component of the Supervisory Severely Adverse Scenario

The 2026 global market shock is characterized by heightened market expectations of persistently high inflation, higher commodity prices, and a global recession. The scenario has certain elements in common with prior episodes of market reactions to periods of expected high inflation combined with low growth, such as the oil crisis of the 1970s. That period was also characterized by commodity price increases.

Both short-term and long-term Treasury rates rise sharply driven by higher inflation expectations. Heightened inflation expectations drive commodity prices upward.

The expected fall in economic activity leads to notable equity price declines across global markets. Concerns about corporate credit defaults in light of the economic slowdown leads to wider credit spreads.

The U.S. dollar strengthens, exhibiting large gains against the euro and moderate gains against the Japanese yen driven by higher yields in the U.S.

Comparison of the 2026 Global Market Shock Component and the 2025 Global Market Shock Component

The 2026 global market shock features expectations for higher inflation, while last year's global market shock was characterized by expectations for lower inflation. Accordingly, the main difference between the 2025 and 2026 global market shocks is the behavior of interest rates, foreign exchange rates, and commodities prices. Treasury yields increase across all tenors in the current global market shock, whereas last year, yields decreased with short-term yields falling more than long-term yields. Inflation breakeven rates increase in the current global market shock, while they decreased in the 2025 global market shock.

The U.S. dollar appreciates against most major currencies in both the 2025 and 2026 global market shocks. However, an exception is that the dollar appreciates against the Japanese yen in the 2026 global market shocks, while it depreciated against the yen in the 2025 global market shock. In both years, the U.S. dollar appreciates with respect to emerging market currencies. Commodities—such as gold, oil, and natural gas—exhibit price increases due to inflationary pressures in the current global market shock, while commodity prices decreased in the 2025 global market shock. Credit spreads widen and equity prices fall in both the 2025 and 2026 global market shocks.

Counterparty Default Component of the Supervisory Severely Adverse Scenario

Large firms with substantial trading or custodial operations are required to incorporate a counterparty default scenario component into their supervisory severely adverse scenario for 2025 and recognize associated losses in the first quarter of the scenario.²⁷ This component involves the unexpected default of the firm's largest counterparty. In identifying its largest counterparty, a firm subject to the counterparty default component will not consider certain entities.²⁸ In addition to sovereign entities and qualified central counterparties, certain multilateral development banks and supranational entities (International Bank for Reconstruction and Development, International Monetary Fund, Bank for International Settlements, European Commission, and European Central Bank) will not be considered for the counterparty default component to better align the treatment of these entities across regulatory exercises. The Board is separately proposing to exclude certain additional sovereign entities from the counterparty default component.²⁹

The counterparty default scenario component is an add-on to the Federal Reserve's severely adverse scenario: firms are required to estimate and report the potential losses and related effects on capital associated with the unexpected default of the counterparty that would generate the largest net stressed losses across their derivatives and securities financing transactions.

Net stressed losses are estimated by applying the global market shock to revalue securities financing transactions and derivatives, including collateral posted or received. The as-of date for the counterparty default scenario component for the is the same as-of date as for the global market shock component.³⁰

Question 9: The Board invites comment on how the separate proposal to exclude certain additional sovereign entities from the counterparty default component should be incorporated into the 2026 global market shock.

Variables for the Supervisory Scenarios

Table 1.A. Historical Data: Domestic variables, Q1:2000–Q4:2025

Table 1.B. Historical Data: International variables, Q1:2000–Q4:2025

Table 2.A. Supervisory baseline scenario: Domestic variables, Q1:2026–Q1:2029

Table 2.B. Supervisory baseline scenario: International variables, Q1:2026–Q1:2029

Table 3.A. Supervisory severely adverse scenario: Domestic variables, Q1:2026–Q1:2029

Table 3.B. Supervisory severely adverse scenario: International variables, Q1:2026–Q1:2029

Notes Regarding Scenario Variables

The following are descriptions of data as released through August 29, 2025. The 2025:Q3 and 2025:Q4 values of all variables are estimates. Additionally, the 2025:Q2 values for the U.S. House Price Index and U.S. Commercial Real Estate Price Index are estimates.

U.S. real GDP growth: Quarterly percent change in real gross domestic product (chained 2017 dollars), expressed at an annualized rate, Bureau of Economic Analysis (NIPA table 1.1.6, line 1).

U.S. nominal GDP growth: Quarterly percent change in gross domestic product (current dollars), expressed at an annualized rate, Bureau of Economic Analysis (NIPA table 1.1.5, line 1).

U.S. real disposable income growth: Quarterly percent change in real disposable personal income (current-dollar values divided by the price index for personal consumption expenditures), expressed at an annualized rate, Bureau of Economic Analysis (NIPA table 2.1, line 27, and NIPA table 1.1.4, line 2, respectively).

U.S. nominal disposable income growth: Quarterly percent change in disposable personal income (current dollars), expressed at an annualized rate, Bureau of Economic Analysis (NIPA table 2.1, line 27).

U.S. unemployment rate: Quarterly average of seasonally adjusted monthly unemployment rates for the civilian, non-institutional population aged 16 years and older, Bureau of Labor Statistics (series LNS14000000).

U.S. CPI inflation: Percent change in the quarterly average of seasonally adjusted monthly levels of the all-items CPI for all urban consumers (CPI-U), expressed at an annualized rate, Bureau of Labor Statistics (series CUSR0000SA0).

U.S. 3-month Treasury rate: Quarterly average of 3-month Treasury bill secondary market rate on a discount basis, H.15 Release, Selected Interest Rates, Federal Reserve Board (series RIFSGFSM03_N.B).

U.S. 5-year Treasury yield: Quarterly average of the yield on 5-year U.S. Treasury notes, constructed for the FRB/U.S. model by Federal Reserve staff based on the Svensson smoothed term structure model (see Lars E. O. Svensson, 1995, "Estimating Forward Interest Rates with the Extended Nelson–Siegel Method," Quarterly Review, no. 3, Sveriges Riksbank, pp. 13–26).

U.S. 10-year Treasury yield: Quarterly average of the yield on 10-year U.S. Treasury notes, constructed for the FRB/U.S. model by Federal Reserve staff based on the Svensson smoothed term structure model (see Svensson, "Estimating Forward Interest Rates").

U.S. BBB corporate yield: Quarterly average of ICE BofAML U.S. Corporate 7-10 Year Yield-to-Maturity Index, ICE Data Indices, LLC, used with permission (C4A4 series).

U.S. mortgage rate: Quarterly average of weekly series for the interest rate of a conventional, conforming, 30-year fixed-rate mortgage, obtained from the Primary Mortgage Market Survey of the Federal Home Loan Mortgage Corporation.

U.S. prime rate: Quarterly average of monthly series, H.15 Release (Selected Interest Rates), Federal Reserve Board (series RIFSPBLP_N.M).

U.S. Dow Jones Total Stock Market (Float Cap) Index: End-of-quarter value via Bloomberg Finance L.P.

U.S. House Price Index: Price Index for Owner-Occupied Real Estate, Z.1 Release (Financial Accounts of the United States), Federal Reserve Board (series FL075035243.Q divided by 1000).

U.S. Commercial Real Estate Price Index: Commercial Real Estate Price Index, Z.1 Release (Financial Accounts of the United States), Federal Reserve Board (series FL075035503.Q divided by 1000).

U.S. Market Volatility Index (VIX): VIX converted to quarterly frequency using the maximum close-of-day value in any quarter, Chicago Board Options Exchange via Bloomberg Finance L.P.

Euro area real GDP growth: Quarterly percent change in real gross domestic product at an annualized rate, staff calculations based on Statistical Office of the European Communities via Haver, extended back using ECB Area Wide Model dataset (ECB Working Paper series no. 42).

Euro area inflation: Percent change in the quarterly average of the harmonized index of consumer prices at an annualized rate, staff calculations based on Statistical Office of the European Communities via Haver.

Developing Asia real GDP growth: Quarterly percent change in real gross domestic product at an annualized rate, staff calculations based on data from Bank of Korea via Haver; National Bureau of Statistics of China via Haver; Indian Central Statistics Office via Haver; Census and Statistics Department of Hong Kong via Haver; and Taiwan Directorate-General of Budget, Accounting and Statistics via Haver.

Developing Asia inflation: Percent change in the quarterly average of the consumer price index, or local equivalent, at an annualized rate, staff calculations based on data from National Bureau of Statistics of China via Haver; Indian Ministry of Statistics and Programme Implementation via Haver; Labour Bureau of India via Haver; Statistics Korea (KOSTAT) via Haver; Census and Statistics Department of Hong Kong via Haver; and Taiwan Directorate-General of Budget, Accounting and Statistics via Haver.

Japan real GDP growth: Quarterly percent change in real gross domestic product at an annualized rate from 1980 to present and percent change in gross domestic expenditure at an annualized rate prior to 1980, Cabinet Office of Japan via Haver.

Japan inflation: Percent change in the quarterly average of the consumer price index at an annualized rate, based on data from the Ministry of Internal Affairs and Communications via Haver.

U.K. real GDP growth: Quarterly percent change in real gross domestic product at an annualized rate, U.K. Office for National Statistics via Haver.

U.K. inflation: Percent change in the quarterly average of the consumer price index at an annualized rate from 1988 to present and percent change in the quarterly average of the retail prices index prior to 1988, staff calculations based on data from the U.K. Office for National Statistics via Haver.

Exchange rates: End-of-quarter exchange rates, H.10 Release (Foreign Exchange Rates), Federal Reserve Board.

Footnotes