Deep Dive: DFAST 2024 Stress Test Scenarios

On Feb. 15, 2024, the Federal Reserve released the severely adverse scenario and the global market shock (GMS) component that will be used to calculate the stress capital charge imposed on covered banks by the results of the supervisory stress tests. The Fed also announced four additional “exploratory” scenarios to evaluate the resilience of banks to funding stress combined with rising interest rates and a severe global recession. The results of these exploratory scenarios will not be used to calculate the stress capital charge.

The severely adverse scenario is designed to assess a bank’s ability to withstand a severe macroeconomic recession and to set a bank-specific capital buffer; the GMS is imposed on banks with significant trading operations and results in additional losses that feed into the bank’s stress capital charge. This charge is determined by the decline in each bank’s common equity tier 1 capital ratio under the severely adverse scenario, including the market shock if applicable. Therefore, a larger decline in a bank’s capital ratio results in a higher stress capital charge. Failure to maintain the required level of minimum capital, including the stress capital charge, severely restricts a bank’s ability to distribute capital to shareholders.

In aggregate, the 2024 stress scenario appears to be somewhat more severe than last year’s scenario, especially with respect to the assumed trajectory of equity prices and corporate bond spreads. In addition, some of the GMS risk factors, such as the decline in equity prices and the rise in Treasury rates, are more severe than last year’s scenario. While many banks begin the exercise with a robust level of net interest income, potentially mitigating the heightened stress severity, the initial higher level of noninterest expenses could offset the anticipated improvement in pre-provision net revenue in this year’s stress tests. As was the case last year, banks are expected to see an increase in the fair value of their available-for-sale securities driven by the decrease in interest rates in the severely adverse scenario. 

photo showing graph

Overall, we project that this year’s stress test will result in a slightly higher reduction in projected bank capital ratios than last year’s stress test for banks in Categories I through IV. The greater severity of market shocks and a rise in provisions for loan and lease losses drive the higher aggregate decline in banks’ capital ratios. It is important to note that because several Category IV firms did not participate in the 2023 stress tests as they are only required to participate biennially, the more relevant comparison would be with the results from the 2022 stress tests.

One unique aspect of this year’s stress tests is the inclusion of the FDIC special assessment imposed in the fourth quarter of 2023. The models employed by the Federal Reserve often extend past data into future projections. Without an adjustment for the one-time FDIC charge in initial noninterest expenses for banks, bank capital requirements could be overstated. It would therefore be more reasonable for the Federal Reserve to exclude this FDIC charge from noninterest expenses at the start of this year’s stress tests when generating the projections of bank performance over the nine-quarter planning horizon.

This year’s stress test features four exploratory scenarios–two macroeconomic scenarios and two market shocks. One of the alternative macroeconomic scenarios assumes a severe global recession coupled with higher interest rates. In addition, the scenario requires banks to increase their reliance on wholesale funding and pay market interest rates on a greater portion of their liabilities. As a result, banks’ revenues under stress will decline, and the increase in interest rates will lead to further unrealized losses on available-for-sale securities. Our projections suggest that the decreases in the common equity tier 1 capital ratio under this scenario, particularly for Category I banks, could equal or exceed those expected under the severely adverse scenario.

The severity of the 2024 severely adverse scenario is comparable to 2023’s . . .

This year’s severely adverse scenario includes, on a start-to-stress basis:

  • A 6.3-percentage-point increase in the unemployment rate (6.4 p.p. in 2023).
  • A 40 percent drop in commercial real estate (CRE) prices (40 percent in 2023).
  • A 4.1-percentage-point increase in corporate BBB spreads (3.6 p.p. in 2023).
  • A 55 percent drop in the stock market (45 percent in 2023).
  • An 8.2 percent fall in real GDP (8.75 percent in 2023).
  • A 36 percent decline in house prices (38 percent in 2023).
  • A 70 peak value in the volatility index (75 in 2023).

The 2024 severely adverse scenario, mirroring that of 2023, features a severe global recession characterized by significant declines in both commercial and residential real estate prices. This downturn spills over into the corporate sector, affecting investment sentiment. The severely adverse scenario also assumes that decreases in CRE prices will be concentrated in properties most vulnerable to a prolonged decline in income and asset values, such as commercial office spaces potentially harmed by the shift toward remote work.

photo showing graph

Exhibit 2 shows four macroeconomic variables that are key drivers of bank performance in the stress tests, each with a severity level that roughly matches or exceeds that of last year: the unemployment rate, the CRE price index, the stock market index and the BBB corporate bond spread. As shown in the top left panel of Exhibit 2, the increase in the unemployment rate under the 2024 severely adverse scenario is about the same as in the 2023 scenario. The top-right panel shows a severe fall in CRE prices, similar to last year’s severely adverse scenario. The fall in equity prices in the 2024 scenario is more pronounced relative to the 2023 severely adverse scenario (lower-left panel). Finally, the BBB spread widens from 1.5 to 5.8 percentage points, marking an increase in severity from last year (lower-right panel).

photo showing graph

Exhibit 3 shows the path of key variables that drive the projections of pre-provision net revenue in the Federal Reserve’s stress testing models.[1] In the severely adverse scenario, the lower term spread—the difference between the yields on 10-year and three-month Treasury securities—creates more of a headwind for pre-provision net revenue in this year’s stress test. Nonetheless, the performance of banks in the year preceding the start of the stress test planning horizon significantly affects PPNR projections. Notably, the net interest margin at the jump-off date of this year’s stress tests is substantially higher than at the beginning of last year’s, as shown in the right panel of Exhibit 3. This increase leads to higher net interest income projections that more than offset the negative impact of the lower term spread throughout the projection horizon.

Exhibits 4 compares the trajectories of macroeconomic indicators that affect bank performance across the severely adverse scenario and the two exploratory scenarios in this year’s stress tests. The chart shows that the severity of the second exploratory scenario (Exploratory B) is almost on par with that of the severely adverse scenario. As shown in the top-left panel, the unemployment rate in the Exploratory B scenario reaches a peak of 10 percent in the fourth quarter of 2025, just one quarter later than in the severely adverse scenario. Real GDP declines 7.8 percent from the fourth quarter of 2023 to its lowest point in the second quarter of 2025, also delayed by a quarter. Moreover, the significant drop in CRE prices mirrors that of the severely adverse scenario.

photo showing graph

A significant distinction between the exploratory scenarios and the severely adverse scenario lies in the trajectory of interest rates. For example, the bottom-right panel in Exhibit 4 shows that the three-month rate in the second exploratory scenario (Exploratory B) reaches a peak of 5.9 percent in the third quarter of 2024, followed by a gradual decrease. In theory, this path of interest rates could further boost net interest income projections in the exploratory scenario. However, the assumption of funding stresses in the exploratory scenario will require banks to raise rates paid on noninterest-bearing deposits. The increase in interest rates also results in further losses on other comprehensive income.

The GMS seems more severe compared with last year’s test. . . The two exploratory market shocks require banks to simulate defaults of their five largest hedge fund exposures.

Banks with significant trading operations are subject to the GMS component in the stress tests. The GMS consists of thousands of large movements in market prices and rates, which are generally calibrated to extreme market moves seen in the second half of 2008 and the period following the Global Financial Crisis. This year’s severely adverse scenario also includes two exploratory market shocks (EMS) to evaluate the trading risk profiles of banks against a broader spectrum of risks.

photo showing graph

Exhibit 5 shows the shocks to a set of risk factors across various important asset classes, such as corporate bond spreads, mortgage-backed securities spreads, commodity prices, interest rates and the S&P 500 Index. The GMS scenario for 2024 differs from last year’s, particularly regarding the shocks to interest rates and commodity prices. In addition, this year’s spot shock to equities is more severe than last year. Large banks with substantial trading and custodial operations are also required to incorporate the default of their largest counterparty. 

The exploratory market shocks introduce different shocks to interest rates and commodity prices and seem to be designed to justify the assumed increase in inflation and assess banks’ sensitivities to a wider range of risks. The remaining shocks are identical to those in the GMS under the severely adverse scenario. Another novel feature of the exploratory market shocks is that both require banks to assume defaults of their five largest hedge fund exposures, representing a significant deviation from the current standard.

Loan losses and provisions are projected to increase slightly. . .

Merely observing the paths of the macroeconomic variables over the stress horizon offers only a partial view of how these scenarios affect bank performance. To establish a more direct link between these macroeconomic variables and bank-level performance, we employ top-down, time-series models. Based on BPI’s top-down models, we estimate the impact of the severely adverse scenario on the projections of loan losses, provisions and PPNR under the supervisory stress tests. In addition, we also present projections for provisions and PPNR under the exploratory scenarios.

Overall, we expect loan loss projections under the 2024 stress scenarios to be slightly higher than last year, as well as significantly higher than the 2020 stress tests before COVID. Total estimated loan losses for the 32 firms participating in this year’s stress tests are projected to reach $542 billion, about $33 billion more than in 2023 for the same group of banks. This increase in loan losses is primarily attributed to losses pertaining to CRE and credit card loans. Total losses are about $120 billion higher compared with the severely adverse scenario in the 2020 stress tests.

Exhibit 6 plots the projected loan loss rates over the past four stress testing exercises. The aggregate loss rate is projected to remain approximately unchanged at 6.4 percent. The loss rate for CRE loans is expected to increase compared with last year’s test, mainly due to an increase in CRE loss rates at launch, likely reflecting a slight deterioration in the portfolio of banks. In contrast, the loss rate for commercial and industrial (C&I) loans is predicted to remain unchanged at 6.7 percent due to a similar outlook for the unemployment rate and real GDP. The loss rate for residential real estate loans is anticipated to decline driven by the slightly less pronounced decline in house prices. The loss rates of other loan categories are expected to change little relative to 2023.

photo showing graph

Exhibit 7 illustrates that projected provisions are expected to increase cumulatively by about $28 billion more than 2023 over the nine quarters of the stress planning horizon. Provisions for Category II and III banks remain largely unchanged, while those for Category I banks are set to increase by $20 billion and for Category IV firms by $6 billion. We also project that provisions for loan and lease losses will be lower than total losses because the level of allowances for credit losses at the start of the stress tests remains elevated.

photo showing graph

In the two exploratory scenarios, the projected provisions for loan and lease losses are lower than those in the severely adverse scenario, with the difference being relatively modest in the more severe exploratory scenario. Overall, when compared with the severely adverse scenario, the projected provisions for loan and lease losses are $39 billion lower in the second exploratory scenario and $214 billion lower in the first.

More than offsetting the increase in provisions is a projected increase in PPNR . . .

Overall, PPNR is expected to remain little changed relative to the 2023 stress test, but the final outcome depends on whether the Federal Reserve adjusts banks’ noninterest expenses to reflect the FDIC’s special assessment. In aggregate, the 32 firms that participate in this year’s stress tests are projected to generate $435 billion in net revenues over the nine quarters of the planning horizon. The recent strong performance of net interest margins for banks continues over the stress horizon, thanks to the backward-looking component of the Fed’s models. This contributes to higher projections for net interest income in this year’s stress tests, even though term spreads are lower. The projections of noninterest income also increase, but noninterest expenses are also projected to rise considerably more, in part due to the higher FDIC charges in the fourth quarter of 2023.

photo showing graph

Under the two exploratory scenarios, projections for PPNR would be initially higher compared with the severely adverse scenario due to the higher path of interest rates. However, the funding stress add-on shock significantly reduces these projections. These scenarios simulate funding stresses where banks must increase the interest rates paid on deposits to preserve their deposit levels. The Federal Reserve’s exploratory scenario description states that 20 percent of noninterest-bearing deposits will transition to time deposits, which results in banks paying market rates on a larger share of their own liabilities.

This funding shock is illustrated in Exhibit 8. For instance, in the first exploratory scenario, we anticipate a $75 billion reduction in PPNR over the nine-quarter planning horizon, attributed to 20 percent of noninterest-bearing deposits shifting to time deposits. This PPNR decrease is distributed as $52 billion for Category I banks, $12 billion for Category II and III banks and $10 billion for Category IV banks. In the second exploratory scenario, we predict a $55 billion decline in PPNR due to the funding stress add-on. The lesser impact in this scenario is attributable to the comparatively lower path of interest rates.

photo of exhibit 9-noninterest expense

Similar to revenue projections, the forecasts of noninterest expense are affected by historical data. Exhibit 9 shows that aggregate noninterest expenses surged in the fourth quarter of 2023, primarily due to the inclusion of the FDIC special assessment. Our top-down models indicate that, over the nine quarters, projected noninterest expenses are about $73 billion higher compared with last year’s stress tests. Without knowing the exact formulation of the Federal Reserve models, accurately estimating the impact of the FDIC special assessment on noninterest expenses is challenging. However, it is expected to cause a further decrease in capital ratios over the planning horizon. To prevent unnecessary increases in banks’ capital requirements, the Federal Reserve should exclude the FDIC special assessment from noninterest expenses for the fourth quarter of 2023.

The projected peak decline in capital ratios is modestly lower than in 2023. . . and the second exploratory scenario could produce greater capital drawdowns for universal banks and custodians.

Our analysis employs the projections described here to estimate the impact of the severely adverse scenario on the peak decline in each bank’s CET1 capital ratio under the supervisory stress test. In making these estimates, we must make further assumptions about other significant components of the stress tests, such as trading and counterparty losses, operational risk losses and changes in accumulated other comprehensive income. For the purposes of this analysis, we assume the following:

  • Operational risk losses remain unchanged from last year across all three scenarios.
  • Trading and counterparty losses increase 10 percent in the severely adverse scenario and the first exploratory scenario. There are no changes in trading and counterparty losses in the second exploratory scenario compared with 2023.
  • Unrealized gains on available-for-sale securities are unchanged due to the similar trajectory of interest rates in this year’s scenario. However, we anticipate an increase in unrealized losses on available-for-sale securities in the second exploratory scenario, while expecting little gains or losses in the first exploratory scenario.

Exhibit 1 presents the projected aggregate decrease in CET1 capital ratios, broken down into Category I, II–III, IV firms, and all banks combined. In the 2024 stress tests, banks within Categories I through IV are expected to experience higher capital drawdowns in the stress tests. It is important to note that, as several Category IV firms did not participate in the 2023 stress tests, the more relevant comparison would be with the results from the 2022 stress tests.

Under the exploratory scenarios, the projected decrease in capital ratios could lead to capital reductions for Category I firms that are similar to or even exceed those seen in the severely adverse scenario. This outcome is attributed to the second exploratory scenario’s dual effect of lowering net interest income through the funding stress add-on and higher unrealized losses on available-for-sale securities for these banks due to the higher path of interest rates. For the time being, further increases in interest rates in the exploratory macroeconomic scenarios will not affect Category III and IV banks regulatory capital ratios through unrealized gains and losses on available-for-sale securities. That will most likely change when the Basel proposal is finalized.

Another element of uncertainty in the exploratory scenarios concerns the losses from the default of the five largest hedge funds that banks will have to assume under the market shocks. It is unclear whether this would result in greater or lesser trading and counterparty losses compared with the severely adverse scenario.

Conclusion

According to our models, this year’s severely adverse scenario may result in increases in capital requirements for large U.S. banks on an aggregate level. The heightened severity of market shocks and a slight increase in provisions for loan and lease losses are expected to be balanced out by higher pre-provision net revenue projections. However, the Federal Reserve should exclude FDIC charges from noninterest expenses at the outset of this year’s stress tests to prevent any undue influence on banks’ projections over the nine-quarter projection horizon. This adjustment could leave the aggregate decline in capital ratios little changed compared with 2023.

Our examination of the exploratory scenarios suggests that the second exploratory macroeconomic scenario might lead to larger reductions in banks’ CET1 capital ratios compared to the severely adverse scenario. There is some uncertainty about this, because we lack all the information necessary to anticipate all adjustments the Federal Reserve might make. However, the difference in severity between the severely adverse and the second exploratory scenarios is small. The inclusion of funding stresses and higher unrealized losses on available-for-sale securities could result in greater capital drawdowns in one of the exploratory scenarios. The impact of the default of the five largest hedge fund exposures, as opposed to the current assumption about the default of the largest counterparty, also remains uncertain.


[1] Our models also include stock market returns and the change in market volatility.