Federal reserve board publishes 2020 stress testing results and additional sensitivity analysis

The Federal Reserve Board released stress test results for DFAST 2020 including additional sensitivity analysis, considering the COVID19 outbreak, to assess the resiliency of large banks under three hypothetical recessions, or downside scenarios, that could result from the coronavirus event. Furthermore, the Board provides guidance for large banks to maintain resiliency during economic uncertainties from the current context going forward.

On June 25, 2020, the Federal Reserve Board disclosed annual stress test results, implemented pursuant to the Dodd-Frank Act (DFAST 2020), of Banking Holding Companies (BHCs) with $100 billion or more in total consolidated assets and U.S Intermediate Holding Company’s (IHCs) to observe if firms are sufficiently capitalized to absorb losses during stressful conditions while meeting obligations to creditors and counterparties over a 9 projected quarter horizon. Below includes a number of key changes to DFAST 2020 compared to years past.

Results of DFAST 2020, in aggregate, show the 33 firms subject to the supervisory stress test would experience substantial losses under the severely adverse scenario but could continue lending to businesses and households, due to the accretion of capital since financial crisis. The stress tests are based on two supervisory scenarios, baseline and severely adverse, released by the Federal Reserve on February 6, 2020. Supervisory scenarios include trajectories for 28 variables that capture economic activity, assets prices, and interest rates in the U.S economy as well as macroeconomic variables such as GDP growth, inflations, and U.S. foreign currency exchange rate. The Severely Adverse scenario is characterized by a severe global recession accompanied by a period of heighted stress in commercial real estate and corporate debt markets. Similar to DFAST 2019, the Federal Reserve applies a global market shock to the trading portfolio of 11 firms with trading and private equity exposures and a counterparty default scenario component to 13 firms with substantial trading, processing, or custodial operations.

Projected aggregate losses of the 33 firms subject to DFAST 2020 would amount to $552 billion in losses under the severely adverse scenario. Of the 18 firms where stress test results were to be disclosed both last year and this year, total losses increased 5% in DFAST 2020 to $443 billion (80% of all DFAST 2020 losses) versus 2019. Aggregate loan losses as a percentage of loan balances were similar in DFAST 2020 to years past, even with higher loss rates. A considerable amount of the 2020 projected loan losses are concentrated across credit cards, commercial and industrial loans, CRE, and other consumer loans (student loans and automobile loans). Aggregate projected pre-provision net revenue (PPNR) in DFAST 2020 under the severely adverse scenario total $430 billion for the 33 firms.

*Excludes common distributions. Sample consists of 33 firms participating in DFAST 2020.

*To reduce uncertainty, the provision of loan and leases is the calculated under the old methodology used in years past and does not use the current expected loss methodology (CECL).

The Board did not object to the five foreign banks whose capital planning practices were evaluated as part of the stress tests, who are still subject to the public disclosure of a non-objection to the qualitative assessment of capital planning. The Board will use results to set new stress capital buffer requirement for large firms, which will go into effect, as planned, in Q4 2020.

Additional Sensitivity Analysis

Due to the deterioration of the economy due to the coronavirus outbreak and response, the Board conducted additional sensitivity analysis to assess the impact and resiliency of large banks. The sensitivity testing included three hypothetical “downside” scenarios beyond the severely adverse scenarios published in February 2020:

• V-shaped recession – Rapid recovery with much of the output and employment lost by the end of the year
• U-shaped recession– Slower recovery with only a small share of lost output and employment is regained in 2020
• W-shaped “double-dip” recession – Short lived recovery followed by a severe drop-in activity later this year due to a resurgence of another COVID event

Results of the sensitivity analysis show aggregate loan losses for 33 banks ranged between $560 billion and $700 billion and aggregate capital ratios declining from 12% in Q4 2019 to between 9.5% and 7.7% across the hypothetical scenarios. Results of V-shaped downside scenario are comparable to the full stress, where all firms remained well capitalized. Under the V-shaped the U and W – shaped scenarios, a majority of firms remain well capitalized, however several firms would approach minimum capital levels.

While the sensitivity analysis does not constitute a full stress test and includes conservative assumptions, it demonstrates a wide range of possible outcomes. Scenarios were developed in early April 2020 and while certain economic and financial market indicators have improved since then, the path of the economy going forward is uncertain. Under the three downside scenarios, the unemployment rate saw peaks between 15.6% and 19.5% that are much higher and much more stringent than any of the Board’s pre-coronavirus stress test scenarios. For context, the COVID event saw unemployment reach peaks as high as 14.7% (post-World War II high) and U.S GDP decreasing at an annual rate of 4.8% for Q1 2020 as equity prices plunged with surging volatility and nominal treasury and longer-term security yields dropped as corporate bond spreads widened. By mid-March 2020, it became evident the COVID event was disrupting the U.S economic activity with more extreme downside outcomes, particularly for near-term unemployment and gross domestic product (GDP) when compared to the severely adverse scenario. The severely adverse scenario features a severe global recession, a rise in unemployment-rate to 10%, 50% decline in broad equity prices, roughly 25% decline in residential house prices, and heighted stress in corporate debt markets and commercial real estate.

Scenario variables in severely adverse and alternative downside scenarios

The scenarios are hypothetical and are not predicators of the likely path of the economy or financial markets in the current environment as well do not incorporate the potential effects of government stimulus payments and expanded unemployment insurance. This includes the effects of assistance businesses through the Payment Protection Program, direct payments from the CARES Act, and the Federal Reserves’ lending facilities not included in the sensitivity analysis. Furthermore, the analysis does not account for capital depletion resulting from common equity distributions over the projection horizons.

Following the results, the Board is taking several actions to ensure large banks remain resilient, which includes:

• No share repurchases will be permitted in Q3 2020. Historically, share repurchases have represented approximately 70% of shareholder payouts from large banks.

• Dividend payments will be capped to the amount paid in the Q2 2020 and further limited based on recent earnings. Banks cannot increase dividends and can pay dividends only if it has earned sufficient income.

• All banks will be required to re-submit and update their capital plans in the second half of the 2020 to reflect the current stresses, where the Board will conduct analysis each quarter to determine if adjustments are reasonable. This will assist firms in re-assessing their capital needs, re-evaluate long term capital plans, and maintain strong capital planning practices.

The restrictions will apply for the third quarter of 2020 and may be extended by the Board on a quarter-by-quarter basis, contingent on the economic situation.

By Arthur Midianga.

[1] Capital ratios are calculated using capital action assumptions provided within the Dodd-Frank Act stress testing rule

[2] Excludes common distributions. Sample consists of 33 firms participating in DFAST 2020. rb