Key Takeaways from the Fed’s 2016 Dodd-Frank Stress Tests

Dan Ryan is Leader of the Financial Services Advisory Practice at PricewaterhouseCoopers LLP. This post is based on a PwC publication by Mr. Ryan, Mike Alix, Adam Gilbert, and Armen Meyer.

Large banks will be less constrained in returning capital to shareholders based on this year’s Dodd-Frank Act Stress Test (DFAST). The DFAST results published last Thursday are the Federal Reserve’s (Fed) first stress test results released in 2016. On June 29th, the Fed will release the more important Comprehensive Capital Analysis and Review (CCAR) results. Those will indicate whether the banks passed both the Fed’s qualitative and quantitative assessments in order to return more capital to shareholders. [1]

1. As always, CCAR results are what matter most. All 33 institutions, including this year’s two new foreign bank joiners, exceeded each minimum capital ratio threshold under the Fed’s stress models (applying the banks’ existing dividend payout rate). However, this does not necessarily foreshadow success for all institutions with respect to CCAR. Still to be revealed are the results of the Fed’s CCAR qualitative assessment, which last year tripped up three banks that had adequate post-stress capital under DFAST and CCAR. [2] Given clarified supervisory guidance on capital planning published in December and January, [3] we could see some failures.

2. Biggest banks fare better, but smaller banks fare worse. This year’s results under the Fed’s severely adverse scenario showed lower minimum CET1 ratios for 19 of the 31 banks involved in last year’s DFAST. However, the minimum CET1 ratios for the eight US globally systemically important banks (G-SIBs) increased by an average of 131 bps from last year to this year. These large banks’ higher minimum ratios were predominantly driven by significantly improved stressed performance, with stronger starting capital ratios also helping in three cases. This improvement occurred even though this year’s severely adverse scenario included a deeper and more severe recession than last year’s (with unemployment peaking over 10% from a lower starting point). In contrast, the minimum CET1 ratios for the non-G-SIBs decreased by an average of 80 bps from last year to this year. A key factor fueling these disparate results between the G-SIBs and other banks is likely the projected negative short-term interest rates, which is unfavorable to more traditional lenders (as a result, non-GSIBs experienced on average a 50-bps decrease in the ratio of pre-tax net income over average assets as compared to last year).

3. No bank will fail CCAR on quantitative grounds. Since 2013, the Fed has privately offered banks an opportunity to revise downward their planned capital distributions after receiving their DFAST results (i.e., “take the mulligan”). Three banks took advantage of this opportunity last year to ensure they met all minimum capital ratios, resulting in no banks failing CCAR on quantitative grounds. [4] This week’s CCAR results will disclose which banks took the mulligan this year. Several smaller banks’ minimum DFAST ratios are low enough to suggest that one or two may have overshot on their capital plans in an effort to satisfy shareholders, and thus had to take the mulligan to pass CCAR’s quantitative test. More interestingly, even banks with high DFAST ratios may have strategically taken the mulligan by submitting large increases in planned capital distributions, expecting to simply revise them downward if needed. But given recent Fed guidance, [5] banks are unlikely to be as aggressive in employing this tactic as in prior years, in order to avoid triggering a CCAR failure for poor qualitative processes.

4. Commercial real estate turns the corner. For the first time since the Fed began disclosing its supervisory stress test results in 2012, the Fed’s projected loss rate for commercial real estate (CRE) loans declined versus the prior year, improving by 160 basis points. This result contrasts with the loss rates for various other types of loans (i.e., 2nd-lien mortgages, commercial and industrial loans, and credit cards) which all increased for the first time since stress testing disclosures began. The Fed did not indicate any changes to its loan loss models that would explain this apparent anomaly, so we believe the lower loss rates can be attributed to higher quality portfolios as legacy problem CRE assets roll off balance sheets.

5. Quantitative test will get significantly tougher for big banks. Two major capital rules will be incorporated into DFAST and CCAR in the future. First, the supplementary leverage ratio (SLR) will apply to banks with over $250 billion in assets starting in 2017. [6] The SLR will not just pose a new capital hurdle, but will also require modeling and operational capabilities that the banks must build, test, and implement. Second, and more importantly, Fed Governor Daniel Tarullo recently confirmed that the G-SIB capital surcharge will be included in future stress testing, thus significantly increasing the minimum CET1 capital ratios that G-SIBs will have to meet. [7] G-SIBs can take some comfort in the fact that their DFAST CET1 minimum ratios are currently high enough to meet the threshold under recent surcharge estimates, but the surcharge will no doubt challenge capital distributions going forward and will prove to be a significant binding constraint if G-SIBs do not do as well under future Fed stress tests as they did this year.

Endnotes:

[1] The primary difference between DFAST and CCAR is that under CCAR the Fed assesses the quantitative impact of the bank’s capital plan and the quality of the bank’s qualitative capital planning processes. DFAST does not assess capital plans (instead assuming prior year dividends and employee stock issuances) and does not qualitatively assess processes.
(go back)

[2] See PwC’s First take, Ten key points from 2015 CCAR (March 2015), discussed on the Forum here.
(go back)

[3] See PwC’s First take, Fed’s supervisory assessment of capital planning and positions (December 2015) and PwC’s First take, Fed’s 2016 CCAR summary instructions and supervisory scenarios (February 2016), discussed on the Forum here.
(go back)

[4] See First take cited note 2 for more information on last year’s results. These three G-SIBs have significantly more post-stress capital this year to distribute more capital.
(go back)

[5] See First takes cited note 3.
(go back)

[6] See PwC’s First take, Key points from the final US supplementary leverage ratio (September 2014). Notably, the enhanced SLR, which adds a 2% surcharge for G-SIBs on top of the 3% SLR, will not be incorporated in 2017, although it may be included in the future given Governor Tarullo’s comments regarding incorporating the G-SIB capital surcharge.
(go back)

[7] See PwC’s First take, Key points from the Fed’s final G-SIB surcharge rule (July 2015), discussed on the Forum here. In a recent Bloomberg interview, Governor Tarullo confirmed that the most stringent version of the G-SIB capital surcharge will be added in full to G-SIBs’ post-stress minimum CET1 ratios. This addition will be somewhat offset by allowing G-SIBs to adjust their capital distributions as a severe economic downturn unfolds. Since this change will likely be implemented by rulemaking, it is unlikely to come into effect before 2018 given the time needed to propose and finalize a rule.
(go back)

Both comments and trackbacks are currently closed.