The largest U.S. banks reported weak first-quarter 2020 earnings, weighed down by large provisions, foreboding the economic pain that is likely ahead. Outside of a hefty buildup in allowance for loan losses, most of the other earnings measures held up reasonably well in the quarter. For example, revenue for the median large banks was up 3.6% because of exceptionally high trading results and two months of the quarter (January and February) that were devoid of pandemic concerns.

The main reason for the banks' large allowance build in the first quarter is the weighting they placed on the more dour economic scenario that lies ahead. This was exacerbated by the adoption of the Current Expected Credit Losses (CECL) accounting methodology, in which banks set aside allowances for lifetime loss expectations.

The need for a substantially larger allowance buildup without commensurate offsetting earnings would likely lead to negative rating actions for the more affected large banks. In this regard, S&P Global Ratings looked at several metrics to gauge the level of allowance build so far, and the ability of large banks' capital to withstand stress if losses were to increase more than bank management teams currently expect. At this time, all of the largest banks have suspended share repurchases, which will help preserve capital. But if allowance levels need to rise substantially from here, capital ratios for some of the banks could decline meaningfully, and downgrades could ensue.

All in all, median earnings for the large banks declined 33% in the first quarter--with some banks experiencing much steeper declines than that--and will likely be under pressure in at least the second quarter as well. One bright spot in the quarter was liquidity. Despite large corporate draws, liquidity for most banks improved, aided by significant deposit inflows on the heels of the Fed's massive quantitative easing measures.

What Could Lead To Further Negative Rating Actions

Supporting the largest banks is that they entered the current crisis in a position of strength, from both a balance sheet and profitability perspective. But their ability to withstand stress is not limitless, and negative outlooks and downgrades could ensue if economic conditions worsen more than our economists currently expect. Specifically, our economist are forecasting U.S. GDP will decline 5.2% in 2020 (changed from their previous forecast of GDP down 1.3%) and then rebound 6.2% in 2021. They believe unemployment could reach 19% in May before declining to 8.8% by year-end and 6.7% in 2021. Our economists also acknowledge a high level of uncertainty in these forecasts at this time and see risks to the downside.

To assess a tipping point of when we may change the outlooks on the largest banks, we recently conducted a stress test (see "Stress Scenarios Show How U.S. Bank Ratings Could Change Amid Pandemic-Induced Financial Uncertainty," March 24, 2020). We concluded that if our severely adverse stress scenario assumptions were to materialize (see Appendix A for our assumptions), and we don't expect a sharp recovery in the economy, this could result in a larger number of negative bank rating actions. We have updated the stress test to incorporate first-quarter results and made a few changes to tailor it to the makeup of individual banks' loan portfolios.

So far, we have taken negative rating actions related to COVID-19 on 30 banks, all of them outlook revisions (see Related Research). All of the large banks we rate--except for two--have stable outlooks. (Our outlook on Wells Fargo was negative before the pandemic, and we recently revised our outlook on Capital One to negative because of COVID-19-related credit issues.) The stable outlooks on the large banks reflect our view that the economy will at least gradually rebound in the latter part of this year and in 2021, and that the largest banks' diversified business lines can help them withstand the stress we foresee.

Duration Of The Downturn And Pace Of Recovery Are Key Factors

The big question from here is what will provisioning look like for banks in the second quarter and beyond? It will depend on the assumptions bank managements made both about the length of the lockdown and the pace of the economic recovery in its aftermath, to derive their first-quarter allowances, versus any meaningful change in these assumptions as the year progresses. The effectiveness of fiscal stimulus and regulatory initiatives will also be determining factors that require time to run their course.

Notably, we recently revised our view of the economic risk trend for U.S. banks to negative--which we incorporate in our Banking Industry Country Risk Assessment (BICRA). In our view, despite the government's extraordinary measures to support the economy, substantial pressure on banks' earnings and asset quality could lead to steep credit losses and capital declines if the downturn is long lasting or the rebound is tepid. Still, even if the economic risk score of the U.S. BICRA declines to '4' from '3', this would not lead to a lowering of the anchor--the starting point for the ratings--for U.S. banks, which is currently 'bbb+'. (For more, see "Outlooks On 13 U.S. Banks Revised To Negative Due To Economic Downturn From COVID-19.")

Whether our severely adverse stress scenario materializes hinges on the economic reality versus our current economic forecast. A worsening of the economy relative to our projections may lead to bank provisions and charge-offs in line with, or perhaps even greater than, our severely adverse stress scenario assumptions. Conversely, a more benign reality (such as a speedier economic recovery than we currently project) suggests the odds of negative actions would be lower.

Using First-Quarter Allowance Levels To Gauge Relative Conservatism Among The Largest Banks

Incorporating CECL, all the banks built significant allowances in the first quarter. Despite the large buildup, there are a few metrics we use to assess how conservative the largest banks' allowance levels are relative to the amount of losses they may possibly face.

One way is to look at a bank's allowances at the end of the first quarter, as a percentage of the Fed's assumed loan losses, as determined by the 2019 Dodd-Frank Act Stress Test (DFAST). In DFAST, the Fed bases its loss rate assumptions on both the composition of a bank's loan book (for example, a higher credit card portfolio will likely equate to more losses) and the bank's underwriting standards on its loans. A higher percentage of first-quarter allowances for the Fed's nine-quarter severely adverse loan loss assumption would reflect greater bank conservatism.

From this perspective, the median bank allowance for credit losses (including both funded and unfunded exposures) versus the Fed's DFAST loan loss assumption is 37%. Citigroup, at 48%, had the highest percentage, and the trust banks and Wells Fargo were at the lower end, below 30%. Still, when banks run their own versions of the Fed's stress test, some of them forecast lower loss rates than the Fed. That could be one of the reasons some banks percentages' of allowances for DFAST losses are lower than peers.

Another way to measure relative conservatism is to look at the buildup in allowance for loan losses, excluding the allowance impact on Jan. 1 that resulted from banks incorporating CECL, which assesses lifetime loan losses for a bank's portfolio. By excluding the Jan. 1 allowance build, we can hone in on the amount of provisions, less net charge-offs, banks took--with most of it directly related to COVID-19 economic deterioration. (In general, the larger the allowance build, the more conservative the approach. Though we would also consider the composition of the loan portfolio--e.g., a bigger percentage of credit cards in a portfolio would need more reserves--and relative underwriting standards.)

A third way to assess the adequacy of allowances versus peers is to look at a bank's tangible common equity (TCE) plus first-quarter ending allowance as a percentage of its risk-weighted assets. This ratio reflects the amount of reserves a bank has taken while also considering its capital levels. We opted to use TCE, rather than common equity Tier 1 (CET1), to exclude the leniency regulators have allowed the banks in calculating their CET1 regulatory risk ratios. In the first-quarter CET1 ratios, regulators have neutralized a portion of banks' allowance build as it pertains to their capital ratios (see Appendix A for more details).

In all of these metrics, the composition of a bank's loan book, perhaps more than management conservatism, is a major factor in determining the allowance that banks have taken so far. That's because CECL is, in general, more punitive for consumer loans than commercial loans, due in part to consumer loans often having longer contractual maturities than commercial loans.

Said another way, if a bank has a higher percentage of consumer loans--particularly credit card loans--its allowance build under CECL will likely be higher because of how CECL requires allowances to be taken. To accentuate how this factors into the allowance, we have reflected the amount of consumer loans and credit card loans (see table 1).

Table 1

Deploying Allowance Analysis To Determine Rating Resilience

We also expect to incorporate allowance analysis to determine whether rating actions may be warranted. The variables we will monitor are:

• Allowance build as a percentage of DFAST assumed loan losses (the higher the percentage, the greater the likelihood that banks are approaching our severely adverse stress scenario);
• The degree of conservativeness in the economic scenarios each bank uses to set its allowance for credit losses (when that disclosure is available);
• Changes to our economic outlook over the coming two years, including our projection of GDP and unemployment and how long we expect these metrics to remain weak; and
• An individual bank's business model and its ability to withstand the economic stresses we foresee.

Another factor we will use to determine whether rating actions are warranted is banks' preprovision net revenue (PPNR), which helps offset capital decline as allowances build. In the first quarter, PPNR handily exceeded our stressed assumptions because two months in the quarter experienced no stress.

But, in the quarters to come, we expect PPNR to decline. If we were to believe PPNR would systematically fall below our stress assumptions (on a sustained basis), that could lead to negative rating actions as well because it would likely result in a significant decline in our projected risk-adjusted capital (RAC) ratios for the banks, even assuming that share repurchases remain suspended.

Updating Our Stress Analysis Based On First-Quarter Results

To help gauge how banks fared relative to the stress tests we previously ran, we reexamined allowance build in the aftermath of first-quarter results. We made a few changes to our previous stress test assumptions.

We refined individual bank loss rate assumptions based on their performance in the Fed's 2019 stress test. Previously, we had assumed a loss rate of 2.5%, absorbed in a single year, for most banks we tested, and higher loss rates for banks engaged in riskier lending, such as credit cards. The Fed's assumed loss rates are, on average, double our 2.5% assumption but occur over nine quarters rather than just a year. When annualizing these rates, most of them are in line with our previous 2.5% assumption, but a few banks differed (see table 5 for each bank's assumed annualized loan rate). The Fed's 2019 severely adverse stress assumptions have some similarities to the current stress, including unemployment climbing to 10% at its peak and GDP falling 8% from its prerecession peak, but the Fed's stress goes on longer than our economists' current assumptions (see table 2).

Table 2

Also, the stress test we previously conducted was prior to the reporting of first-quarter results, meaning we did not have clarity on the impact of CECL on bank allowances and capital. We now know ending first-quarter allowances and can estimate the neutralization of CECL-driven allowances regulators provided to banks regarding regulatory capital ratios. We estimate it amounted to roughly 30 bps in the first quarter but varied depending on the amount of the allowance taken. Specifically, regulators put forth an interim final rule allowing banks to neutralize the impact of CECL, versus the incurred loss model, by allowing them to add back the estimated difference to regulatory capital (completely for two years and partially during the three-year transition).

We are not incorporating further capital neutralization when projecting banks' CET1 ratios under stress, so as to be conservative, but we do provide the impact of the neutralization in table 3. (In table 3, it's shown as the difference between the "Ending nine-quarter CET1 ratio with Q1 deferral" column and the "Stress CET1 ratio decline excluding deferral" column.)

According to our stress analysis, the median large bank CET1 ratio declines by about 80 bps at the end of the nine-quarter stress period starting in second-quarter 2020 (see table 3) when incorporating the stress CET1 ratio deferral due to CECL. Excluding the regulatory benefit of the CET1 ratio deferral due to incorporating CECL, the median large bank CET1 ratio declines by about 110 bps at the end of the nine-quarter stress period. The capital decline is larger for some banks during the nine quarters.

Table 3

Factors That Could Worsen Our Stress Test

The median CET1 ratio decline, after nine quarters of stress, of 80 bps (110 bps excluding CECL deferral) from banks' current capital levels seems manageable. But, if we see signs that our severely adverse stress scenario will materialize, it could result in negative rating actions for most of the largest banks. That's largely because at that juncture, we would not know with any certainty whether losses will be higher than those in our severely adverse stress assumptions.

Factors that could lead to results being worse than our stress assumption are:

Steeper and sustained economic decline

Should GDP and unemployment deteriorate more than envisaged in the Fed's severely adverse stress scenario (e.g., unemployment climbing to 10% at its peak and GDP falling 8% from its prerecession peak), and remain at these levels longer than our economists currently project, the loan loss rates and PPNR figures we used in our stress test could prove too optimistic.

Extended economic decline

The longer the economy stays in decline, the higher the loan loss rates will likely be. An extended U-shaped or L-shaped recovery could test the limits of the fiscal stimulus relief provided through the CARES Act (for example, additional unemployment insurance or forgivable Small Business Administration loans).

Trading losses and counterparty defaults

Our stress test does not incorporate any trading losses or counterparty defaults. Specifically, the construct of our stress test is especially hard hitting for banks that rely heavily on loans and net interest income and have limited business activity outside of these areas.

Related Research

• Outlooks On 13 U.S. Banks Revised To Negative Due To Economic Downturn From COVID-19, May 4, 2020
• Outlooks Revised To Negative On Several Spanish Banks On Deepening COVID-19 Downside Risks, April 29, 2020
• Outlook On MUFG Americas Holdings Corp. Revised To Stable From Positive Following Outlook Revision On Parent, April 24, 2020
• BNP Paribas And Most Core Subsidiaries Outlooks To Negative On Higher Industry Risks Amid COVID-19; Ratings Affirmed, April 23, 2020
• COVID-19 Deals A Larger, Longer Hit To Global GDP, April 16, 2020
• Who The U.S. Government Plans Help, Who They Don't, And What That Means For Financial Institutions, April, 16, 2020
• Comparative Statistics: U.S. Banks (April 2020), April 13, 2020
• U.S. Financial Institutions Face A Rocky Road Despite A Boost From Government Measures, April 8, 2020
• Six U.S. Regional Banks Outlooks Revised To Negative On Higher Risks To Energy Industry, March 27, 2020
• Credit FAQ: Most U.S. Banks, Helped By Fed Actions, Are Well Positioned To Meet Corporate Borrowers' Demand For Cash, March 24, 2020
• Stress Scenarios Show How U.S. Bank Ratings Could Change Amid Pandemic-Induced Financial Uncertainty, March 24,2020
• IBERIABANK Outlook Revised To Negative, First Horizon National Corp. Ratings Removed From CreditWatch Positive, March 24, 2020
• Outlooks On Six Banks Revised To Stable From Positive On Emerging Economic Downturn; Ratings Affirmed, March 23, 2020
• Credit FAQ: The Fed's New Rules Change Capital Management Dynamics For U.S. Banks, March 19, 2020
• The Fed's Crisis Actions Will Further Bolster Liquidity For U.S. Banks, But Earnings And Asset Quality Are Set To Worsen Substantially, March 18, 2020
• COVID-19 And Falling Rates Cloud The Outlook For U.S. Financial Institutions, March 10, 2020

Appendix A: Stress Test Assumptions

Table 4

Table 5

Appendix B: Banks' First-Quarter Allowance Levels

In the first quarter, all of the largest U.S. banks posted significant provisions, building hefty allowances with varying degrees of conservatism. The main reason for this is all of the large banks decided to adopt CECL, in which allowances are set aside for lifetime loss expectations. According to statements bank management teams have made, the main determinants in building allowances in the first quarter were:

• The weightings management teams placed on various macroeconomic scenarios, including expected GDP declines and elevated unemployment;
• The length of time management teams expect GDP and unemployment to remain pressured; and
• Internal analysis of the creditworthiness of loan books and the likelihood of credit deterioration of these loans (such as energy, airline, hotel exposure).

On Jan. 1, due to the adoption of CECL, all of the banks took a one-time CECL allowance build, reflecting their view of lifetime losses under what was--at that time--a more benign macroeconomic scenario. The allowances taken did not flow through banks' income statements. Rather, they were reflected as a balance sheet entry, reducing retained earnings. At the end of the first quarter, the banks then took additional provisions, based large on the more dour economic outlooks, with some of it also attributable to loan growth. The additional provision, less net charge-offs in the quarter, amounts to an "allowance build"--allowances built largely for the pandemic-induced credit issues that lie ahead.

To sum up, banks' allowances at the end of the first quarter included:

• The previous level of allowances at year-end 2019 calculated under the incurred loan loss method;
• An additional allowance taken when they adopted CECL on Jan. 1; and
• An allowance build at the end of the first quarter.

The allowance build for CECL purposes does not have a one-for-one relationship with regulatory capital ratios--meaning regulatory capital ratios will not decline commensurate with the hefty allowance build. In March 2020, as the impact of the pandemic grew, regulators put forth an interim final rule, allowing banks to temporarily delay the effects of CECL on regulatory capital for two years, followed by a three-year transition period--to help support lending activity.

The interim final rule provides a uniform approach for estimating the effects of CECL on regulatory capital compared with what the impact would have been under the previous and less conservative incurred loss model. The rule, in effect, allows banks to negate the impact of CECL versus the incurred loss model by allowing them to add back the difference to regulatory capital (completely for two years and partially during the three-year transition).

First, banks can add back the day one impact CECL had on retained earnings (when it was implemented on Jan. 1, 2020) to regulatory capital. Second, they can assume that allowance build in 2020 and 2021 would be 25% lower under the incurred loss model than under CECL and add back the difference to regulatory capital. Those amounts added back to regulatory capital will be phased back in at 25% per year, beginning Jan. 1, 2022.

We estimate that this rule boosted banks' CET1 capital ratios by about 30 bps at the median in the first quarter (inclusive of the day one CECL impact and the provision taken in the quarter) and will continue to benefit the banks, probably at a slower pace per quarter, depending on the level of their provisions and net charge-offs going forward.

We will not incorporate such neutralization when calculating banks' RAC ratios. If the incurred loss model was being used instead of CECL, a similar allowance build would have occurred, just likely at a later time (probably a year later than CECL). As such, the gap between S&P Global Ratings RAC ratios and regulatory capital ratios will likely widen, particularly over the next two years, and could weigh on ratings. We acknowledge though that the deferral of CECL reserves will help keep banks' regulatory capital ratios above their minimum levels, which is a positive for creditors because it will help banks avoid headline risk.

Appendix C: Summary Of Individual Bank Results

Table 6

Table 7

Table 8

Table 9

Table 10

Table 11

First-quarter earnings comments

Bank of America Corp.'s   first-quarter earnings were hurt by an outsize credit provision because of the economic fallout from the COVID-19 pandemic, much like its large bank peers. Pretax income was $4.5 billion, down 48% from first-quarter 2019, primarily because of the $3.6 billion loan loss reserve build. Net interest income fell 2% year over year due to the adverse impact of rate cuts on mark-to-market loan portfolios as well as 18 basis points (bps) of margin compression. Loan balances were up 11% as a result of sizable drawdowns on commercial commitments.

Positively, higher market volatility and favorable market valuations cushioned earnings to some extent as they led to higher asset management and FICC (up 13%) and equity trading revenues (up 40%). Also, investment banking revenues reflected strong debt underwriting (up 27%) and equity underwriting revenues, offset by reduced advisory revenues. Credit quality remained stable compared with the previous quarter, although we expect losses to accelerate throughout the year. Despite a 41 bps decline in CET1 ratio from last quarter, we believe the capital ratios will remain above targeted minimums, supported by the announced suspension of share repurchases.

Analyst: Rian Pressman 

Bank of New York Mellon Corp. (BK)  posted net income of $944 million and a good pretax operating margin in the first quarter of 30% as fee revenue benefited from elevated volumes in most of its businesses, including foreign exchange trading, its Pershing clearing and custody business, and BK's clearing and collateral management segment. Otherwise, net interest revenue was down slightly, mainly reflecting lower rates, partially offset by higher loan and non-interest-bearing deposit balances.

Given the market turmoil and the influx of deposits, period-end assets grew 23% and deposits rose 15% during the quarter. Loans continue to account for less than 20% of assets, but they increased during the quarter, reflecting $3 billion in drawdowns from committed credit facilities and elevated overdraft loans from the market volatility. BK took an outsize $169 million loan loss provision. As a result of the turbulence in the short-term markets, BK purchased more than $3 billion in assets from money market funds as support to them, including those it manages. Illustrating BK's importance to the infrastructure of the banking system, the Federal Reserve activated the Primary Dealer Credit Facility through BK's triparty repo services.

BK's CET1 ratio (standardized approach) of 11.3% was down 20 bps, driven by payouts and growth in risk-weighted assets (RWAs). Pro forma for new regulatory rules as of April 1, the supplementary leverage ratio (SLR) was 7.6%, but the Tier 1 leverage ratio was down about 60 bps during the quarter to 6.0% because of asset growth, and this capital may continue to be pressured.

We expect BK's earnings to remain satisfactory for the rating over the next few quarters, but profitability will probably be somewhat constrained by possible lower average market values on assets under custody and administration (AUCA) and assets under management (AUM), potential money market waivers because of low rates, and net interest margin (NIM) compression.

Analyst: Barbara Duberstein 

Capital One Financial Corp.   posted a quarterly loss of $1.3 billion, primarily because of a $3.6 billion loan loss reserve build reflecting the expected economic fallout from the COVID-19 pandemic, as well as credit deterioration in its oil and gas portfolio. The largest portion of the reserve build was attributed to the card portfolio. In addition, earnings were hurt by a decline in noninterest income due to lower fee-based revenues across all business segments. Net interest income was sequentially flat and the NIM compressed 17 bps year over year, largely as a result of lower yields on earning assets. Nevertheless, preprovision net revenues were $3.5 billion, up 8% sequentially due to significantly lower operating and marketing expenses.

On balance, asset quality was satisfactory, although we expect losses to increase throughout the year. Capital ratios were stable and comfortably above targeted minimums, but we expect some deterioration in CET1 ratios as additional credit provisions are built and the impact of CECL adoption is gradually absorbed.

Analyst: Rian Pressman 

Citigroup Inc.  reported a 46% drop in first-quarter earnings from the prior year as its provisions for loan losses more than tripled following the onset of the coronavirus pandemic. That, along with the day one impact from CECL, drove Citi's allowance for loan losses up by about 70%, with the allowance equating to 2.9% of loans at the end of the quarter. The company reported a return on common equity of 5%. The surge in the provision more than offset strength in sales and trading, which led to a 36% rise in noninterest income.

Trading in both FICC and equities was up 38% compared with last year because of higher market volatility and activity. On the balance sheet, deposits rose a very robust 11% from the prior quarter due to the Fed's quantitative easing measures, clients drawing on revolvers and redepositing the funds in the bank, and other factors. Loans rose 3% from the prior year and quarter, mainly as a result of revolver draws on corporate loans. With that increase in the balance sheet and share repurchases in January and February, Citi's CET1 ratio declined to 11.2% from 11.8%.

Analyst: Brendan Browne 

The Goldman Sachs Group Inc.  reported a 50% year-over-year decrease in pretax income to $1.3 billion, mainly because of net losses in the asset management segment amid challenging market conditions and a decline in global equity prices, which offset strong performance in other business segments.

Within trading, FICC revenues were up 33% because of strong client activity in currencies, commodities, and rates partly offset by weakness in mortgages due to credit spread widening. Equities revenue increased 22% year over year, mainly reflecting strength in derivatives and commissions and fees due to higher volumes and client balances and equity market volatility. Within investment banking revenue, higher leveraged finance activity and an increase in investment-grade activity contributed to strong debt underwriting and equity underwriting trends reflecting favorable comparisons from a weak first quarter last year, whereas advisory fees fell because of a decline in industrywide transactions. Wealth management benefited from an increase in transaction volumes and investment management fees, reflecting higher average assets under supervision (up $219 billion due to inflows in liquidity products).

Higher net interest income stemming from growth in consumer credit was offset by a $600 million reserve build during the quarter. Most of the loan growth was attributed to the $19 billion in corporate drawdowns in investment banking. Still, liquidity remained solid, with global core liquid assets averaging $243 billion, up $6 billion versus the fourth quarter. Non-compensation costs increased 24% as a result of higher business volumes and litigation-related accruals in the quarter, while compensation expenses were essentially flat from first-quarter 2019. The Basel III CET1 ratio (advanced approach) was 12.3%, down 140 bps due largely to higher RWAs and credit spread volatility.

Analyst: Stuart Plesser 

JPMorgan Chase & Co. (JPM)  reported a nearly 70% decline in first-quarter earnings, mostly because its provision for credit losses rose by more than 4.5x, triggered by the coronavirus pandemic. With that provision and the day one CECL impact, the company increased its allowance for credit losses by 77% from year-end 2019, in particular because of credit card and wholesale lending. Preprovision profit also dropped, by 10%, due to mark-to-market losses in its bridge book and widening of credit spreads in derivatives. The company reported a 4% return on equity. Net interest income was flat year over year, but the NIM declined 20 bps, reflecting overall lower yields on earning assets.

Positively, sales and trading in fixed income and equity rose 34% and 28%, respectively. Investment banking fees rose 3% as a sharp drop in advisory revenue partially offset a strong quarter in debt and equity underwriting.

On the balance sheet, the company had an outsize 18% rise in deposits from the prior quarter related to the Fed's quantitative easing measures, clients drawing on revolvers and redepositing the funds in the bank, and other factors. Loans rose 6% from the prior year and quarter, mainly because of revolver draws on wholesale loans. With that increase in the balance sheet and share repurchases in January and February, JPM's CET1 ratio declined to 11.5% from 12.4%.

Analyst: Brendan Browne 

Morgan Stanley's (MS)  first-quarter earnings fell by 30% compared with the prior year, in large part on certain mark-to-market losses on loans held for sale, a significant rise in its provision for credit losses, and losses on investments related to employee deferred-cash based compensation plans. It also had significantly lower net interest income and transactional revenue in its wealth management business and a reversal of carried interest in the investment management business. Those items more than offset strength in sales and trading, which benefited from high volumes and volatility. Despite the drop, the company still reported a return on equity of 8.5%, which was stronger than most other banks.

MS continues to focus on expense management, with total expenses flat year over year. MS reported $13 billion in gross drawdowns, $5 billion in new corporate lending facilities, and $3 billion in retail commitments.

The advanced approach CET1 ratio declined to 15.3%, mainly because of growth in the balance sheet.

Analyst: Brendan Browne 

Northern Trust Corp. (NTRS)  posted solid results amid the market turbulence, reporting $326 million in net income and a 29% pretax operating margin. Versus the fourth quarter, revenue increased 7%, with noninterest income up 5%, driven by higher foreign exchange trading revenue and higher custody and administration and investment management fees that are based on lagged AUCA and AUM. Meanwhile, AUCA and AUM dropped 10% and 9%, respectively, during the quarter, which will probably pressure fee revenue in the second quarter. NTRS took a somewhat elevated $61 million credit loss provision under CECL, although we currently expect that asset quality will remain superior to most commercial banks in NTRS' moderate-size, high-quality loan portfolio. Because of clients' flight to quality, deposits were up sharply, and total assets grew 18% during the quarter.

Capital was lower but remained adequate. The CET1 ratio declined 100 bps to 11.7%, and the Tier 1 leverage ratio declined 60 bps to 8.1%, because of asset growth and $297 million in common share repurchases early in the quarter. We expect that NTRS will prudently manage its capital ratios during the crisis and will continue to suspend share buybacks.

Analyst: Barbara Duberstein 

PNC Financial Group Inc.'s  first-quarter results were hurt by a large reserve build, similar to its regional bank peers. Net earnings, excluding the $693 million reserve build, were fairly stable, and asset quality remained stable, although we expect deterioration for the remainder of the year. Preprovision net revenue was about $2 billion, up 7% from the prior quarter, primarily as a result of lower operating expenses. Net interest income was flat but the NIM expanded 6 bps, reflecting lower cost of funding. We expect pressure on the NIM due to lower anticipated yields on PNC's earning assets in 2020.

Nonperforming assets declined 7 bps to 0.66% of total loans and other real estate owned, while net charge-offs remained stable at 0.35% of average loans. The CET1 ratio marginally declined to 9.4% because of the impact of the phase-in of CECL, partially offset by opt-out of AOCI reporting. We expect further pressure on capital and earnings going into 2020 because of the economic fallout from the COVID-19 pandemic.

Analyst: Rian Pressman 

State Street Corp. (STT)  posted satisfactory first-quarter results, with net income of $580 million, as its pretax margin improved from recent quarters to 25%. Revenue was aided by a sharp rise in foreign exchange trading revenue amid the high market volatility, and asset servicing and investment fees held up well, while operating expenses were down 2% year over year. Still, AUCA was down 7% and AUM was down 14% from the fourth quarter because of the equity market drop, and these declines (taken into account on a lagged basis) will probably hurt fees somewhat in the coming quarters. Net interest revenue declined slightly, reflecting lower rates partly offset by higher loan and noninterest deposit balances. The NIM declined 24 bps to 1.30%.

Given the market turmoil and the influx of deposits, period-end deposits surged 41% because of a flight to quality from STT's asset manager clients. Elevated year-end assets were also driven by STT's key role in the Fed's Money Market Mutual Fund Liquidity Facility. STT has a small loan portfolio but needed to take a $36 million provision, mainly related to a leveraged loan portfolio.

The CET1 ratio (standardized approach) dropped 100 bps to 10.7%, and the Tier 1 leverage ratio declined 80 bps in the quarter to 6.1%, reflecting the bloated balance sheet and share repurchases early in the quarter. We expect that capital ratios will remain adequate for the rating, particularly assuming that STT will continue to suspend common share repurchases during the crisis.

Analyst: Barbara Duberstein 

Truist Financial Corp.   reported its first full quarter of post-merger earnings. Net income was $1.2 billion adjusted for aftertax one-time charges (merger and COVID-19) and included a $582 million loan loss reserve build. Noninterest earnings for the quarter were supported by higher insurance, residential mortgage, and wealth management revenues. The core NIM (excluding the impact of purchase accounting) was down 8 bps from last quarter to 3.06% because of continued pressure on earning asset yields, as well as higher cash balances maintained at the Fed. Loans and leases increased by 5%, primarily reflecting an $18 billion drawdown in commercial commitments. Asset quality metrics were stable, although we expect deterioration throughout the year. The CET1 ratio declined to 9.3%, primarily because of asset growth.

Analyst: Rian Pressman 

U.S. Bancorp (USB)  reported first-quarter results in line with the median of the largest banks, with earnings down roughly 32% versus last year. Still, return on average assets and return on average common equity were a decent 0.95% and 9.7%, respectively. The decline in earnings reflected 25 bps margin compression due mainly to lower earning asset yields and higher operating expenses.

Notably, USB incurred a $100 million future delivery expenses related to COVID-19 as it pertains to its merchant processing exposure. Noninterest income was up 10% year over year, largely because of higher mortgage banking revenues and fixed income, partly offset by a 7% decline in payment services, which were hurt by lower sales volume due to global lockdowns affecting clients. Loan growth mainly reflected drawdowns on funded commercial commitments.

Asset quality metrics remained stable. During the quarter, USB increased its provisions for loan losses, with a $600 million reserve build reflecting recent economic stress. Capital metrics were relatively stable versus the previous quarter, with its CET1 ratio down 10 bps to 9.0%.

Analyst: Stuart Plesser 

Wells Fargo & Co. (WFC)  reported weak first-quarter earnings performance, negatively affected by a number of nonrecurring items. Excluding these items, results remained subpar year over year. In addition, WFC reported a sizable reserve build of $3.1 billion during the quarter, and we expect further provisioning in 2020. The NIM compressed 33 bps, reflecting balance sheet repricing in a low rate environment.

Operating expenses declined, and the efficiency ratio improved during the quarter. Loans grew 6% from the prior year, mainly because of $80 billion drawdowns in commercial commitments. The asset cap remains a concern going into second quarter because, at the end of the first quarter, total assets were $1.98 trillion (two-quarter daily asset average was still below $1.95 trillion) due to loan growth outside of the Paycheck Protection Program, which the Fed excludes from the asset cap threshold.

Asset quality metrics remained well below historical levels but reflected an increase in both net charge-offs and nonperforming assets driven by the negative impact of market conditions on both commercial and consumer portfolios. WFC's CET1 ratio declined sequentially by 40 bps to 10.7% but remained 170 bps over the targeted minimum.

Analyst: Stuart Plesser 

Appendix D: Capital One And Wells Fargo Outlooks

Capital One

The negative outlook reflects our view that Capital One's financial performance could be more sensitive to the economic fallout from the COVID-19 pandemic than the average U.S. bank because of its sizable concentration in consumer lending, including to subprime consumers through its credit card and auto lending businesses. We believe the increase in unemployment and weakened consumer confidence (leading to lower consumer spending and lending activity) from the COVID-19-related economic shutdown has resulted in substantial headwinds for U.S. consumer lenders like Capital One. Our outlook also incorporates the fluidity of the situation and downside risks to the economic forecast over our two-year outlook horizon.

We expect the recession triggered by the COVID-19 pandemic could result in Capital One modifying a meaningful portion of loans and taking substantial credit provisions. We could lower the ratings if the economy has a more severe impact on Capital One's financial performance than we anticipate.

This could be caused by the length and severity of the recession being greater or the rebound being slower than we currently expect. For example, it is possible that government programs designed to support consumer finances will prove inadequate for the severity of the downturn for the bank's borrowers, leading to a rapid acceleration of loan deferrals and eventual loan losses. Borrower behavior in terms of willingness to repay or collateral values may also differ from the norms that Capital One experienced in prior downturns or simulated during stress testing, which could put additional pressure on credit losses.

At the same time, ultra-low interest rates and higher on-balance-sheet liquidity will weigh on net interest income (which accounts for more than 80% of total revenues), and suppressed business activity will hurt fee income, although operating expenses rationalization may provide an offset.

The adequacy of Capital One's loss buffer, including capital and loan reserves, to absorb the evolving ramifications of the COVID-19 pandemic will be a key factor in determining the ratings. Other factors include the maintenance of stable funding, robust on-balance-sheet liquidity, and conservative capital management, including limiting share repurchases until economic conditions improve. We could lower the ratings if Capital One's S&P Global Ratings RAC ratio declines and remains sustainably below the lower bounds of our 7%-10% adequate range, or if we believe that credit reserves will be insufficient to absorb the losses that we expect will materialize.

We would revise the outlook to stable if we see convincing evidence of abatement in problem assets and normalization of earnings in line with an economic rebound. To revise the outlook to stable, we would need to see Capital One demonstrate a track record of financial resilience through the economic downturn and a clear path to rebuilding its loan reserves, capital, and liquidity buffers to pre-pandemic levels. An upgrade is unlikely based on comparisons with higher-rated peers.

Wells Fargo

Our negative outlook reflects the possibility that we could lower our ratings on Wells Fargo & Co.'s holding company over the next two years. This could happen if:

• Wells does not show timely progress in improving its governance and operational risk management deficiencies and is unable to meet the requirements to lift the asset cap and show progress in removing the consent orders.
• Operational control issues become even more material--for example, if we expect substantial additional fines that are large relative to earnings, or if sizable additional operational controls, compliance, or governance weaknesses surface.
• The leadership transition results in a substantively different strategy, or if culture clashes result in significant operational disruptions.
• Capital ratios decline more than we expect as a result of more aggressive distributions without corresponding offset from consistent earnings generation.

Our outlook on Wells' main operating subsidiaries remains stable because we expect those ratings to be supported by the company's substantial buffers of loss-absorbing debt, even if the company's group stand-alone credit profile (group SACP) were to decline to 'a-' from 'a'. That said, we could lower those ratings if Wells' additional loss-absorbing capital (ALAC) ratio, by our measure, falls below our two-notch uplift threshold, a scenario we do not expect.

We could revise our outlook on the holding company to stable if Wells seamlessly navigates its leadership transition and resolves the deficiencies that the Federal Reserve identified in the bank's risk management, governance, and compliance practices, demonstrated by removal of the asset cap. We would expect the company's financial performance to remain stable and its strategy to remain consistent with its earnings generation power, its leading market positions in various businesses, and its longstanding, very strong deposit profile.