Is there a Bull Case for US Bank Stocks?

The IRA Bank Book Q4 2020

Review & Analysis

In this edition of The IRA Bank Book for Q4 2020, we ask a basic question: Is there a bull case for US banks? Whether from a perspective of an equity or debt investor, or a risk counterparty, the answer is unclear. Markets have taken up the valuations of bank stocks and debt because of the deliberate asset scarcity contrived by the Federal Reserve Board and other global central banks. But is there any hope of restoring US bank earnings and payouts to investors via share repurchases in the short-term?

Bank Group: ALLY, AXP, BAC, BK, C, COF, DB, DFS, FRC, GS, HSBA, JPM, MS, OZK, PNC, SCHW, TD, TFC, USB, WFC

Suffice to say that while nominal prices for bank equity and debt have largely recovered from the difficult market environment seen in Q2 2020, investors are depending upon greatly reduced earnings and business volumes to support their investment valuations. Indeed, with the cessation of share buybacks by many banks, investors are receiving far lower cash returns than even a year ago.

Share repurchases for the top 25 US banks were worth more than $120 billion annually in 2019. Dividends have been essentially tracking around 50% of bank pretax income, in some notable cases much more. To put that into round figures, the top 25 US banks were returning something approaching $350 billion annually to investors via dividends and share repurchases.

In Q3 2020, bank dividends were just $14 billion for the entire industry, $13 billion in Q2 2020 and $41 billion in Q1 2020. Despite these sharp reductions in cash flow from bank stocks, since April’s market carnage investors have rushed in to buy these assets with reckless abandon. There seemingly is no longer any reliable correlation between bank stocks and financial performance, although sudden changes in bank credit loss provisions, for example, would still hurt market pricing for bank stocks and bonds.

Source: FDIC

The sharp decrease in bank loss provisions reported in the third quarter – from $61 billion in Q2 2020 to just $14 billion in Q3 – came about as much due to the inability to quantify future losses and also the government-mandated loan forbearance granted to millions of consumer and thousands of businesses. But the sad fact is that many small and medium size enterprises (SMEs) that stayed afloat in 2020 due to government loans and subsequent forgiveness will likely fail in 2021.

We anticipate that just as the early period immediately following the Great Crash of 1929 was relatively stable, but was then followed by years of wrenching credit deflation, in 2021-2023 we are likely to see substantial restructuring of business assets and commercial real estate. These once blue-chip assets have been rendered moribund by the social distancing requirements of the response to COVID.

Just imagine how empty office and retail buildings in downtown Manhattan or Chicago or Los Angeles will be revalued in the next 24 months and you begin to appreciate the future impact on commercial real estate credit and related public sector obligors. We cannot even begin to list the clients and associates at different financial firms that are planning to decentralize their operations into hotel type structures that give employees maximum flexibility and protection. And in every case, the firms are moving their employees out of major metro areas such as New York.

Below we show the components of US bank income through Q3 2020. As you can see, interest expense is falling rapidly. Interest earnings are also falling. Since the Federal Open Market Committee greatly increased open market purchases of Treasury debt, and agency and government mortgage backed securities (MBS), the spreads on these assets over funding have steadily compressed, reducing the flow of income to banks. Indeed, banks and other investors are experiencing negative returns on loans, MBS and servicing assets.

Source: FDIC

The remarkable fact is that despite the compression of spreads of assets over funding costs, despite the other vagaries and credit expenses, the US banking industry did manage to report $128 billion in net interest income in Q3 2020. The problem, which lies 12-18 months out, is that the FOMC must eventually pull up on the monetary control stick and allow short term interest rates to trade freely or risk doing significant harm to the US banking sector.

Just as a large commercial airliner flying at low altitude must maintain a minimum speed or stall, the FOMC needs to understand that the compression of bank interest margins could drive US banks into a net loss position on their interest rate book, essentially a replay of the S&L crisis of the 1980s.

In the 1980s, the funding costs of S&Ls spiked over the yield on assets, causing mass insolvency in the industry that funded residential mortgages. This time around, the yield on earning assets could dip below the artificially low funding costs created by QE. As the chart below suggests, the return on earning assets (ROEA) is near a 40-year low.

Source: FDIC/WGA LLC

In addition to the effects of QE, US banks are fighting an environment where they are seeing portfolios running off due to loan prepayments even as deposits rise artificially due to QE. Total banking system now exceeds $21 trillion and earning assets are $19.3 trillion vs $10.5 trillion in Q1 2007. But please remember that the folks on the FOMC say that inflation is too low.

The basic problem for the US banking industry is that the FOMC honestly believes that diverting the income from $6 trillion in Treasury obligations and MBS to the US Treasury is somehow a form of economic stimulus. Certainly, the impact on the US economy of a boom in 1-4 family mortgages is enormous, but we’d argue that the positive income benefit of QE and mortgage refinance opportunities on US households is offset by draining trillions of dollars annually in income for bank depositors and bond investors.

For investors in MBS and whole loans, the current environment of high prepayment rates and falling bond yields promises only negative returns. And for large lenders, the lending market is daunting. Even as the FOMC has gunned deposit growth rates, bank loan portfolios are running off on net, as shown in the chart below.

Source: FDIC

One area where lending is surging is 1-4 family mortgages, where volumes in 2020 could come close to $4 trillion in new loans, near the 2004 record. As we’ve noted in recent comments on PennyMac Financial Services (NASDAQ:PFSI) and Rocket Companies (NYSE:RKT), lending volumes are likely to remain strong in 2021, but secondary market spreads are likely to compress further.

As SitusAMC noted in their most recent presentation on trends in the market for mortgage servicing rights (MSR), coupon spreads vs the Treasury yield index have been almost cut in half since April due to the open market operations of the FOMC. While lenders are still capitalizing the on-the-run 3% Ginnie Mae MBS coupons at 3x annual cash flow, we’d argue that a more realistic valuation is closer to 1.5x given current prepayment rates.

By no surprise, the banking industry is only benefitting modestly from the mortgage boom. Banks lost important market share in April due to the decision to shut-down purchases of third-party production in 1-4s. The chart below shows the US mortgage servicing sector, representing nearly $12 trillion in unpaid principal balance (UPB) of residential mortgages.

Source: FDIC, MBA, FRB

As the chart illustrates, the US banking sector still controls more than 75% of the servicing of residential mortgage loans, although the nonbank sector is growing quickly. The assets serviced for others (ASFO) remains just below $6 trillion or half of total outstanding UPB. Notice in particular that the decision by many banks to suspend third-party loan purchases via correspondent and wholesale channels is causing the $2.4 trillion bank retained portfolio of 1-4s to shrink. Nonbanks continue grow their share, largely in the Ginnie Mae market. In addition, sales of 1-4s and most other loan categories continue to fall, an ominous sign for future bank revenue and earnings.

Assets Securitized & Sold

Source: FDIC

One reason that the banks are retreating from 1-4 family lending, loan aggregating and servicing is that the cash flows are negative. As one veteran nonbank CEO told us last week, “Servicing is problematic for the banks. The cost of forbearance or loss of cash flow and increased future expense plus runoff creates the accounting loss. If you sell servicing and retain the infrastructure, your fixed overhead becomes burdensome. The runoff of the MSR is the real GAAP issue. Critical mass is a big challenge. Banks will struggle with the choice of selling MSRs or staying in because the cash flow may remain positive, even in the face of losses.”

Notice in the chart below that the FV of bank owned MSRs was stable Q2-Q3, but net servicing income plummeted in Q3 to -$2 billion – the worst performance since 2011 for industry. We suspect that the expense of COVID and the Cares Act is the culprit. Cash flush nonbanks may find some ready sellers of servicing among large commercial banks as year 2020 draws to a close.

Source: FDIC

While some of the stronger names in our bank coverage group such as JPMorgan Chase (NYSE:JPM) and U.S. Bancorp (NYSE:USB) have retraced much of the losses of Q2 2020, many names in the group remain depressed. Again, investors are not buying these names based upon current returns but in the hope of higher future returns. Notice that even the lowly Goldman Sachs (NYSE:GS) has managed to crawl out of the basement to book value thanks to the generosity of the FOMC. A summary of our bank coverage group is below:

Source: Bloomberg (12/9/2020)

Credit Analysis & Charts

As we note above, credit costs for US banks moderated in Q3 2020 as provision expenses fell. Actual charge-offs of the $10.9 trillion in loans held by US banks fell, however, as the real cost of credit remains muted. Part of the reason that provisions expense fell in Q3 was that the Fed and other prudential regulators have allowed banks to slow-walk the process of recognizing losses on loans under forbearance due to COVID, the Cares Act, as well as state-mandated loan moratoria. We do not expect this seemingly benevolent situation to continue.

Total Loans & Leases

The Chart below shows past due loans and loans charged-off through Q3 2020 for all loans held by US banks. Notice that the non-current rate is rising rapidly compared with the downward trend of the past several years. At the same time, the actual rate of charge-offs fell in Q3 2020, reflecting somewhat the regulatory forbearance that has been rolled out due to the COVID pandemic. For the same reason that provisions for future loss have fallen in Q3 2020, the reported levels of charge-offs are also down for many loan categories.

Source: FDIC

In addition to the impact of regulatory forbearance by the Fed and Federal Deposit Insurance Corp, particularly with respect to the rapid growth of bank assets and the related decline in capital levels, the big positive impact on US banks from QE has been the inflation of asset values. Note that LGD for all US banks fell 5% in Q3 2020, an extraordinary skew.

Although the level of loss given default (LGD) appears relatively normal, the price appreciation seen in real property, residential assets and commercial assets such as facilities of various types has greatly improved the credit exposures of US lenders. Office buildings in New York may not be particularly attractive, but other assets in superior locations are demanding premium prices. In the rare event of a loan default actually going to liquidation and sale of assets, the recovery rates are very good, sometimes more than 100% of the loan amount. The chart below shows loss given default for all US bank loans.

Source: FDIC/WGA LLC

Real Estate Loans

In the case of real estate loans, the events of the past nine months have begun to force up default rates on real estate generally, in large part due to commercial loan defaults. The default rate on commercial and industrial (C&I) loans rose almost 40% in Q3 2020, which suggests a lot of volatility in corporate and commercial real estate credit. Since default rates on 1-4s and multifamily loans, as well as construction and development loans, are tracking near zero at present, it seems safe to assume that CRE loans are driving this increase. Notice that charge-off rates for real estate loans are still miniscule.

Source: FDIC

Another perspective on the changes underway in the world of commercial real estate is shown in the chart below, where LGD has spiked upward in the past several quarters. Whereas LGD on all real estate loans was actually negative at the end of 2019, today post-default loss rates on the $5.1 trillion in real estate loans are strongly positive once again at 67%, almost precisely the long-term average loss rate going back 40 years.

Source: FDIC/WGA LLC

1-4 Family Loans

When we next move to the $2.4 trillion in residential mortgage loans, the situation looks very similar. Non-current loans are accumulating, but charge-off rates remain very low. More important, post default loss rates are still negative, reflecting the strong market for residential homes. So long as asset prices in the residential mortgage space remain buoyant, LGDs for 1-4s are likely to remain extremely low by historical standards. As shown in the two charts below, charge-off rates for 1-4s were actually negative in Q3 2020 and LGD was -8.4% vs the 50-year average loss rate post-default of 67%.

Source: FDIC

Source: FDIC/WGA LLC

In addition to the charts above, another significant indicator of delinquency in the 1-4 market is early buyouts (EBOs) of delinquent loans from Ginnie Mae MBS. As we noted in our recent update on PFSI (“Nonbank Update: PennyMac Financial Services”), buyouts of delinquent loans by Ginnie Mae issuers such as Wells Fargo & Co (NYSE:WFC) are growing rapidly, an indicator of the deterioration of underlying government-insured loans.

The chart below shows Ginnie Mae EBOs by banks through Q3 2020. When a loan is repurchased out of a pool, it is considered “rebooked” and shows up on the bank’s balance sheet as an NPL. With delinquency rates on FHA and VA loans in the teens in many regions, we expect to see this indicator rising over the next few quarters. If buyers of EBOs are successful in modifying the loan, for example, they may then sell the note into a new Ginnie Mae MBS and capture the gain on sale. But EBOs also contain significant downside risk for servicers.

Source: FDIC

Multifamily Loans

As with the single-family portfolio, the delinquency and loss rates for bank owned multifamily loans are also exhibiting considerable volatility. The $480 billion in total loans is showing virtually no net charge-offs, but the period of negative LGDs seems to be at an end, as shown in the charts below.

Source: FDIC

Source: FDIC/WGA LLC

Commercial & Industrial Loans

After real estate, the next largest loan category for US banks is the $2.5 trillion in C&I loans held in portfolio. The C&I book is less affected by the market manipulation of the FOMC, thus both the loss rates and LGD display a more normal, pre-2020 pattern. The divergence between the abnormal pattern seen in real estate exposures and the C&I book was noted in the most recent Quarterly Banking Profile from the FDIC:

“The net charge-off rate declined by 5 basis points from a year ago to 0.46 percent. Net charge-offs decreased by $418.2 million (3.2 percent) year over year. The annual decrease in total net charge-offs was attributable to a $1.3 billion (15.9 percent) decline in credit card net charge-offs. This decline offset increases in charge-offs for the commercial and industrial (C&I) loan portfolio, which increased by $898.5 million (39.3 percent). The C&I net charge-off rate rose by 8 basis points from a year ago to 0.49 percent, but remains well below the post-crisis high of 2.72 percent reported in fourth quarter 2009.”

Source: FDIC

Notice in the chart above that the delinquency rate on C&I loans is continuing to rise, but the net charge off rate for the series actually declined like the bank loan book overall. We suspect that this decline is an anomaly and is related to regulatory forbearance from the Fed, OCC and other regulators. That said, the economic reality within bank C&I portfolios is continuing to deteriorate, as illustrated by the rising LGDs. The fact that LGD dropped from 90% loss given default in Q2 2020 to only 85% loss in Q3 does not address the more general concern.

Source: FDIC/WGA LLC

Credit Card Loans

Like most of the other loan series discussed in this report, the $979 billion in credit cards are also displaying the impact of the Cares Act and various other types of official and unofficial loan moratoria tied to COVID. Non-current rates have fallen as a result of loan forbearance, but net charge off rates have remained steady ~ 4%. We suspect that when the legal mandates regarding COVID loan forbearance have lapsed, the net-charge off rates and LGD for credit cards will begin to rise again. More than many other loan types, credit card default rates and LGD are tied closely to levels of unemployment.

Source: FDIC

Source: FDIC/WGA LLC

Outlook for Q4 2020

As we head into year-end for 2020, the Street has earnings estimates for most of the bank group that reflect a decidedly downbeat outlook for Q4 2020 and, more important, for 2021. But that does not mean that Street analysts are unwilling to stretch -- even as a 1930s deflation threatens the US financial system.

The consensus earnings estimate for JPM, for example, has the bank delivering almost $7.50 per share in earnings in 2020 and $9 in 2021, a remarkable performance given the bank’s significant commercial exposures. USB, to take another industry exemplar as an example, is believed to be headed for $3 per share in 2020 and slightly higher in 2021. Both of these optimistic estimates depend crucially on the level of loan loss provisions for US banks in Q4 2020.

Given that both USB and JPM are trading near 1.5x book value and given that cash flows to investors have basically been cut by two-thirds, these valuations are quite remarkable. And since the FOMC seems intent upon keeping interest rates low for the foreseeable future and given that asset returns and sources of fee income are also under growing pressure, we have a hard time making the bull market case for US banks at this point in time.

It may be several years before that Fed and other prudential regulators allow banks to resume full dividend payments let alone share repurchases. And on the horizon, investors face the probability that major banks will be required to raise capital to support the “temporary” expansion of bank balance sheets due to the effect of QE. When the Fed buys Treasury securities or MBS from banks, they credit their master account at the central bank. But these funds are fleeting and can run off quickly as the MBS and Treasury paper is redeemed.

We suspect that the FOMC will maintain the current size of the Fed’s portfolio ~ $7 trillion, meaning that continued asset purchases will artificially inflate the size of bank balance sheets but will also retard lending and asset returns. In such an environment, look for earnings to come under strong downward pressure. Also, there is likely to be significant dilution for equity investors as the banks struggle to support their bloated balance sheets with new capital, this even as net interest income falls. Investors may have convinced themselves that bank earnings will rebound in the near-term, but instead we could be headed for a protracted period of low earnings and cash returns to bank investors.

The IRA Bank Book (ISBN 978-0-692-09756-4) is published by Whalen Global Advisors LLC and is provided for general informational purposes. By accepting this document, the recipient thereof acknowledges and agrees to our copyright and the matters set forth below in this disclaimer. Whalen Global Advisors LLC makes no representation or warranty (express or implied) regarding the adequacy, accuracy or completeness of any information in The IRA Bank Book. Information contained herein is obtained from public and private sources deemed reliable. Any analysis or statements contained in The IRA Bank Book are preliminary and are not intended to be complete, and such information is qualified in its entirety. Any opinions or estimates contained in The IRA Bank Book represent the judgment of Whalen Global Advisors LLC at this time, and is subject to change without notice. The IRA Bank Book is not an offer to sell, or a solicitation of an offer to buy, any securities or instruments named or described herein. The IRA Bank Book is not intended to provide, and must not be relied on for, accounting, legal, regulatory, tax, business, financial or related advice or investment recommendations. Whalen Global Advisors LLC is not acting as fiduciary or advisor with respect to the information contained herein. You must consult with your own advisors as to the legal, regulatory, tax, business, financial, investment and other aspects of the subjects addressed in The IRA Bank Book. Interested parties are advised to contact Whalen Global Advisors LLC for more information.