Insights | June 18, 2020

No Need to Bank on a Rebound

The Current Opportunity in Financials

Executive Summary

The pandemic has created an extraordinary risk/return trade-off for the shares of high quality U.S. banks. We believe there is the potential for decent returns for bank investors without improvement in the current environment, and the potential for enormous returns if the rate of change in the economy remains positive.

As we watch the markets heading back toward February highs, we note that returns have been driven by technology companies with sturdy balance sheets, defensive consumer staples, and health care. Travel, energy, and credit-exposed financials, among others, have had a bad year.

U.S. bank shares are, we believe, subdued by three causes of fear resulting from the global pandemic. The first and most significant cause for concern is the potential for loan losses in a dramatic, GFC-beating recession. The second is the fear that the Fed will mandate an added regulatory capital cushion in the current environment, resulting in dividend cuts and putting a lid on growth. Finally, there is the fear of a what a move into negative rates might spell for banks’ ongoing profitability. In this note, we outline the case for investing in high quality U.S. banks.1

Banking systems are built to absorb negative news. The ratio of bad to good news for banks is inherently skewed to the downside. For the vast majority of a bank’s assets, the return is capped at an agreed rate – and the actual outcome will be worse after taking into account the inevitable write-offs – while a bank’s liabilities are resolutely dependable and must be paid. However, elevated levels of bad news are only truly problematic for long-term shareholder returns if they necessitate dilutive capital raising, like that seen most recently during the Global Financial Crisis (GFC). It is from this perspective that we address the current fears for the banks; we conclude that the risk of large-scale dilution, for high quality banks such as U.S. Bancorp and Wells Fargo at least, is remote and that the prices on offer at the time of writing are attractive.

Fear of Loan Losses

Yes, there will be significant credit losses for the banks. Q2 will be grim. Whereas Wells Fargo and U.S. Bancorp have set loan loss reserves for low double-digit unemployment, initial pandemic data tells us this is probably not going to suffice. Financial markets have short memories, but not that short – we can understand why many investors’ first urge is to mutter “here we go again” and move on to other opportunities with less hair.

Our work on banks over the years tells us that the devil is in the detail. It’s not the credit losses that investors should fear per se – credit losses are an inevitable part of banking – but a mismatch between those losses and the ability to absorb them. We believe that this time around, for the higher quality banks at least, this is no rerun of the GFC. Why?

In short, we are seeing the upside of tougher post-GFC regulation. In the years and even decades that preceded the GFC, it was not uncommon for banks to run a model of very high leverage, 20-25x assets to equity, etc. Today bank leverage is much lower; they are flush with regulatory capital and in most cases are far, far above their required minimum levels. It’s one thing to run into a biblical flood of credit losses at GFC-era 25x leverage – take a hit of 2.5% of your assets and you lose over 60% of your equity – but another at today’s 10x equivalent to 25% of your equity over a couple of years by offsetting against profits from the rest of the business. Wells Fargo, for example, just added $3 billion to loan loss reserves in Q1, but can withstand another $21 billion in credit losses (cumulatively at least half a biblical flood) before troubling its regulatory minimum and without the assumption of profits from elsewhere in its book.

In addition to the stronger overall balance sheet poise, we believe that the quality of the collateral that the banks hold in the event of client defaults today compares very favorably with the last cycle.

Banks hold collateral against the various categories of loans, but it is on the key consumer mortgage side where the difference this time around is most plain to see.

  • Today the average consumer is significantly less levered than she was in 2007. Her personal savings rate was just 2% then; today she saves 8% of her income.2
  • There was a clear housing bubble then, with the U.S. homeownership rate pushing toward 70% as the marginal borrower – those sadly most at risk economically in 2020 – stepped in. Today subprime is more or less absent from the books of both U.S. Bancorp and Wells Fargo.
  • Housing prices in the U.S. went to ~5x price to income prior to the GFC vs. 3.5x historically. Today we sit much closer to historical norms at 4x.
  • Loan-to-Values (LTV) are markedly lower than during the GFC. There is no home equity lending frenzy in 2020. Even for Wells – despite its recent brush with scandal, always one of the more conservative banks – greater than 40% of loans in 2008 had an LTV of greater than 90%. Today the equivalent figure is just 11%.3

And, just think of this: with levered consumers, overpriced homes, and high LTVs, it took a full 25-30% national correction in home prices to get to a 2.5% mortgage loan loss ratio for banks. With home equity today representing the majority of household wealth, we tend to the view that homeowners make serious efforts to protect their equity. While it’s easy to walk away and default when you’re underwater on your loan, it’s a different ballgame when 20-25% of the value of the house is yours. We don’t anticipate a major correction in housing prices today, but even if we’re wrong, current LTVs will leave bank losses rather low.

A similar logic applies to the commercial half of the banks’ real estate books. While credit losses are looming, we believe that reduced bank leverage, reduced consumer leverage, and better collateral make the risk of dilution for the most conservative banks sufficiently remote.

Fear of Regulatory Edicts on Capital

The flip side of the tougher post-crisis regulation manifests itself in a fear of further capital constraints being placed on the banks. In the face of deteriorating economic conditions, for example, the payment of dividends could be curtailed and we believe that weighs on stock prices today. The banks have already agreed to suspend buybacks, but the results of the Federal Reserve’s stress tests this month provide another opportunity for the Fed to intervene on capital plans. There is a school of thought in the Fed in favor of building a temporary extra cushion of regulatory capital given the Covid-19 crisis. While we can see why that might make a good sound bite, we believe that this is a relatively unlikely outcome as it runs counter to almost every effort by the government machine to accommodate the stresses in the economy today. If banks require even more capital on their balance sheets, absent an enforced capital raise, they would need to limit lending – an outcome that would be, we suspect, intolerable in an election year. If our dividend suspicions are misplaced, the banking stocks might well take a short-term knock but cutting dividends would have almost no bearing on long-term returns; if a bank earns and accrues a dollar to book value, whether that dollar is paid out today, or retained or paid out in a couple of years, barely changes the discounted value of the franchise.

Fear of Low Rates

The final concern is the impact of low rates. Should rates push into negative territory, the possibility that deposit costs get snagged on the zero-bound rate while loan rates are less constrained implies that banking margins could come under pressure. In Japan and Europe, deposits rates – for individuals at least – have not generally gone negative, while government bond yields have done so emphatically over the last year.

While it is mechanically true that tightening spreads are, all things equal, negative news, the connection between net interest margins and bank return on capital is only a loose one. Banks can adjust the rates charged, fees charged, expenses incurred, funding sources, etc., to address return on capital – the market for banking in the U.S. is sufficiently competitive that we would expect the market to clear at an economic rate over time.

But perhaps more importantly, the low rates are part and parcel of a remarkable package of policy responses to this pandemic-induced downturn around the world. The amount of stimulus and backstopping being thrown at the system is unprecedented and, unlike during the GFC, almost immediate. Whether it’s checks to individuals, enhanced transfer payments, paycheck protection, or forbearance programs, it’s clear that the system is bending over backwards to keep households liquid. Again, we interpret this through the lens of dilution risk, and we believe that the overall set of policy actions is a positive. If the price of buttressed borrower health is a couple of years of diminished spread while the system clears, we remain confident shareholders at the prevailing stock prices for U.S. Bancorp and Wells Fargo.


In the face of huge economic disruption from the pandemic, investors are concerned about the possibility of GFC-like trouble for the banking sector. The concerns have some merit – banking is about managing disappointment, after all. A focus on credit losses, the potential for capital constraints, and narrowing spreads, however, must be balanced by an assessment of the probability of dilutive capital raisings – by far the most impactful potential determinant of long-term shareholder returns. We believe that those probabilities are low, much lower than during the GFC. At price-to-book valuations close to GFC lows, the risk/reward profile of the stronger U.S. banks represents an unusual opportunity for the quality-seeking investor.

Download article here.


Register for full access to GMO's Research

Please enter a valid email address

Please enter First Name

Please enter Last Name

Please enter Company

Investor Type Required

Country Required

State is required

* Indicates required field