A Healthy Outlook for U.S. Banks

Credit Analyst Addison Maier shares his views on the outlook for U.S. banks now that the first quarter is behind us and discusses the likelihood of future dividends being paid.

Key Takeaways

  • First quarter earnings reports saw a rise in loan loss provisioning and a decline in capital levels; however, we think the majority of U.S. banks entered the crisis with strong balance sheets and remain well positioned to weather the recession.
  • We believe most U.S. banks are unlikely to cut their dividends, as the capital impact is not overly significant relative to the risk of signaling unwarranted instability. But cuts could occur if proactive regulators – “out of an abundance of caution” – request dividend reductions.
  • The magnitude and duration of the current recession are highly uncertain, but banks’ ability to meet the Federal Reserve’s extreme stress test scenarios inspires confidence.

First quarter earnings season for U.S. banks has now wrapped up, and while banks generally reported large increases in loan loss reserves and saw capital levels decrease versus year-end levels, we believe that when you peel back the layers, you find that banks are still well positioned to weather the COVID-19 storm. The massive loan loss increases were as much a function of a transition to a new accounting standard as they were the result of the coronavirus’s effects on the banks’ loan books. We do not expect the decline in their capital ratios to continue at the same level, and they may even reverse in future quarters: Draws against revolving credit by bank customers have slowed, and banks generally have stopped share buybacks.

We will likely see banks continue to build reserves against future loan losses in the second quarter due to the further deterioration of macroeconomic forecasts since the end of the first quarter. However, we believe that the banks’ strong profitability and capital levels going into the crisis should enable them to withstand a range of severe scenarios.

Getting a Head Start on Provisioning for Bad Loans

While provision charges were large last quarter, and management teams generally indicated that the second quarter could see another large provision build, it is important to note that many banks still generated positive net income. This means that their earnings capacity was more than enough to offset the billions that were set aside for future credit losses.

A wrinkle this quarter was the adoption of a new accounting standard called Current Expected Credit Loss (CECL). Unlike past provisioning practices where banks only had to set aside provisions once it was probable a loss was going to occur, under CECL banks are required to run their own macroeconomic models and, based on those models, provision for total expected losses over the life of the loan. Using the CECL approach means that if the banks forecast accurately, they will not have to provision for any more losses on their existing loan book. In this scenario, future provisioning could be minimal and, all else equal, suggests earnings would be higher in future quarters.

The banks’ forecasts are not, of course, going to be perfectly correct, but it is notable that – given the timing of quarter end – they could include some of the direct impacts of COVID-19 in their forecasting models. The banks did broadly indicate that the outlook has deteriorated since they closed the first quarter books and thus we should expect more sizable provisions going forward. But, given that their forecasts have already considered, at least in part, the current recession, second quarter provisioning could look similar, but the provisioning taken in the first quarter should not be extrapolated over the coming year.

After first quarter provisioning, allowances for loan losses are now a bit under 2% of loans (on average), though the exact figure depends on the particular bank’s loan portfolio. Loss reserves will likely be higher if they have more credit card debt and lower if they have more jumbo mortgages, for example. But the 2% figure equates to the typical bank having set aside roughly 40% of the credit losses that were modeled under the “severely adverse” scenario provided by the recent Federal Reserve (Fed) Stress Test (known as DFAST). And if we consider banks’ excess capital above regulatory minima, as well as their ability to continue to set aside provisions in future quarters while maintaining profitability, we are comfortable that the banks can withstand credit losses across a wider range of severe economic scenarios.

Capital Levels Should Also Be Resilient

Despite positive net income over the quarter, capital levels and ratios were generally lower. But we do not expect this level of capital degradation to occur again in the coming quarters for a couple of reasons. First, banks were still buying back their own stock through March 15. As the big banks have now suspended buybacks (likely until there is clarity on a recovery), buybacks are not likely to be a drag on capital levels in the near term. And the impact could be significant: The large banks operate with Common Equity Tier 1 (CET1) capital ratios of around 11%, so suspending stock buybacks saves them in the ballpark of 125 basis points per year in capital.

Second, last quarter saw many corporations draw on credit lines provided by the banks. These large revolver draws increased the denominator (risk-weighted assets) of the Common Equity Tier 1 (CET1) ratio, putting further pressure on the first quarter CET1 ratios.1 However, the banks generally indicated that the number of draws had flattened into the second quarter, keeping a lid on the risk-weighted assets denominator. Also, it is worth noting that loans made under the Paycheck Protection Program (PPP) are zero risk-weighted, so those loans will not have an impact on capital ratios when they are disbursed in the coming quarter.

Dividend Cuts Are Unlikely, Supporting the Case for Preferred Stock

Questions about the sustainability of banks’ common and preferred stock dividends are understandable, given that large reserves against loan losses (which depresses profitability), stock buybacks that have already occurred and loan growth all put significant pressure on capital ratios in the first quarter.

The uncertainty has been heightened by the suspension of common stock dividends by UK and European banks. One important difference between the EU and the U.S., however, is that the bulk of the return on European bank capital comes from dividends, while the majority of the return in the U.S. comes from share buybacks. In Europe, the shareholder capital return equation is about 90% dividends, whereas in the U.S., it is more like 33%. The lower percentage of capital return from dividends was intentional in the U.S., a decision resulting from the Global Financial Crisis (GFC). The thinking was that allowing banks to suspend buybacks quickly would help preserve capital (as it has in the first quarter of this year) without the risk of negative signaling effects that could accompany dividend restrictions.

Cutting common stock dividends is mathematically a credit positive, as more capital is retained to absorb losses rather than paid out (we estimate cutting dividends to zero would save ~50 basis points per year of capital on average). However, we would prefer banks do not cut their dividends due to the possibility of it unintentionally signaling concern about capital strength. One solution to this concern would be a proactive request by regulators for banks to suspend dividends “out of an abundance of caution.” This scenario is, we believe, more likely if political pressure builds against banks paying dividends and could result in a win from a creditor’s perspective: a dividend cut without negative signaling repercussions.

Even if common stock dividends do get cut (which is not our base case), we view any preferred stock dividend cuts as extremely unlikely, because the latter are much smaller than common stock dividends and they are a regulatory capital instrument. Because preferred stock dividends are typically about 20% of the size of common stock dividends, we think the negative signaling effects of cutting preferred dividends would far outweigh the capital retention benefits (which we estimate at about 10 basis points a year). We would also note that in Europe, when common dividends were cut, coupons on the European equivalent of preferred stock were not, and while those security prices wobbled a bit after the dividend cut announcement, they quickly retraced.

Looking Ahead

Despite our expectations for U.S. banks to remain healthy, the magnitude and duration of the current recession are highly uncertain. Indeed, the current crisis is so different from those of historical periods, including the Global Financial Crisis (GFC), that the traditional usefulness of looking to the past for worst-case scenarios is diminished. Simply put, U.S. banks have not seen such a rapid collapse in corporate revenue or personal income in modern history. Indeed, this is why the Fed presented its stress tests, and in our view, the banks’ ability to meet those standards inspires confidence.

Looking further ahead, if the recession requires even lower (and possibly negative) interest rates, this could weigh on bank profits. While our base case is that rates remain low but rise as the economy stabilizes, banks will have to adapt to a sustained low yield environment. However, as a result of heightened oversight after the GFC, and conservative management, the strong balance sheets and resilient earnings of U.S. banks gives us confidence they have not only the ability to weather the current storm but also to provide much-needed stability in a time of great uncertainty.

 

1Tier 1 capital refers to core capital that includes equity capital and disclosed reserves. CET1 is a component of Tier 1 capital that consists mostly of common stock held by a bank or other financial institution. It is a capital measure that was introduced in 2014 as a precautionary means to protect the economy from a financial crisis.

Bank financial statistics sourced from Bloomberg and Janus Henderson analysis.