BY BRIAN GRAHAM
The recent rescue of NYCB through the infusion of private capital comes almost exactly a year after the collapse of Silicon Valley Bank. In the intervening 12 months, sentiment has seemingly gyrated between “the sky is falling” (i.e. a lot of banks will fail) and “nothing to see here” (i.e. banks are fine). As is usually the case, the truth lies in the middle: the banking industry faces severe stress but, at least so far, it looks like few banks are insolvent and, assuming regulators don’t stand in the way, private capital (and not FDIC bailouts) can be the answer.
The banking industry is facing two big challenges. First, higher interest rates have caused the value of fixed rate assets owned by banks like securities and loans to decline significantly. That pressure is evidenced by $699 billion dollars of unrealized losses and, consequently, poor earnings (as the spread between the yield on those assets, which were purchased when rates were zero, and the cost of deposits to finance them, which are decidedly not zero anymore). The bad news is obvious. The good news, such as it is, is that the extent of this challenge is in no way a mystery; any analyst or investor can easily see in public disclosures precisely how big the unrealized rate-driven losses are for each publicly traded bank.
Second, banks are big lenders on commercial real estate (or CRE) with $2.6 trillion in CRE loans on the industry’s balance sheet. CRE is under stress, both due to secular changes in social patterns accelerated by the COVID pandemic (such as work-from-home, which has materially impacted demand for office space) and to the impacts of higher interest rates and related inflation (such as declines in the net cash flows generated by a given property as the cost of financing, personnel, utilities, insurance and related factors have increased materially). It is much harder for an outsider to assess the strength or weakness of a given bank’s CRE portfolio. A wide array of factors that are not transparent to analysts or investors can cause an otherwise similar CRE portfolio to be perfectly fine or horrible - including geography, property type, underwriting, pricing and servicing. This makes it challenging to assess the extent of potential credit losses from CRE lending for any bank based solely on publicly available information. The best illustration of the opacity of CRE risk in the banking system is the fact that the credit problems simmering in the NYCB multifamily loan portfolio were unknown to the market (and indeed it appears to management as well) until very recently. Surprise.
The difficulty of analyzing bank CRE exposure is no excuse to ignore the simmering challenges that have been facing the banking system for a couple of years now. Let’s do the best we can with the information available to outsiders. As noted, we need to look at the combined implications of both higher interest rates and CRE stress.
For the rate side, it is pretty simple. For publicly traded banks, one need only do some simple math. Start with tangible book value (TBV), which is disclosed by all banks and already reflects unrealized losses on securities that banks own in an accounting category called “available for sale”. We only need to adjust for unrealized losses that are not already in TBV and conveniently public banks disclose these numbers too: losses on securities owned in an accounting category called “held-to-maturity” as well as the change in the fair value of loans held for investment. The only complexity here is for banks that are not public. Where banks aren’t required to disclose the fair value loss on loans, we simply apply an average from the public banks to guess at the embedded losses there.
We take the result, which is basically TBV adjusted for all unrealized losses, and divide that by the similarly adjusted assets of the bank. Regulators require banks to have minimum leverage ratio capital of at least 4% (and in practice expect much higher levels). So we divide banks into those with above (relatively stronger) and below (relatively less strong) that 4% level.
For CRE, we have little choice but to take a crude and imprecise approach by making the unfair assumption that all CRE is the same. We take the total amount of CRE loans for a given bank and divide that amount by the bank’s TBV. Regulators have defined a CRE concentration greater than 300% of capital to be of concern, so we use that to delineate banks with relatively higher or lower CRE exposures. I know this isn’t precise or accurate, but it is the best approach we have given the limited public information available.
The chart below, based on data from Q4 2023 call reports, sorts the banking industry by size and the rate and CRE metrics above. Note that we filtered out outliers like foreign owned banks and branches and special purpose charters (such as nondepository trust companies). Despite the imprecision, it is reasonable to look at banks with high levels of both unrealized rate-driven losses and CRE exposure of greater concern. This analysis tags 282 banks with higher levels of both, most of which are below $10 billion assets in size.
Caveat: it’s important to be cautious when driving to specific bank names. The CRE analysis in particular is imprecise.
Despite the gyration of public attention, the challenges facing the banking system - though real and serious - are not, at least so far, existential. The NYCB situation is a good example. It should come as no surprise that NYCB screened high on both rate-driven losses and CRE exposure; it is the “1” bank above in excess of $100 billion in assets. But, despite its challenges, our analysis suggested that NYCB was materially undercapitalized (when these exposures were taken into consideration), but critically not insolvent. As a result, private capital - and not a bailout from the FDIC - could seek to address the problem through a capital infusion. Which is exactly what happened a couple of weeks ago.
When looking at the other 281 banks, we should be concerned but there is good news as well. Only 14 of those banks, less than 0.3% of the industry by institution count and less than 0.1% by assets, screen as potentially insolvent on a mark-to-market basis. Even among the more challenged banks, the problems seem tough but solvable. A lot of incremental capital is required, through stand-alone capital raises for and, more likely for most , acquisitions by stronger, better capitalized banks. Given how small the institutions in question are, consolidation would in no way reduce competition. So far at least, the number of failures we should expect, though greater than zero, is not that daunting.
Image by Benjamin Sutter