Against bank stocks

Good morning. US regional banks indices rose yesterday, to everyone’s relief. Yes, First Republic shares got cut in half, even as other banks considered throwing the troubled lender another lifeline. But for most of us, it’s signs of systemic stabilisation we’re looking for, and it was a good day in that sense. More on banks below. Email us your thoughts: robert.armstrong@ft.com & ethan.wu@ft.com.

Why own banks?

The last couple weeks have served to remind everyone that it is not easy being a bank shareholder. Yes, Credit Suisse’s shareholders got something, which is better than the nothing that its AT1 bondholders received, but they have hardly had it easy. Silicon Valley Bank shareholders have been zeroed; First Republic shareholders are down 90 per cent; even shareholders in mighty JPMorgan Chase are down 11 per cent.

In the FT yesterday, the fund manager Terry Smith took these ugly results and ran with them, presenting an argument against ever owning bank shares. It’s a compelling case, and raises important questions. Smith’s main points:

1. Banks are highly leveraged. Many banks have $20 in assets for every $1 of equity, so if the assets fall in value by just 5 per cent, the equity is wiped out. This is insanely risky, and other businesses are not like this.

2. Even a solvent bank can be destroyed by a run by depositors, another risk more or less unique among banks.

3. Banks have systemic vulnerabilities: a bank can face a run, or an impairment of the value of its assets, because of other banks’ mistakes. In almost every business, having incompetent competitors is a blessing. In banking, it’s dangerous.

4. Banks, which basically just move electrons around in computer systems these days, are very vulnerable to disintermediation by technology.

5. For all this risk, banks’ return on equity is not particularly high.

The fifth point is the most interesting, especially if it applies not just to return on equity, but to shareholder returns, as well. Lots of businesses are risky, but back in finance school we were taught that shareholders would receive higher expected returns for taking on these risks. To a certain extent, bank shareholders are promised higher returns: bank stocks always trade at a meaty discount to the rest of the market on, for example, a p/e multiple basis (at present, the discount is very large: the S&P 500 banking sector trades at nine times earnings, compared to 19 for the wider index). 

Over the past 20 years or so, however, the cheapness of bank stocks has not translated to good shareholder returns. Here is the performance of the S&P 500 banking sector, and the somewhat broader KBW Nasdaq Bank index, against the S&P 500 from October 2002 (the bottom of the dotcom bust bear market) until year end 2021 (the top of the long post-crisis bull market): 

Smith appears to be on to something. And what is striking is that timing bank stock purchases perfectly with the cycles would not have helped much. Here is bank performance during the 2002-2007 bull run — a little outperformance early in the recovery, but underperformance overall:

Of course, at the end of that bull run, bank shareholders were absolutely crushed. Next, consider the 2009-2021 bull run that followed that shellacking — moments of outperformance, but ultimately, a lot of beta for market-like returns:

Note that last chart begins at the banks’ absolute low point!

What these charts suggest to me is that, as cheap as banks often are, they should be even cheaper. Why aren’t shareholders getting paid, even over the long run, for taking the many special risks bank ownership entails?

I think, or rather speculate, that this has something to do with the fact that banks are especially hard to understand. Most non-bank businesses are straightforward, at least from an accounting point of view: they are income statement businesses. The company provides goods or services in return for money; expenses, interest expense and taxes are deducted from that money; what remains is profit. Simple, in principle.

Banks are balance sheet businesses. Of course, they have income statements, too. But assessing the banks requires properly careful analysis of balance sheet assets and liabilities, which come in a bewildering variety of flavours. The accounting is picky. There are loads of acronyms. The filings are very long. It’s all hard. After thinking about banks on and off for more than a decade — sometimes as a full-time job! — I still feel I’m in over my head much of the time. Given this, it seems reasonable to suggest that bank risks are systematically and persistently under-appreciated by investors as a whole.

Alternatively, there may be something about the past 20 years that has made bank investing particularly difficult — we did have a banking crisis in the middle of that span, after all. Or perhaps there is another explanation? Or a better way to look at the performance data? We’d be keen to hear from readers on this — bankers in particular.

The CRE scare

Even before anyone much cared about the health of US regional banks, there were two worries about commercial real estate. One is rising rates, a danger for levered assets like CRE. The other is remote work. Unhedged wrote this newsletter in the Financial Times’ Soho office, but attendance was sparse, as it is most Mondays. Manhattan office occupancy is still a third below pre-coronavirus pandemic levels, and in other major cities, it’s even lower. This is a big problem for office owners, and for the people lending to them. Year over year, CRE prices contracted in December and January:

Since the run on Silicon Valley Bank began, listed real estate investment trusts, already battered by rising rates last year, have taken another leg down, led by offices:

This may owe to a third fear: a nasty tightening in credit conditions. As we mentioned yesterday, US regional banks do the bulk of CRE lending, something like 80 per cent. Their balance sheets are sensitive to what happens in CRE, too. Among smaller banks (ie, not in the top 25 by assets), CRE loans make up 29 per cent of total assets, versus just 7 per cent for the largest banks. Regional banks were already pinched by rising funding costs, and now face deposit outflows and lower share prices. That will push them to lend more cautiously and at higher rates.

What could the spillover to CRE look like? The tail-risk disaster scenario, says Kiran Raichura, property economist at Capital Economics, is that stricter lending (in combination with rising rates and working from home) jacks up defaults among property owners, forcing regional banks to seize the underlying assets. Banks don’t want fallow real estate assets clogging up their balance sheets, and will probably feel pressed to unload them at a discount. Forced selling would pile even more downward pressure on prices.

That, again, is a pessimistic outcome. Raichura’s base case is a 15 per cent peak-to-trough decline in CRE prices, which have already fallen 8 per cent from the peak last July. Ari Rastegar, who runs a real estate private equity shop in Austin, Texas, told us that outside of a brutal correction in offices, CRE has been broadly supported by the resilient labour market. As for rising rates, he says, “There’s always accretive financing for great deals, but maybe not for good deals” anymore.

But investors are antsy. Spreads on triple-B commercial mortgage-backed securities, the lowest rung of investment grade, are shooting up:

And so are lenders. The number of unique active lenders in CRE is falling, notes Jim Costello, chief real estate economist at MSCI. He writes:

The turmoil with the closure of New York-based Signature Bank may, in a microcosm, highlight the challenges that the market can face with fewer lenders active. Looking across loans originated for apartment assets in New York City in 2022, Signature was the second-largest lender.

Add CRE, alongside regional banks and residential real estate, to your watchlist of shoes that might drop. (Ethan Wu)

One good read

Hope springs eternal.