Is Schwab cheap?

Good morning. Apple is now in the business of monetary policy transmission. Its new Goldman Sachs-run savings account offers a 4.15 per cent interest rate. Bank of America and Chase are still only paying a few basis points. Could a concerted Apple advertising push massively increase deposit competition? Email us: robert.armstrong@ft.com and ethan.wu@ft.com.

Schwab is not SVB. But is it cheap?

Here’s a stock chart for Charles Schwab, the low-cost broker, since the start of the pandemic. It’s been a roller-coaster ride, particularly lately:

For a cogent explanation of all that volatility, read Madison Darbyshire’s excellent piece from over the weekend. A summary:

• In 2019, Schwab eliminates trading fees, a move that shakes the brokerage industry.

• Part of the strategy for making up lost trading fee revenue is attracting new clients who would park cash on the brokerage platform — which is to say, place deposits at Schwab’s bank subsidiary. Schwab could then invest the cash at a higher rate than it paid clients in interest.

• It did so by investing it largely in medium-term government-backed bonds.

• Rates rose fast, causing those bonds to lose value, and clients to move their money out of low-yielding bank accounts and into other products such as money-market funds, either at Schwab or elsewhere.

• Schwab shares are down 40 per cent from their peak in January.

Part of the reason the stock has been hit quite so hard is that the story rhymes with that of Silicon Valley Bank. A flood of cheap deposits and investment in “riskless” securities is a theme for both. At both companies, securities are 55-60 per cent of total assets. But there was no run on Schwab’s bank, in part because the great majority of its depositors are Federal Deposit Insurance Corporation insured. Another crucial difference: the duration of Schwab’s securities portfolio is shorter than SVB’s was, mitigating the losses from higher rates. Here is chief executive Walter Bettinger, sounding a little shirty on a call with investors yesterday:

It’s important not to confuse, as unfortunately some less than savvy alleged researchers and analysts have, that maturity or weighted average life is not the same as duration. We have many floating rate securities that have a long life, but essentially zero duration and therefore, do not contribute to negative marks with higher interest rates . . . We have historically managed our bank investment portfolio to a duration range between 2.75 and 4 years. And we were approximately 3.5 years as rates began to rise in mid-2022 . . . our overall duration across the firm’s aggregate balance sheets, which includes the banks and the broker-dealers, was about 2.5 years.

Here is a chart of securities and deposits at Schwab (the acquisition of TD Ameritrade, which increased the size of the company’s balance sheet, is marked):

The fact that the securities have shrunk its portfolio without Schwab having to take mark-to-market losses bears out Bettinger’s comments.

All that said, deposits are leaving Schwab, so it will have fewer earning assets on its balance sheet, even as its cost of funding rises. This has brought revenue and earnings down and will continue to do so. Here is a chart of Schwab’s two big revenue streams, net interest income and fee and trading revenues:

It does not seem crazy to expect interest revenue (the blue line) to return to something like its pre-pandemic level. Schwab took a wild trip during the pandemic boom, and the ride is ending. The question is how much of this round-trip is priced into the stock. Schwab’s forward price/earnings ratio passed 25 in 2021; it is now down to a pre-pandemic 13.

I’m not yet sure if that makes Schwab cheap or not. But I am sure it is an intriguing post-pandemic story. The large-scale, low-cost broker model has worked well for investors historically. Even after the recent drawdown, the stock has compounded at 13.3 per cent a year, beating the S&P’s 12.1 per cent. I will write more about Schwab in days to come.

A tail-risk tiff

A week or two ago, Bloomberg’s Justina Lee published a widely read piece scrutinising Universa, the tail-risk hedging fund founded by Mark Spitznagel and advised by Nassim Taleb. We interviewed Spitznagel back in October and found his story sensible enough, so we thought it worth having a look at what his critics are saying.

Remember the basic pitch for tail-risk hedging. Nothing hurts long-term compound returns on stocks like sudden, sharp losses. So you buy protection, accepting a steady performance drag in calm markets to trim your downside in the next crash. Most of the time you bleed money slowly. But every once in a while, things melt down and your tail hedge will (hopefully!) perform so well that it’ll offset years of small losses.

A tricky question, though, is how a tail-risk fund should present its historical results to investors. It’s not as simple as showing annual returns. Tail hedges are meant to protect other equity investments, so you want to see the performance of the portfolio containing both the equities and the protection. But which equities? With how much protection? Over what timeframe?

So non-standard returns presentation is needed. But rivals have long complained that Universa goes too far, using methods that, they say, are clearly misleading. Most hotly disputed are the staggering returns Universa has touted during market panics, such as its 3,612 per cent return in March 2020. Lee’s piece gets at this (you should read the whole thing):

Using standard fund maths, a 3,612 per cent return on $6bn — roughly Universa’s stated assets under management at around that time — would’ve translated into a mind-boggling $217bn windfall for Universa clients, and that’s how many people read it. In fact . . . What Universa in effect does is calculate the return on an insurance policy using only one month of premium. Conveniently ignored there is the reality that clients typically pay Universa for protection for years.

In other words, Universa picks a very small denominator to create those huge returns figures. The fund’s loudest critic, Boaz Weinstein, who does a different flavour of tail-risk hedging at Saba Capital Management, has gone as far as suggesting these are “fake returns”. He also objects to Universa’s claim that, “an investor with 98 per cent in the S&P 500 and 2 per cent in Universa would have seen their money grow an average 11.8 per cent annually in the firm’s 15-year history, compared with 9.6 per cent for the index alone”. Weinstein tweeted earlier this month:

That’s true for every tail fund — there’s nothing about Universa’s strategy that makes this possible.

Universa chief operating officer Brandon Yarckin had this to say: “We love that people want to discuss our returns publicly, even if we can’t [for regulatory reasons]. The more people think about returns on invested capital and portfolio effects, the better off they’ll be in the long run.”

The four-digit returns figures do smell of marketing puffery, even if they’re technically accurate. But here are some clearer returns from an investor letter circulated in April 2020, shortly after the pandemic market crash. Note that the aftermath of a crash is precisely when a tail-risk fund’s returns should look strongest:

Lee also reports this figure:

The firm estimates its insurance costs on average about 1.7 per cent of a protected portfolio annually, according to a person who has heard the firm’s pitch and who declined to be identified as the discussion was private.

Put together, it seems to us that Universa’s product is what it says on the tin (alongside a bit of overwrought marketing). You are paying 1.7 per cent a year to be up 0.4 per cent during the month that King Kong climbs the Empire State Building (or whatever) and the S&P is tanking.

How impressive is it that, from inception through April 2020, a Universa-insured portfolio has beaten a 75/25 stocks/Treasuries portfolio by 2.6 percentage points? You might think of that as compensation for taking on execution risk — the risk that Spitznagel (or Weinstein or someone else) can’t pull off their complex derivatives trades in the next crash. Investors know what they are getting with a bond hedge. But when paying a very clever person to build a derivatives hedge, investors have to reckon with the possibility that when things go south they will end up holding little but a very complex explanation for why the hedge didn’t work this time, for reasons that were impossible to anticipate.

As one investor who has worked with both Universa and Saba Capital Management told us: “Even when you’re good, you can’t make [upwards of] 150 per cent in one month, even when you’re right, and not take very significant counterparty risk with your capital . . . There are all kinds of structural bias that preclude [short positions]”.

That 75/25 portfolio has the virtue of dumb old simplicity. Unhedged finds tail-risk hedging fascinating, but for our money, boneheaded is best. (Ethan Wu)

One good read

The Pareto principle, shoplifting edition.