How Insane Was Silicon Valley Bank’s Zero Hedge Strategy?

The most famous example of the carelessness Silicon Valley Bank is that they foolishly bought a $124 billion bond portfolio.

But is that true? FT Alphaville dug into the balance sheets of SVB and Credit Suisse incredibly sad geeky comparison-and-contrast. The tl;dr may not be quite as idiotic as many people assume, but it’s pretty dumb. Be warned, the following is acronym-heavy.

The first thing to remember is that SVB’s bond portfolio basically fell into two different balancing groups. At the end of 2022, it held $91.3 billion in a “held-to-maturity” portfolio — bonds you want to hold on to until repayment — and $26.1 billion in an “available-for-sale” portfolio, which marks is market.

Here is a snapshot of SVB’s financial accounts at the end of 2022.

Let’s take the larger HTM portfolio first. Securities in the HTM basket may be held at their par value as they are expected to be held until full redemption.

As the table below shows, the majority of SVB’s $91.3 billion HTM portfolio consisted of very long-term, agency-guaranteed, mortgage-backed securities with maturities of at least 10 years ($56.6 billion, to be exact).

The credit rating of this stuff is extremely high, but it’s also very interest rate sensitive (for bond nerds, the average HTM portfolio duration was 6.2 years).

Because of rising prices actually The market value of the HTM portfolio was about $76 billion at the end of 2022, according to someone who saw the portfolio details and shared it with FTAV — an unrealized loss of $15.1 billion.

Yes, the SVB had no protection on this piece. But to do so would be silly. Keep in mind that the entire HTM portfolio is held at face value, but the value of the hedge would obviously fluctuate with the market.

So when interest rates rise, a bank makes money on the hedge, but the bonds remain at face value. If interest rates fall, they lose money on hedging, but they can switch bonds from HTM to AfS and sell them at the higher price. That means it’s basically a directional interest rate bet that goes straight to the income statement, something most banks loathe.

For example, Credit Suisse’s HTM portfolio of 1- to 5-year Treasuries stood at a fairly minimal $992 million at the end of 2022. The market value was around $949 million, but again, there doesn’t seem to be any insurance against the unrealized loss.

FTAV collects that some large commercial banks often Do Still, hedge a little against interest rate risk, just in case. But generally they’re just trying to hold mostly shorter-dated bonds to minimize interest rate sensitivity.

The SVB definitely didn’t do that – they actually added one as of c.2018 much of duration by accumulation in 30-year MBS. But in practice, not hedging HTM’s interest rate risk was probably not Silicon Valley Bank’s biggest mistake.

However, let’s turn to the AfS side. Unfortunately there are dragons here.

This is what the SVB AfS portfolio looked like at the end of 2022. As you can see, it was mostly Treasuries. Remember that these are valued at fair value, ie at market prices.

That’s pretty big. For comparison, Credit Suisse held $70.5 billion in “trading assets” at the end of 2022 — it constantly buys and sells securities of all types on behalf of clients — but its actual AfS portfolio (mostly corporate bonds) stood there $860 million.

The AfS bucket is definitely where most self-respecting banks lugging around a large portfolio of bonds will hedge their interest rate risk. Otherwise, the income statement would fluctuate from quarter to quarter in line with the market.

The SVB seems to have been aware of the danger. Here’s what CFO Daniel Beck said to analysts in early 2021:

. . . We are certainly positioning ourselves at this point for the potential for higher rates. So, during the quarter, we placed almost $10 billion worth of swaps on this portfolio that’s for sale. And we will continue to do more to protect ourselves and mitigate the impact of any further interest rate moves.

And as of late 2021, SVB’s financial accounts indicate that it held $15.26 billion in interest rate swaps on the AfS side to hedge against the impact of rising interest rates on its large bond portfolio. So what happened?

Well, it looks like profitability will falter in 2022 as the tech world pushed SVB to do something really stupid. In the first quarter, the company unwound $5 billion of AFS hedges to post a profit of $204 million, and in the second quarter it shed another $6 billion of hedges to post a profit of $313 million .USD to secure.

Or, as the bank put it in a July 2022 presentation to investors, “it shifted focus to managing downrate sensitivity.” (H/T Antoine Gara of the FT for the following slide):

You can see the shift here in the SVB Annual Report 2022. At the end of last year, it only had $563 million worth of hedges on its books. For comparison, the notional value of Credit Suisse interest rate swap hedges was $135.7 billion at the end of 2022.

To optimize its P&L near-term, SVB strolled almost entirely unhedged into 2023 — essentially a massive multibillion-dollar bet that interest rates were nearing their peak.

Ironically, it was somehow To the right! The 10-year Treasury yield peaked at about 4.29% in October last year, rising to just 4% in early March after a sharp drop in January (and has since fallen back below 3.5% on the chaos unleashed) . from SVB).

However, the Achilles’ heel of the SVB balance sheet was not the assets side, but the liabilities side. Specifically this:

Gormless, SVB had accumulated a massive pile of uninsured deposits, almost entirely in just one industry, which burned up its deposits as VC funding dried up.

Deposits are typically considered very stable, sticky funding, but in the case of SVB it was anything but. As money flowed freely last Friday, there was no way unsecured HMT securities could be sold without incurring an even bigger loss Than realizing the $1.8 billion raised in the March 8 sale of most of the AfS portfolio required the FDIC to intervene.

Bank balance sheets are a tricky business and FTAV hopes we haven’t garbled anything here. But if so, let us know in the comments.

Credit Suisse’s core problem clearly appears to be its struggling business and its exposure to higher interest rates is minimal. In contrast, SVB could be narrowly forgiven for no longer hedging its HTM book but locking in low interest rates and leaving its AfS portfolio almost bare to boost profits – despite a clearly unstable deposit base – looks like an asset -Liability snafu, this is going to be a cautionary tale for bank treasurers and regulators.