Money Stuff, by Matt Levine- Interest-rate hedging: SVB, and Schwab

Old_Joe

One question people have asked is: Why didn’t Silicon Valley Bank hedge its interest-rate risk? SVB, like other regional banks, got a lot of deposits and invested them in long-term US government and agency bonds with fixed interest rates. As interest rates went up, those bonds lost value, eating through all of SVB’s equity. This was bad, people noticed, they withdrew their deposits, and SVB ran out of money. This was all pretty predictable, or at least a known risk. Why didn’t SVB hedge?

We have talked about a couple of answers to that question:

• 1) SVB had expenses, and it needed to make money. It had to invest its depositors’ cash to make that money. In 2022, if it had been earning short-term interest rates on that cash, it would not have made enough money to cover its expenses. The way that it made money was by investing at long-term interest rates, which were higher.[1] So it invested in long-term bonds, earned higher rates, and made enough money. “Hedging” would have meant swapping its long-term rates to short-term rates, which would have defeated its main purpose, making money. And in fact SVB did have some interest-rate hedges in place in early 2022; it took them off, though, to increase its profits.

• 2) SVB thought that it was hedged: It was buying long-term bonds, yes, but it was funding those purchases with deposits. Those deposits are technically very short-term: Depositors could take their money back at any time, and eventually they did. But it is traditional in banking to think of them as long-term, to think that the “deposit franchise” and the deep relationship between banker and customer would make customers unlikely to take their money out. SVB invested a lot in good customer service and good relations with its depositors; it also made loans to startups that required them to keep their cash on deposit at SVB. So it figured it had pretty long-term funding, and it matched that long-term funding with long-term assets. If it had swapped the assets to short-term rates, and then rates fell, it would lose money, and SVB thought that was the bigger risk. When SVB got rid of its interest-rate hedges in early 2022, it did so because it had become “increasingly concerned with decreasing [net interest income] if rates were to decrease”: It worried that the hedges would hurt it if rates fell.

Those are, I think, the main answers. But there is one other sort of dumb accounting answer. Most of SVB’s interest-rate risk came in its portfolio of “held to maturity” bonds. The idea here is that SVB bought a lot of bonds and planned to hold them until they matured. If it did that, the bonds — which were mostly US-government backed and so very safe — would pay back 100 cents on the dollar. So SVB didn’t need to worry about mark-to-market fluctuations in their value. If interest rates went up, and the value of these bonds dropped from 100 to 85 cents on the dollar, SVB could ignore it, because the value would definitely go back up to 100, as long as it held the bonds to maturity. (The problem is that it couldn’t: There was a run on the bank long before the bonds matured.)

This is a standard assumption in banking, that the bank is making loans or buying bonds and planning to hold them for life, so fluctuations in their market values don’t matter. And bank accounting reflects this: Held-to-maturity bonds are held on the balance sheet at their cost, and fluctuations in their market values do not affect the bank’s balance sheet, or its income statement, or its regulatory capital. And thus for a while last year SVB was mark-to-market insolvent — if you subtracted its liabilities from the market value of its assets, you got a negative number — but its regulatory capital was fine, because regulatory capital doesn’t subtract that way.

But now add hedging. SVB had, call it, $120 billion of held-to-maturity bonds. When rates went up, they lost something like $15 billion of market value.[2] If SVB had fully hedged those bonds — if it had put on $120 billion notional amount of swaps, say — then the hedges would have perfectly offset that loss. But if rates had instead gone down, the hedges would have lost money. Obviously last year rates probably had more room to rise than to fall, but even a 0.25% decline in long-term interest rates could have cost SVB something like $2 billion in this scenario.[3]

Of course in that scenario its bonds would have gained $2 billion of market value, offsetting the loss on the hedges. But this is where the accounting is a problem. If you have a held-to-maturity bond, its fluctuations in value do not affect your income statement or balance sheet: When the market price of the bond goes up (or down), the book value of your assets does not go up (or down), and you do not have income (or loss) from the change. But if you have an interest-rate swap, its fluctuations in value do affect your income statement and balance sheet: When its market value goes up (or down), the book value of your assets goes up (or down), and you have income (or loss). An interest-rate derivative is sort of naturally a mark-to-market asset, and so changes in its value are reflected in income.

And so if SVB had hedged and rates had gone down, it would have reported a huge loss: A $2 billion loss on interest-rate derivatives would have wiped out more than all of SVB’s profit last year. Hedging the held-to-maturity bond portfolio would have made SVB economically less risky, but it would have made its reported financial results far more volatile. The hedge would have made SVB look riskier. And banking is a business of confidence, so you don’t want to look riskier. (Also: The hedge would have made SVB’s regulatory capital more volatile, and banking is also a business of regulatory capital.)

Now, an obvious response is: “This is dumb, why should hedging make you look riskier?” And accountants are aware of that, and there is a thing called “hedge accounting” where you basically get to take some asset and the derivative that you use to hedge it, offset them against each other, and neutralize the accounting effect of fluctuations in their values. The hedge makes your financial statements look less risky, which makes sense.

The problem is that this is specifically not allowed for held-to-maturity assets. PricewaterhouseCoopers explains:

The notion of hedging the interest rate risk in a security classified as held to maturity is inconsistent with the held-to-maturity classification under ASC 320,[4] which requires the reporting entity to hold the security until maturity regardless of changes in market interest rates. For this reason, ASC 815-20-25-43(c)(2) indicates that interest rate risk may not be the hedged risk in a fair value hedge of held-to-maturity debt securities.

Again, here the accounting standards line up with the way banks have historically thought about themselves, which is basically that they are in the business of holding long-term assets for the long term. “Why would a bank hedge interest-rate risk on its held-to-maturity portfolio,” the accountants ask, “if it is just going to hold that portfolio to maturity?”

That said, you can hedge your bonds that you treat as “available-for-sale,” and if you do that you will get hedge accounting treatment, so your income statement (and capital) will look less volatile rather than more. (This is what SVB was doing when it did have interest-rate hedges in place last year.) And if you are a US bank in spring of 2023, you will be keenly focused on the risk of rising interest rates, perhaps more keenly focused than you were back when interest rates were about to rise rapidly. Never too late I guess. Bloomberg’s Annie Massa reports:

Charles Schwab Corp. started using derivatives to hedge interest rate-related risk during the first quarter.

The derivatives had a notional value of $3.9 billion as of March 31, the Westlake, Texas-based company said in a regulatory filing Monday.

Schwab, which runs both brokerage and bank businesses, has been ensnared in the tumult ravaging US regional banks after the Federal Reserve embarked on its most aggressive interest rate tightening cycle in decades last year.

The firm confronted swelling paper losses on securities it owns and grappled with dwindling deposits as customers moved cash into accounts that earn more interest. Schwab executives have said those withdrawals will abate. The pace of cash withdrawals is already starting to slow, Chief Financial Officer Peter Crawford said in a recent statement.

It has $3.9 billion of swaps to hedge $3.9 billion of available-for-sale securities, out of a total of about $141 billion of available-for-sale and $170 billion of held-to-maturity securities.

Derf

@Old_Joe : Interesting article in WSJ today about shorting banks.

Old_Joe

@Derf- Yes, I caught that- thanks Derf