Financial news

To assess the possibilities, it is important to understand how changes in interest rates affect financial institutions. A bank’s balance-sheet is the mirror image of its customers’. It owes depositors money. Loans people owe it are its assets. At the beginning of 2022, when rates were near zero, American banks held $24trn in assets. About $3.4trn of this was cash on hand to repay depositors. Some $6trn was in securities, mostly Treasuries or mortgage-backed bonds. A further $11.2trn was in loans. America’s banks funded these assets with a vast deposit base, worth $19trn, of which roughly half was insured by the fdic and half was not. To protect against losses on their assets, banks held $2trn of “tier-one equity”, of the highest quality.

Then interest rates leapt to 4.5%. svb’s fall has drawn attention to the fact that the value of banks’ portfolios has fallen as a result of the rise in rates, and that this hit has not been marked on balance-sheets. The fdic reports that, in total, America’s financial institutions have $620bn in unrealised mark-to-market losses. It is possible, as many have done, to compare these losses with the equity banks hold and to feel a sense of panic. In aggregate a 10% hit to bond portfolios would, if realised, wipe out more than a quarter of banks’ equity. The financial system might have been well-capitalised a year ago, so the argument goes, but a chunk of this capitalisation has been taken out by higher rates.

The exercise becomes more alarming still when other assets are adjusted for higher rates, as Erica Jiang of the University of Southern California and co-authors have done. There is, for instance, no real economic difference between a ten-year bond with a 2% coupon and a ten-year loan with a fixed 2% interest rate. If the value of the bond has fallen by 15% so has the value of the loan. Some assets will be floating-rate loans, where the rate rises with market rates. Helpfully, the data the researchers compiled divides loans into those with fixed and floating rates. This allows the authors to analyse only fixed-rate loans. The result? Bank assets would be worth $2trn less than reported—enough to wipe out all equity in the American banking system. Although some of this risk could be hedged, doing so is expensive and banks are unlikely to have done much of it.

But as Ms Jiang and co-authors point out, there is a problem with stopping the analysis here: the value of the counterbalancing deposit base has not also been re-evaluated. And it is much, much more valuable than it was a year ago. Financial institutions typically pay nothing at all on deposits. These are also pretty sticky, as depositors park money in checking accounts for years on end. Meanwhile, thanks to rising rates, the price of a ten-year zero-coupon bond has fallen by almost 20% since early 2022. This implies the value of being able to borrow at 0% for ten years, which is what a sticky, low-cost deposit base in effect provides, is worth 20% more now than it was last year—more than enough to offset losses on bank assets.

The true risk to a bank therefore depends on both deposits and depositor behaviour. When rates go up customers may move their cash into money-market or high-yield savings accounts. This increases the cost of bank funding, although typically not by all that much. Sometimes—if a bank runs into severe difficulties—deposits can vanish overnight, as svb discovered in ruinous fashion. Banks with big, sticky, low-cost deposits do not need to worry much about the mark-to-market value of their assets. In contrast, banks with flighty deposits very much do. As Huw van Steenis of Oliver Wyman, a consultancy, notes: “Paper losses only become real losses when crystallised.”

This is why Larry Fink, boss of BlackRock, a big asset-management firm, has warned of a “slow-rolling crisis”. He expects this to involve “more seizures and shutdowns”. That high interest rates have exposed the kind of asset-liability mismatch that felled svb is, he reckons, a “price we’re paying for decades of easy money”. Mr Conti-Brown of UPenn points out that there are historical parallels, the most obvious being the bank casualties that mounted in the 1980s as Paul Volcker, the Fed’s chairman at the time, raised rates.

Higher rates have exposed problems in bond portfolios first, as markets show in real-time how these assets fall in value when rates rise. But bonds are not the only assets that carry risk when policy changes. “The difference between interest-rate risk and credit risk can be quite subtle,” notes Mr Conti-Brown, as rising rates will eventually put pressure on borrowers, too. In the 1980s the first banks to fail were those where asset values fell with rising rates—but the crisis also exposed bad assets within America’s “thrifts”, specialist consumer banks, in the end. Thus pessimists worry banks now failing because of higher rates are just the first domino to collapse.

The result of all this is that the banking system is far more fragile than it was perceived to be—by regulators, investors and probably bankers themselves—before the past week. It is clear that smaller banks with uninsured deposits will need to raise more capital soon. Torsten Slok of Apollo, a private-equity firm, points out that a third of assets in America’s banking system are held by banks smaller than svb. All of these will now tighten up lending to try to strengthen their balance-sheets.

That medium-sized banks can be too big to fail is one lesson regulators should learn from svb. The episode has upended other parables of post-crisis finance as well. “After 2008 investors thought deposits were safe, and market funding was risky. They also thought Treasuries were safe and loans were risky,” says Angel Ubide of Citadel, a hedge fund. “All of the post-crisis rule books were written on that basis. Now the reverse looks to be the case.” One parable remains intact, however. Problems in the financial system never emerge from the most closely watched places. ■