Banks brace for higher rates

The Federal Open Market Committee held the federal funds rate at 3.50–3.75 per cent at its 29 July meeting. Bank treasurers on Wall Street took a different message from the vote. Three voting members broke ranks and called for an immediate rate rise, a sign that banks should get their balance sheets ready for tighter money.

Deposits cost more, fast

For bank finance chiefs, higher rates would squeeze margins in an awkward way. In theory, higher benchmark rates lift the yield on floating-rate commercial loans. But deposit betas are already at cyclical highs. Businesses and corporate treasurers are moving cash out of non-interest-bearing accounts into money market funds that pay a return, and banks are having to raise deposit rates sharply to hold on to their funding.

A deposit beta measures how much of a change in the policy rate a bank passes on to its depositors. Early in a tightening cycle betas are low, because customers are slow to move and much of the funding base pays nothing, so a rise in rates widens the gap between what a bank earns and what it pays. Late in a cycle that cushion has gone. With betas already high, each further quarter-point would reach the funding bill almost in full.

Why the loan book lags

Asset yields rise more slowly than the textbook suggests. Commercial and industrial loans priced off SOFR or the prime rate reset within weeks, but residential mortgages, many commercial property loans and most securities carry fixed coupons set when rates were lower. Those assets reprice only as they mature or are refinanced, which can take years. The mismatch is one of timing: the cost side of the income statement adjusts within a quarter, the revenue side over several.

Bond books take another hit

A further round of rate rises would also bring fresh mark-to-market losses on the bonds banks hold as held-to-maturity (HTM) and available-for-sale (AFS), eating into their common equity tier 1 capital ratios. The label on a bond matters less than it appears. Losses on securities held to maturity stay out of the income statement only while a bank can afford to keep them, and a run on deposits can force a sale at the worst possible moment.

What SVB taught

The last tightening cycle showed how this goes wrong. Silicon Valley Bank failed in March 2023 after uninsured depositors, many of them technology start-ups, pulled their money within days of the bank announcing a loss on securities bought when yields were low. Its bonds had not defaulted but had fallen in value as rates rose, and a funding base of concentrated business deposits proved far less sticky than its models assumed. The savings-and-loan crisis of the 1980s followed the same logic on a larger scale, as thrifts holding fixed-rate mortgages were crushed by the cost of short-term deposits once rates soared.

The pain would not be spread evenly. The largest banks, with big retail franchises, fee income and loan books weighted towards floating rates, are best placed. Regional lenders that rely on commercial deposits and hold heavy concentrations of fixed-rate commercial property loans are worst placed. Money market funds would collect more of the cash that leaves the banking system, and commercial property owners facing refinancing would find credit dearer and scarcer.

Backstops have been withdrawn

After Silicon Valley Bank failed, the Fed created the Bank Term Funding Program, which lent to banks for up to a year against Treasuries and agency securities valued at par rather than at market prices. That let lenders meet withdrawals without selling underwater bonds and crystallising the loss. The programme stopped making new loans in March 2024. A bank facing outflows in a fresh tightening cycle would have to turn to the discount window or to advances from the Federal Home Loan Banks, both of which lend against collateral at a haircut, or else sell the securities and book the loss.

The discount window also carries a stigma that makes treasurers reluctant to use it until the need is obvious to everyone, by which point depositors have usually noticed too. Some banks have spent the past three years paying fixed on interest-rate swaps to protect the value of their bond portfolios, and those hedges gain as rates rise. The protection is partial, it costs income while rates are steady, and a hedging programme demands collateral and expertise that smaller lenders are less likely to have.

The case for asset-sensitive banks

The counter-argument deserves a hearing. Since 2023 low-yielding bonds bought in the era of cheap money have been maturing, and the proceeds are reinvested at higher yields, so some banks are more asset-sensitive than they were in 2022. A rate rise that is quickly reversed would also do little lasting harm. The column's case still holds for most lenders, because deposit betas at cyclical highs leave little to gain on the funding side and plenty to lose.

The signposts sit in regular data. The Fed's weekly figures on commercial bank deposits, weekly totals for money market fund assets and the net interest margins reported with third-quarter results in October will show how fast funding costs are moving, and the September FOMC meeting will show whether the three dissenters have become a majority. If the Fed does raise rates again, banks will probably see funding costs climb faster than loan yields and bond losses grow, so margins are more likely to shrink than widen over the next few quarters.