A thought about bank profitability

I am certainly not an expert on bank balance sheets, so take today’s thoughts with a grain of salt and I invite experts to help me out if I made any mistakes in my thinking. But as I reflect on the banking turmoil in March this year, I think we are currently not paying attention to the unintended consequence of that event for bank profitability going forward.

To recap, here is what happened in March in a very simplified way:

Interest rates rose significantly in the US in 2022 and early 2023. This was great news for banks since it made deposits more profitable for them. After all, the business of a bank is to borrow short-term (deposits from clients that earn low short-term deposit rates) and lend long-term (mortgages and business loans that have long duration and earn long-term interest rates).

However, as interest rates rose, two things happened. First, the long-term assets of the banks (loans and investments in bonds) lost value. That’s not a problem as long as a bank has a risk management department that buys interest rate swaps that swap long-term fixed rates for short-term floating rates. Every bank does that except the amateurs at SVB Bank who forgot about that and hence put the bank into insolvency. Second, bank deposit holders look at money market funds and see that they can get higher interest rates there than in their bank deposits. And sooner or later, deposit holders vote with their feet and start withdrawing money from banks.

If the move from bank deposits to money market funds increases from a trickle to a flood, a bank run starts. And banks are notoriously at risk from a bank run. If the deposits march out the door too fast, banks cannot liquidate their loans quickly enough and eventually become insolvent due to a liquidity crisis. The risks of deposits walking out the door is particularly great for deposits that are so large they are not insured by the FDIC, which is why a bank run can become self-sustaining. Once large deposit holders start to panic, they withdraw deposits, creating a liquidity crunch in affected banks, which in turn creates an incentive for other large deposit holders to withdraw their deposits as well, etc.

The way to stop such a bank run is to either insure all deposits against losses, no matter how large they are or ensure that banks have enough liquidity at hand to pay all depositors who want to withdraw their money. That is exactly what regulators did in March of this year and why the bank run was stopped relatively quickly.

Ok, so far so good. But what are banks going to do now? They know they will not have the extraordinary support measures from the Treasury and Federal Reserve forever. They also know they have to keep deposits and that means they have to increase the rates they pay on their deposits to compete with money market funds. That is what usually happens when interest rates rise. Banks become less profitable because they have to pay higher interest rates on deposits and their net interest margin between deposit rates and the income on the loans they offer declines.

Another thing banks can do in reaction to a bank run like the one we have seen in March is to tighten their lending standards (lend less or at higher interest rates) and shorten the duration of their loan book. The former will reduce their revenues while the latter will keep revenues going but reduce profit margins. In my view, both of these things will happen, but reducing the duration of their loan book may create unintended consequences going forward.

Think about it this way. If you have shorter loans and invest in shorter maturity bonds you still make money while interest rates are high and you reduce your liquidity risk because if interest rates rise your assets will lose less in value and your asset-liability mismatch will not be as extreme, thus reducing losses and the risk for a bank run to occur.

That’s great, but what happens if the Federal Reserve starts to cut interest rates and does so just as fast as it increased rates in the first place?

In this case, the loans on the bank balance sheet will become worth more, but their value increases at a lower rate than if the bank had kept its duration longer in the first place. Meanwhile, when interest rates decline, deposits become less profitable for banks. Yes, the bank has to pay less interest on these deposits but the margins on these deposits also decline. In particular, if short-term interest rates drop close to zero again, deposits will become loss-making for banks, just like they were in the years before 2022.

Normally, the losses on deposits can be offset by the profits on the loan book and the profits on bonds held as assets on the balance sheet, but if the bank has shortened the duration of its loan book and bonds held on the balance sheet, these profits may not fully compensate for the losses on deposits. And the larger the deposit book of a bank is, the bigger the problem will become.

What I am saying is that if interest rates decline fast enough, banks that have hedged their risk against a liquidity crunch from a bank run may run into a profitability problem and become loss making very quickly. And ironically, these losses may create another bank run.

If you know your deposits at Bank X are not fully insured and Bank X has a massive deposit book that is becoming less profitable as short-term rates drop, what are you going to do? I would start to withdraw my deposits simply because this bank may very quickly become insolvent. Initially, my bank may welcome this move since getting rid of loss-making deposits will enhance profitability. But if enough deposit holders take their money out of the bank, you tend to get another bank run.

Thus, if the Federal Reserve cuts interest rates too soon and too fast, it may risk another banking crisis like the one we saw in spring this year, though this time for very different reasons.

Meanwhile, banks are caught between a rock and a hard place. If they want to reduce the risks from a bank run in times of high interest rates, they have to tighten lending standards and shorten the duration of their loan book, so assets and liabilities are better matched. But by doing that, they increase risks to their balance sheet and profitability when interest rates decline. And these risks increase the faster interest rates decline. Looks like banks can’t win either way.