Anatomy of a Bank Collapse

When depositors worry about a bank’s ability to make good on its deposits, it can quickly become rational to panic, move deposits out, and ask questions later. We have the FDIC to guarantee the value of deposits smaller than $250,000, so there is no need to run if you’re under that limit.

But if you’re a business with a very large bank balance that you’re using to fund payrolls or other operating expenses, it’s important to have some idea of your bank’s safety and soundness.

Last week, a bank run that was magnified by chatter on Twitter caused a liquidity crisis at Silicon Valley Bank (SVB).

SVB did not have enough liquid assets available to meet deposit withdrawal demands. So, the FDIC stepped in and took it over last Friday. Jim will have more to say on the topics of bank bailouts and contagion in the days ahead.

For now, allow me a few moments to unpack SVB’s failure and walk you through some of its implications for the banking sector at large.

SVB’s failure brought to light the risks that longer-duration, held-to-maturity (“HTM”) securities pose to banks. “Long duration” refers to bonds with ten, twenty, or thirty years to maturity. I’m referring to Treasury, corporate, and mortgage-backed bonds that move up and down in price when interest rates move.

SVB shows what can go wrong when any investor (institutional or retail) buys a truckload of low-yielding, long-duration bonds right ahead of a big spike in interest rates.

These types of bonds are fairly safe to buy now. They were not safe to buy in 2020 and 2021. Back then, they offered rock-bottom yields. That’s when they had the most interest rate risk. Why?

Because the Fed massively overdid it with its mortgage and Treasury bond purchases. The Fed thought it was helping by suppressing long-term interest rates. It was not helping. It was storing up trouble for later.

Remember, bond yields and bond prices move in opposite directions. 2020 and 2021 were historically terrible years for a buy-and-hold investor to be buying long-duration bonds. Concerns about rapid Fed rate hikes and inflation caused a jump in bond yields and a crash in bond prices. In 2020 and 2021, SVB’s internal treasury department loaded up on long-duration securities, including various types of mortgage-backed securities.

These securities suffered price declines in the secondary market. SVB did not buy derivatives to neutralize the risk that its securities portfolio might fall in price and be smaller than expected to satisfy a big bank run. Derivatives are widely available at large banks to hedge its interest rate risk. Many bank securities managers use them as a risk management tool — but not at SVB. Who knows why not?

Importantly, losses on long-duration securities are not considered a big problem unless a bank is forced to fund a liquidity need. During last week’s run on the bank at SVB, its managers liquidated their longer-duration securities portfolio to raise enough cash to meet deposit withdrawals. Unfortunately for SVB shareholders, this crystallized what had been a temporary loss on these securities.

Why did depositors run on SVB? Some of them ran due to concern over the quality of SVB’s assets. SVB was heavily invested in loans and assets tied to the Silicon Valley ecosystem.

Some depositors ran because SVB’s treasury department was not managing interest rate risk.

And some depositors ran simply because they saw other depositors run — whether they knew about interest rate risk or not.

Some blame for this fiasco goes to the federal government and the Federal Reserve for running multi-trillion-dollar deficits and QE post-Covid. In other words, Congress and the Fed set the stage for SVB’s failure.

Trillions of dollars in deficit spending and QE stuffed the banking system with too many excess bank deposits. Some bankers were bound to do foolish things with these deposits. Management finally acted out the play — a Greek tragedy – after the stage was set.

Some of the blame goes to regulators who should have been aware of this risk. I hope the response in Washington, D.C. will not be, “We need to hire more regulators to do the job that the current regulators were not doing!”

Did you know that the Federal Reserve was SVB’s primary regulator? Yeah, “that” Federal Reserve. The one where career employees can fail upward. The one that somehow manages to get more power from Congress after each boom-bust cycle that it enables.

But most of the blame goes to SVB’s management team. It could have avoided failure if it understood the risk of loading up on long-duration securities when these securities offered yields below 2%. But it gorged on these securities ahead of the 2022 wreckage in bond prices. Now we see where that led.

What does this mean for banks and bank stocks going forward?

It’s likely to cause investors to put a premium on bank stocks with the stickiest, most loyal deposit bases. And it’s likely to cause a de-rating of banks that use wholesale or brokered deposits (flighty money).

It may also raise the demand for interest rate hedges as banks look to protect their securities portfolios from the wild price swings experienced by SVB’s portfolio. The second-order consequence of last week’s drama at SVB may be to shorten the expected path to Fed rate cuts.

Other bank stocks sold off — not due to exposure to SVB losses, but due to concerns that they may have to dramatically boost the rates they offer on deposits. For the past year, deposits were so plentiful and undesired by banks that the banks offered rates far below rates on Treasury bills. They often took the cash given to them by depositors, bought Treasury bills, and pocketed the difference.

Last week may have accelerated the move higher in bank deposit rates until they are closer to Treasury bills. If so, the banking sector’s net interest margin will shrink. And a shrinking net interest margin means there will be less income for banks to cover future loan defaults.

Finally, if banks have less income to cover defaults, this might concern the Fed enough that it will contemplate a rate cut.

What I just outlined is a sequence of events that could take a year or so to unfold – or it may only take a few weeks or months. Jim and I will look for signs that this process is happening slowly or quickly.

However quickly it unfolds, the key point is that the Fed is not going to cut interest rates for reasons that favor the stock market.

Conditions are changing rapidly. On Sunday night, The Fed announced a lending program that resembles the programs they rolled out in 2008 and in March 2020. It allows banks like SVB who are caught with unrealized losses in their longer-duration securities portfolio to borrow against them with no haircut. And with no penalty rate.

Unlike 2008, there is a clear, liquid market for agency-guaranteed (Fannie and Freddie) mortgage-backed securities. People know what the current market value is for these securities. The issue is who is going to bear the losses. And with this lending facility — which will be funded by extra FDIC insurance premiums — it looks like it will be borne mostly by responsible banks.

In short, Sunday night’s announcement went a long way toward socializing risk in the commercial banking system, unfortunately. By this, I mean responsible banks — those that were prepared for interest rate risk — will start to subsidize irresponsible banks.

This is a very generous lending program that will stem bank runs for the time being. However, in Monday morning trading, smaller and medium-sized banks that are stuck with unrealized securities portfolio losses were selling off dramatically.

Monday’s bank sell-off despite the Fed’s new loan facility means that investors view the reputation of these banks and the stickiness of their deposit franchises as questionable.

It’s too early to draw that conclusion now, but Jim and I will stay on this rapidly unfolding story.

Finally, from a macro perspective, expectations for Fed rate hikes have changed dramatically in the last few days, and gold is gaining momentum. It could be the beginning of a substantial multi-month gold rally.