SVB and Interest Rate Risk Basics

Ksusha McCormick

March 14th, 2023


The collapse of SVB is a sobering reminder of a risk that lay dormant for the better part of the past decade - interest rate risk. Prior to Covid and the Fed’s hiking cycle, interest rates were effectively pegged at zero, and it was during those years than many lending and fintech startups were launched. By not explicitly measuring and managing their interest rate risk, many of these companies effectively baked in the assumption that the happy state of affairs would last forever.

SVB was a large, sophisticated institution, and its failure was multifactorial. Differences between big bank and regional bank regulation, depositor demographics and behavior, and the inverted yield curve all played a role. However, if you think of risk as a layer cake, interest rate risk is the bottom, most fundamental layer, and it’s instructive to look at the composition of SVB’s investment portfolio in order to gain an intuitive understanding of the way interest rate risk should be managed by lenders of all types, even smaller, less traditional ones.

In 2008, a lot of the trouble with financial institutions stemmed from the fact that they held tranches of securities where the underlying instruments weren’t rated AAA but the tranche itself was considered to be AAA because it was last to participate in any default-driven losses. In the case of SVB, however, nothing of the sort occurred. SVB invested in Treasuries and mortgage-backed securities – just about the highest quality portfolio that the US dollar can buy. Why, then, was that a problem? The problem is duration risk, which is the sensitivity of bond prices to changes in the prevailing market yield. A bond is essentially a stream of future payments, exactly known in advance and guaranteed to occur if the bond issuer doesn’t default. These cash flows are discounted to the present at the prevailing market rate. When the rate rises (as it does when the Fed commands it to), the NPV of these cash flows falls. This happens even if there’s no change whatsoever to the perceived credit worthiness of the borrower, i.e. the bond issuer.

When I was a wealth manager, I had an elderly client who, after much discussion, invested in a portfolio of high quality municipal bonds. The next day, rates went up, the value of his portfolio was marked down, and he called me, demanding an explanation. At that point, I had almost a decade of experience including seven years as a bond trader and portfolio manager, but I found myself struggling to explain what happened in purely non-financial terms. In the end, I told him: “Let’s say you bought a tie for $100. A week later, you walk by the store and find that that same tie is on sale for 20% off. Are you happy? No! But if you’re wearing your tie and don’t need to sell it to pay your bills, you don’t have a problem.”

This is far from a perfect analogy, but the fact remains: when you buy a highly rated bond, you are guaranteed to receive exactly the fixed coupon payments and the principal back at maturity. And barring default, mark-to-market losses will only turn into realized losses if you sell this bond prior to maturity. I’ll skip over mark-to-market risk and the nuances of bank accounting rules, but SVB suddenly found itself needing to sell its portfolio of investments in order to meet the needs of depositors, who had demanded their money back en masse.

A bank’s assets are the loans and investments on its books, and its liabilities are the deposits, so what happened to SVB is in some sense a classic case of asset/liability mismatch. To translate this into bond math-ese, deposits, which the bank must return whenever depositors want them, are ultra-short duration instruments, whereas Treasuries and mortgage-backed securities have a duration of many years. A related problem here is the mismatch between fixed-rate and floating-rate debt. Since the interest that banks pay depositors typically “floats” with changes in the prevailing short-term rate, most regional banks prefer to issue floating rate loans. This way, sudden changes in interest rates will pass through, impacting assets and liabilities in roughly equal measure. If interest rates rise and the bank has floating rates on both deposits and loans, then it’s easier to pay depositors more because the payments from your loan book will have increased. This is a simplified version of reality, of course – the curve can change shape and so forth, but as a general rule, it’s better to be matched than mismatched. The long duration, fixed rate risk of SVB’s portfolio was the primary driver of its unfortunate collapse.

What are the lessons here for smaller, non-traditional lenders? Even high-quality assets are not necessarily “safe” if there’s a fixed/floating mismatch or an asset/liability mismatch in your business. Interest rate risk is the most fundamental risk for bond and loan portfolios, and requires robust management even during periods of low volatility. Good risk management is not a guarantee that you’ll never lose money, and it’s hard to make money without some risk. Therefore, the right set of tools should make it easy for you to determine whether the risks you’re taking are the ones you mean to take.

"Black Red" by Clem Onojeghuo is marked with CC0 1.0.

Previous
Previous

Decameron Technologies Case Study: Applications for Life Insurance Companies

Next
Next

Fixed Income ETFs: Empowering New Managers to Succeed