“Street Speaks in Swap Land” — Marcy Stigum
The collapse of Silicon Valley Bank (SVB), and its aftermaths, in March 2023, showed a structural change in the business model and the risk structure of deposit-taking institutions. In 2008, Great Financial Crisis (GFC) revealed that the banking system’s balance sheets are ingrained with liquidity risk. Bankers borrowed in the short-term, liquid money markets to invest in long-term illiquid assets with high yields. In contrast, the March 2023 crisis showed that deposit-taking institutions had shifted their gear towards investing in liquid assets such as government bonds. In doing so, they have become hedge funds in disguise. Nonetheless, instead of noticing such changes in banking structure, regulators assessed commercial banks based on the lessons of the GFC. They were considered safe as long as they had healthy liquidity and leverage ratios and were funded by deposits. By over-relying on the lessons of the GFC, the regulators and bankers’ risk managers alike disregarded a risk that every hedge fund manager and fixed-income investor is alerted by: interest rate risk.
Interest rate risk was disregarded in the narrative of financial stability. Moreover, GFC was partially to blame. GFC exposed the extent of liquidity risk in the banking system. The liquidity mismatch between the banks’ assets (long-term illiquid assets) and liabilities (short-term money market instruments) became known as the hot spot of banking. The liquidity mismatch would cause a solvency problem if the bank, for instance, needed to sell some of its assets quickly to manage its daily survival constraints and cash flows. In this case, illiquid assets would go through a fire sale process not because they had lost their potential income or become less attractive but simply because they did not have a liquid market. Such circumstances led to Bagehot’s dictum that to avert panic, central banks should lend early and freely (i.e., without limit), to solvent firms, against good collateral, and at “high rates.” In this context, it should not be surprising that the Fed and FDIC provided extensive liquidity provisions, including offering blanket deposit insurance, after the March 2023 banking crisis started.
Nevertheless, such facilities have yet to calm the market. The failure of liquidity provision to stabilize the market in the March 2023 banking crisis is partially generated by the intellectual mismatch between the root of the banks’ vulnerability (interest rate risk) and the proposed remedies (liquidity backstops). The reason is the change in the banks’ business model. Commercial banks started to hold excessively safe assets, such as government bonds, to prevent a GFC-like crisis and escape regulatory pressure. Government bonds may not have default risk. They are also liquid. However, their market value goes down when rates rise. In addition, rising interest rates generally force banks to raise deposit rates or lose funds to alternatives such as money-market funds. For banks, that was only an issue if the bonds were not adequately hedged and had to be sold to redeem deposits, which is exactly what happened to SVB. The signature relied more on loans, but it also experienced a run on its uninsured deposits.
Regional banks’ business model exposes them to unusually high interest rate risks. Regional banks’ liabilities are mostly deposits from modern corporates such as Venture Capitals (VCs), Startups, and Crypto firms. These corporate depositors do not need a bank loan. Instead, they can raise cheap funding through equity, IPOs, and other capital market techniques. As a result, banks use their cash to purchase a very high level of interest-sensitive fixed-income assets such as bonds. The classic problem with holding a large portfolio of fixed-income assets is that when rates go up, they fall in value, as with SVB’s assets. At the same time, as deposits pay competitive rates, higher rates increase the value of banks’ liabilities. This is called “interest rate risk.” All types of deposit-taking activities involve a certain level of interest rate exposure. However, commercial banks’ portfolios used to be more diversified as they also made floating-rate corporate loans. As a result, their balance sheets were less sensitive to interest rate changes in the past.
The heightened sensitivity of banks’ assets to interest rate risks could stay unrecognized by more traditional depositors who treat checkable deposits as safe as government liability. However, like equity investors, corporate clients use all the available information to continuously mark-to-market bank assets. When interest rate is volatile, as was the case in the past few years as a result of the Fed’s policies, bond prices change dramatically. If these assets are marked to market, their fair value sometimes falls below their book value. More informed and rational depositors are impatient and reactive to such developments. As was the case for the SVB, they could hugely penalize the banks by collectively withdrawing their funds and creating a run on a bank. The SVB-derived banking crisis showed that compared to other depositors in history, these new and individually rational types of corporate depositors could collectively create a more unstable banking system.
As a result, regulators are deciding how to secure this segment of the banking system that is unusually exposed to interest rate risk. Nonetheless, regulators should consider more innovative risk management approaches instead of returning to the standard regulatory toolkits, such as stress tests. For example, they could require the banks with such a business model to use interest rate swaps (IRS). First, from a risk management perspective, IRS can provide interest rate hedges. As corporate depositors continuously mark-to-market banks’ financial positions, neutralization could help calm their nerves when interest rates are highly volatile. IRS could also act as a cash management tool. The parallel loan structure of the IRS synthetically transforms the banks’ fixed-income assets into floating-rate- assets to match deposits’ cash flows.
From a classic risk-management perspective, swaps would neutralize the interest rate risks. To understand this point, let us go through an example. Suppose a regional bank tends to issue a $ 1 million deposit at a floating rate. The bank uses this fund to purchase a fixed-income bond. However, additional liability (deposit) at the variable rate will undermine compliance between interest rate-sensitive assets and liabilities. In the event of rising interest rates in the market, banks’ cash outflows and cash inflows increase. The cash outflow increases as the banks make higher interest payments to the depositors. The value of the cash inflow increases too. Even though the bonds generate fixed cash flows, these payments will be reinvested at a higher interest rate and earn a higher income. However, let us assume that the value of the liabilities will be greater than the increase in the income value by one million dollars. The result is a decline in the net interest margin and bankers’ profitability.
To avoid this risk, the banker can convert $ 1 million of liabilities with variable interest rates in the $ 1 million liability insensitive to interest rate movements, tiding interest-sensitive assets to interest-sensitive liabilities. Entering into an interest rate swap will enable her this. Therefore, the banker will contract an interest rate swap under which she will be required to pay at a fixed rate and receives at a variable rate. Variable income from the swap will equal the losses from the additional variable liability, and the net result will be a fixed obligation from the swap. In other words, profit/loss in swap would neutralize variable income from bonds, and the net result will be an interest-insensitive asset and liability because of the swap.
Another way to think of the swap is as a tool that matches cash inflows and outflows synthetically. The mechanics of the swaps can allow the banks to convert their bond holding (that earns fixed income) into repo lending (that earns a floating rate), albeit synthetically. Buying an IRS (being a fixed-payer, floating-receiver) by the bank is like borrowing in the bond market to lend the proceeds in the short-term money market. Banking is the equivalent of borrowing short and lending long. In contrast, IRS is equivalent to borrowing long and lending short. In this scenario, the swap position increases in value when the floating interest rate rises and generates the cash flows required to neutralize the cash flow mismatch between the banks’ assets and liabilities.
Historically, risk managers and regulators have often tried treating such risk as an “accounting” problem. As a result, positions were converted into risk equivalents and added together. For example, in fixed-income markets, participants have, for many years, scaled their positions into units of a common duration. Each position is converted into a basis—for example, a number of “10-year duration equivalents”—which should have equal sensitivity to the main source of fixed income risk, a parallel movement in interest rates. In this case, risk managers and regulators use indicators such as the delta (the net interest rate sensitivity), the vega (the net volatility sensitivity), and the gamma (change in delta concerning a one bp change in interest rates).
While these bits of information are essential to understanding and managing the position, they do not provide an adequate basis for risk management. Over the past several years, the accounting approach to risk management has been largely supplanted by using “stress” tests. Stress tests are the output of an exercise in which positions are revalued in scenarios where the market risk factors sustain significant moves. No doubt, using stress tests improves a situation of not knowing what might happen in such circumstances. However, significant limitations in stress testing need to be recognized.
What are their important limitations? First, it is sometimes unclear which dimensions of risk need to be considered. Also, stress tests do not reveal the relative probabilities of different events. For example, a position with negative gamma that loses money in significant moves in either direction will look bad in extreme scenarios but generally look very attractive when only local moves are considered. In any case, the shrewd banker can tailor his positions to look attractive relative to any particular set of scenarios or, given the opportunity, can find a set of scenarios for which a particular set of positions looks attractive. Moreover, in complex portfolios, there are many scenarios to look at; in fact, it may be virtually impossible to know which risk factors need to be considered. Furthermore, even if an exhaustive set of scenarios is considered, how does the trader or risk manager know how to consider the risk reduction resulting from the diversification of the risk factors? Thus, while stress testing is useful, it often leaves large gaps in understanding risk.
The ongoing banking crisis shows that while we mistakenly disregarded liquidity risk as an anomaly before the GFC, we made the same mistake regarding the interest rate exposure until the collapse of the SVB. As long as banks held safe, liquid assets, the interest rate mismatch between assets and liabilities was considered systemically unimportant. It is true that in modern finance, liquidity kills banks quickly, and liquidity facilities save lives. Nonetheless, the SVB crisis shows that we are also moving towards a parallel world, where interest rate risk evaporates a whole banking ecosystem. In this environment, looking at other instruments, such as swaps, and other players, such as swap dealers, that can neutralize the interest rate risk would be more logical. The neutralization aspect of the IRS can provide a robust financial stability tool to hedge against systemically critical hot spots.
Nonetheless, at least two critical issues should be extensively discussed to better understand swaps’ potential as a tool to strengthen financial stability. First, the economic benefit of the interest rate swaps results from the principle of comparative advantage. Interest rate swaps are voluntary market transactions by two parties. Nonetheless, this comparative advantage is generated by market imperfections such as differential information and institutional restrictions. The idea is that significant factors contribute to the differences in transaction costs in both the fixed-rate and the floating-rate markets across national boundaries, which, in turn, provide economic incentives to engage in an interest-rate swap, is true “a market failure.” Second, any systemic usage of swaps would engage swap dealers and require expanding the Fed’s formal relationship with such dealers.