by Donald van Deventer
Are we on the brink of a new banking crisis? In March 2023, California’s Silicon Valley Bank collapsed. Rising interest rates had made refinancing existing loans expensive and difficult. Although the US government intervened to avert a widespread banking crisis, the financial services industry cannot ignore the alarming signs. Asset liability management must be central in months ahead.
When it emerged in March that Silicon Valley Bank was in troubled waters, customers rushed to withdraw their deposits. Within the span of 48 hours, the bank run forced a California regulator to close Silicon Valley Bank and appoint the Federal Deposit Insurance Corporation (FDIC) as receiver. Nasdaq suspended SVB Financial Group’s stock on the stock exchange.
To prevent an even bigger crisis, the US government announced that it would guarantee Silicon Valley Bank’s uninsured deposits without cost to taxpayers. This intervention was necessary to protect the economy and secure public confidence in the American banking system.
Of course, Silicon Valley Bank’s default risk had been growing for some time. And given the subsequent failures of Signature Bank and Credit Suisse, it seems clear that the elevation of such risks is not isolated. As one of Warren Buffet’s most famous quotes goes: “Only when the tide goes out do you discover who’s been swimming naked.” There have been many warning signs pointing toward increasing deficiences in interest rate risk management.
One-factor interest rate
In recent conversations with banking risk managers from around the world, it’s common for me to hear statements like:
- “We don’t use default probabilities in our interest rate risk analysis”
- “We use a one-factor yield curve model to manage risks”
- “We do interest rate risk once a month without fail”
The quote about the one-factor interest rate risk framework is of particular concern. Bankers who rely on one-factor interest rate models are at serious risk in the current environment. Valuations are wrong, capital adequacy calculations are wrong, hedges are wrong. And the correlation of rates, oil prices and commercial real estate is completely ignored. Yet the latter has already devastated banking and loan industries twice in the past.
How to ensure optimal financial risk management
In the halcyon days of historically low interest rates, most analytical professionals involved in interest rate risk analysis moved to other banking disciplines or left for non-regulated financial services firms. This is a significant problem for financial firms now that the interest rate tides have turned and asset liability management has become of paramount importance. So, how can the industry promote optimal financial risk management?
First, financial institutions must take stock of their overall risk posture, starting with liquidity and deposit balance sheet risks in light of rising interest rates. They should also study data from fallen firms to better understand the underlying causes of their failures and the data patterns (like deposit outflows) that may signal growing default danger.
Second, integrated balance sheet management should be much more than ticking the regulatory boxes. In fact, regulatory compliance is a bare minimum. Banks must hold themselves to a higher standard – and they must also remember that they work for shareholders, not regulators.
Finally, banks cannot afford to shortcut robust stress testing and scenario analysis. AI-powered statistical models and machine learning algorithms can help banks to simulate different scenarios and predict potential impacts. Relaxing the rules has consequences, and resilience necessitates vigilance.
A holistic, granular approach to balance sheet and liquidity risk management will be vital for financial institutions navigating the current economic uncertainty. Asset and liability management, supported by the right technology and analytical expertise, can help banks to proactively manage risks and avoid the next Silicon Valley Bank scenario.
The author, Donald van Deventer, is Managing Director of Risk Research & Quantitative Solutions (RQS) at SAS.