by Kasper Halevy (’24) | March 31, 2023
One million dollars a second. That is how fast clients of Silicon Valley Bank (SVB) withdrew deposits on March 10, causing the Federal Deposit Insurance Corporation (FDIC) to take over the failed bank by midday. This was the second-largest bank collapse in United States history since the demise of Washington Mutual in 2008.
The closest thing to a rationale that explains the bank run is SVB’s poor risk management. Having experienced a meteoric rise in deposits from 2020 to 2022, the bank had to decide how to use its newfound money. Lending was a limited option because SVB primarily caters to an already cash-flush Bay Area ecosystem of tech companies and start-ups. Instead, the bank purchased longer-term bonds with higher yields to provide competitively higher rates for clients. However, rising interest rates due to record inflation diminished the market value of the government-backed securities, forcing the bank to sell its bond portfolio at a $1.8 billion loss in an effort to improve liquidity. There had been no hedge against the possibility of rate hikes, which is a fairly elementary financial mistake. Alarmingly, SVB also operated with no chief risk officer for its final nine months.
Despite the evident fiscal irresponsibility, were both the depositors’ and regulators’ responses appropriate?
As noted countless times recently, the banking system is more reliant on trust than assets. If the vast majority of customers at the “too big to fail” banks such as JP Morgan or Bank of America lost confidence and simultaneously withdrew their funds, they too would collapse in spectacular fashion. For a bank to flourish, its client base must believe that its deposits will be available whenever they need them.
Unsurprisingly, Twitter is not conducive to achieving that level of trust. The publicization of SVB’s management woes was overwhelmingly amplified by the platform, resulting in prominent venture capitalists sounding the apocalypse alarm publicly and privately to their portfolio companies. As René Girard’s theory of mimesis posits, our desires are usually not intrinsic or “spontaneous” but rather often imitative of wants and suggestions from others. For most clients, the idea that their bank could be in peril or that there would be an urgent need to take back their funds would not have risen without the storm of cautionary tweets. “As a startup…we are facing all kinds of risks, technology risks, product risks, GTM risks, competition risks, and the list goes on…we never thought about bank risks,” said Sam Liang, founder and CEO of Otter.ai. While SVB’s imprudence should have addressed, the prolific and intentionally provocative circulation of the idea that SVB was in trouble was a debacle waiting to happen. SVB’s collapse was driven more by a herd mentality imitating models for action such as Peter Thiel and Jason Calacanis than sound economics; the bank’s wobbly balance sheet alone was not enough to justify its implosion.
Another complication with the fallout of SVB was that it created a classic Prisoner’s Dilemma scenario where self-preservation made more sense than cooperation. As an individual depositor, there was little incentive not to defect and withdraw your funds. The fear of having money erased or frozen in a defunct bank outweighed the gamble of participating in a joint heroic effort to support a financial institution. Although I believe that the criticism of the lack of coordination among an ostensibly loyal client base is valid, the game theory asserts that each individual depositor’s “betrayal” was statistically the most rational decision—especially when levels of trust are low.
The government responded by promising to reimburse all deposits, including approximately 90% of those that were over the 250k insurance threshold, at no cost to taxpayers. More specifically, the FDIC invoked the “systemic risk” clause, asserting that the failure and loss of all uninsured deposits at SVB would threaten to upend the broader economy. Dissenters pointed out the dangerous precedent that the Fed is setting in regards to moral hazard: if a mismanaged SVB can be bailed out, what is deterring other institutions from taking on additional risk at no cost?
I believe that the Fed chose the lesser of two evils. There are several other ways to alleviate the moral hazard problem, including the reinstatement of regulations similar to those of Dodd-Frank and the investigation into and punishment of executives. Rightfully, equity holders, including executives, will incur full losses. On the other hand, not guaranteeing depositors their funds would send a much more problematic message—that one cannot be confident in the banking system. While the collapse of one or two mid-size banks is likely not systemically significant in purely economic terms, the swift spread of the SVB mimetic wildfire facilitated by social media is already being felt by First Republic Bank. In 2008, despite the unjust bailouts of reckless lenders, relief of some sort was necessary to prevent the onset of a second Great Depression. Similarly, relief in the form of deposit repayments, even if it involves a bailout and may not technically be the correct economic move, coupled with more stringent regulations and stress tests can be crucial in staving off broader panic.
