Fed’s Bowman Challenges Social Media Story Behind SVB Collapse
Silicon Valley Bank’s collapse became shorthand for the digital bank run. Billions of dollars left quickly while warnings and speculation circulated online. Three years later, an independent review is challenging the connection between those two events. “[W]hile many postmortem accounts have asserted that social media fueled the run at SVB, none of these accounts made […] The post Fed’s Bowman…
An independent review is challenging the notion that social media played a role in the Silicon Valley Bank (SVB) collapse. Federal Reserve Vice Chair for Supervision Michelle Bowman announced initial findings on Friday, stating that social media did not trigger the bank run and there was no evidence that it accelerated the run. The review, conducted by Charles River Associates at the request of Starling Advisory Group, found that 96% of social media discussion about the run appeared after SVB's failure was inevitable.
The study highlights the importance of understanding the characteristics of a bank's deposit book, such as uninsured deposits, large balances from similar industries, and the potential for simultaneous withdrawals. SVB had a run-prone deposit base that was largely uninsured and concentrated in venture capital-backed technology companies.
The bank also had unrealized securities losses exceeding its capital and lacked operational readiness to borrow from the Fed's discount window when it needed liquidity. While the Fed recognized these vulnerabilities as early as March 2022, it did not take prompt and decisive action to address them, according to the review. The Fed is now implementing changes to its supervisory process, including monthly reports to supervision leaders and reserve banks when examiners are uncertain about the need for action.
The policy debate surrounding deposit insurance and potential changes to coverage continues, with implications for commercial customers who hold balances above the standard insurance limit.
Written by urgent.news from PYMNTS's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.