Customers continue to enter and exit the Menlo Park branch on Sand Hill Road, which has been repainted in First Citizens green, as if nothing had happened. That is a portion of the issue. The physical evidence has disappeared three years after Silicon Valley Bank disappeared in about thirty-six hours, and the muscle memory has followed suit. Today, if you walk into any San Francisco founder dinner, the topic of discussion has shifted to hiring in Bangalore, model training expenses, and AI compute costs. Suddenly, banking risk seems like a 2023 issue.
It isn’t. Eighty-six percent of corporate financial decision-makers still hold more than the FDIC insurance limit at a single institution, according to the Ampersand Cash Confidence Survey published last year. If that bank failed tomorrow, the average runway would be less than three months. Those figures ought to be startling. They aren’t because everyone believes that regulators will use the systemic risk exception and reimburse depositors within a weekend, as they did in March 2023. It’s a fair wager. It’s a wager as well.
No one had really factored in the velocity of the initial collapse. On March 9, 2023, SVB lost $42 billion in deposits in a single day, and its founders witnessed the panic spread via Signal chats and WhatsApp groups. The bank was the sixteenth largest in the nation, with $209 billion in assets. Fraud or even poor lending wasn’t the issue. On long-dated Treasuries, there was a duration mismatch during a rate cycle that no one had stress-tested for. The danger of textbooks. The one that regulators believed had been resolved following the savings and loan crisis decades ago. Since then, Stanford researchers have contended that the same blind spots still exist, albeit in different locations.
Observing this from the outside, it’s interesting to note how little the underlying setup has changed as a result of the tech sector’s behavioral response. The founders relocated their funds. The architecture was not altered. The majority still focus their operating cash at one or two banks that they deem “tech-friendly,” which these days refers to newer fintech sweep accounts rather than SVB. The names were altered. The pattern didn’t. Speaking with city treasury consultants, it seems like the post-SVB era produced a lot of “Banking Diversification” Slack channels but very few actual policies.
There is a structural component to the problem. Large legacy banks continue to be uncomfortable partners for startups, slow to open accounts, and unconcerned with venture-backed cash flow. This was stated clearly by the UK founders during the initial collapse, and the trade-off hasn’t really been settled since. Although they focus on the threat, the specialized banks comprehend their customers. While avoiding the risk, the high-street incumbents hardly ever return calls. Nothing less than persistent regulatory pressure may be able to close that disparity.
The question of contagion is the other incomplete piece. SVB failed because depositors used social networks that the bank was unable to monitor and moved at a rapid pace. That infrastructure is still in place. With AI-driven financial commentary, founder group chats, and viral threads that can move billions in a matter of hours, it’s arguably more potent now. Rules for a slower world are still being written by regulators. The type of run SVB had was never intended to be addressed by the Basel III endgame proposals.
It’s difficult to avoid thinking that the lesson has been filed away rather than absorbed as you watch this play out. Until the bleeding stops, the tech industry adores a post-mortem. After that, it continues.
