Federal Reserve Vice Chair for Supervision Michelle W. Bowman said an independent review of Silicon Valley Bank’s 2023 collapse found that Federal Reserve supervisory staff identified, or should have identified, the bank’s critical vulnerabilities well before its failure but did not move quickly enough to force corrective action. Speaking in London on September 18, Bowman said the Starling Advisory Group’s initial work points to a breakdown not only at SVB but also in the Fed’s supervisory process, particularly around escalation, decision-making authority and the willingness of examiners to act when risks were material but the precise supervisory response was uncertain.

The findings reopen one of the most consequential bank-supervision episodes of the past decade. SVB failed in March 2023 after a rapid loss of depositor confidence exposed a balance sheet that was highly sensitive to rising interest rates and heavily dependent on uninsured deposits. The new review identified three central vulnerabilities: unrealized accounting losses on securities that exceeded the bank’s capital, a deposit base that was 94% uninsured and concentrated among venture-capital-backed technology companies, and insufficient operational readiness to borrow from the Federal Reserve’s discount window when emergency liquidity was needed. Bowman said supervisors knew, or should have known, about those weaknesses as early as March 2022.

The most consequential conclusion for the supervisory framework is that awareness did not translate into timely intervention. According to Bowman’s summary, supervisors did not take prompt and decisive steps to encourage or require SVB to reduce its interest-rate exposure or the concentration of vulnerabilities that made the bank especially susceptible to a run. The review attributes part of that delay to a longstanding culture of risk aversion inside supervision. Staff, Bowman said, often believed it was professionally safer to refrain from acting unless they could be certain that the contemplated action was exactly right. That hesitation was compounded by unclear decision rights and by a separation of responsibility, authority and accountability that left staff unsure who could validate a proposed supervisory response.

That diagnosis matters because bank supervision is designed to work before a problem becomes a resolution event. Capital and liquidity rules set minimum requirements, but supervisors are expected to identify institution-specific weaknesses, challenge management, escalate concerns and require remediation when a bank’s risk profile threatens its safety and soundness. The independent review suggests that the key failure at SVB was not simply a lack of information. Rather, the system had information about material risks but did not convert that information into action with sufficient speed and force. For banks, investors and regulators, that distinction shifts attention from whether warning signs were visible to how supervisory organizations make decisions under uncertainty.

The new findings also differ in an important respect from the Federal Reserve’s April 2023 internal review led by then-Vice Chair for Supervision Michael Barr. That earlier review concluded that SVB’s board and management failed to manage basic interest-rate and liquidity risks, that supervisors did not fully appreciate the extent of the vulnerabilities as the bank grew, and that supervisors were too slow and insufficiently forceful after problems were identified. It also said the post-2018 regulatory tailoring framework and a shift in supervisory policy impeded effective supervision by reducing standards, increasing complexity and contributing to a less assertive posture.

Bowman’s September 2026 account reaches a different preliminary conclusion on that last point. She said the independent review found that delays in supervisory action were not caused by the 2018 Economic Growth, Regulatory Relief, and Consumer Protection Act’s tailoring mandate or by a directive from the former vice chair for supervision to reduce supervisory intensity. Bowman noted that the prior vice chair had left that role in October 2021, before the period in 2022 when the review says SVB’s vulnerabilities became most apparent. The contrast between the two reviews is significant because it affects how policymakers may distinguish failures of regulation—the rules that applied to SVB—from failures of supervision—the way examiners used their authority within the rules that existed.

The independent review also challenges a prominent explanation for the speed of SVB’s collapse: that social media materially triggered or accelerated the run. Bowman said Charles River Associates, working at Starling’s request, found no evidence that social media triggered the run and no evidence that it accelerated the failure. The analysis cited in her remarks found that 96% of social-media discussion related to the run occurred after SVB’s failure had already become inevitable. That does not make digital communication irrelevant to modern liquidity risk, but it redirects the causal focus in this case toward the bank’s underlying funding structure, balance-sheet losses and inability to withstand a rapid loss of uninsured deposits.

For bank risk officers, the practical message is broader than the historical dispute. Large concentrations of uninsured deposits can leave an institution vulnerable to rapid outflows, particularly when depositors share similar business models, investors or information networks. Duration risk in securities portfolios can weaken economic capital even when accounting rules delay recognition of losses in regulatory capital. Contingency funding plans are only useful if institutions are operationally prepared to mobilize collateral and access central-bank liquidity under stress. The combination of those factors can produce a nonlinear liquidity event in which confidence deteriorates faster than management can sell assets, raise capital or replace funding.

Bowman said the Fed has already begun changing its supervisory approach rather than waiting for the independent review to be completed. The central bank’s supervisory operating principles place greater emphasis on identifying significant threats to bank safety and soundness and U.S. financial stability early, then taking timely and proportionate action to mitigate them. The framework directs examiners to prioritize risks capable of causing significant financial harm instead of giving excessive weight to procedural or documentation deficiencies. It also broadens the range of supervisory responses by using observations alongside matters requiring attention and formal enforcement actions, giving examiners more ways to calibrate responses to the seriousness of a problem.

The Fed is also adding a new escalation mechanism aimed directly at the cultural issue identified in the review. Bowman said examination teams will submit monthly reports to senior supervision officials and the relevant Reserve Bank leadership identifying cases in which examiners are uncertain whether a supervisory threshold has been met or whether a proposed action is consistent with leadership expectations. The stated goal is to make it easier for examiners to surface uncertainty without fearing that an imperfect judgment will carry personal consequences, while giving senior officials real-time visibility into areas where policy or decision rights are unclear. If implemented as described, the process could shorten the time between identifying a material risk and obtaining a decision on how forcefully to respond.

In a separate speech the same day on stress testing, Bowman used SVB to argue for more forward-looking scenario analysis as a supervisory tool. She said a plausible recession scenario that combined rising interest rates with fair-value treatment of investment securities would have shown SVB’s capital falling below minimum required levels as early as the fourth quarter of 2021. That example was presented as an illustration of how additional scenarios could help supervisors identify idiosyncratic and emerging risks earlier. For the industry, the implication is that stress testing may increasingly be viewed not only as a capital-setting exercise for the largest banks but also as an analytical discipline that can inform supervisory prioritization before vulnerabilities become acute.

The review therefore lands at the intersection of two competing demands on financial oversight: supervisors must act early enough to prevent avoidable failures, but they also need clear standards so interventions are consistent, proportionate and defensible. A system that demands certainty before action can become too slow; a system that rewards intervention without disciplined thresholds can create unpredictability for regulated institutions. The Fed’s current response attempts to address that tension by focusing examiners on material financial risks, clarifying escalation channels and expanding the set of supervisory tools available before formal enforcement becomes necessary.

For regional banks and their investors, the near-term significance will depend on how these principles affect examination behavior in practice. Banks with concentrated funding, large unrealized securities losses, rapid balance-sheet growth or weak contingency liquidity capabilities may face greater pressure to remediate risks sooner. Management teams may need to demonstrate not only compliance with formal liquidity and capital ratios but also credible plans for adverse scenarios involving simultaneous deposit outflows, collateral demands and changes in market value. Supervisory findings that once might have remained in extended internal deliberation could be escalated more quickly if examiners believe the issue threatens viability or broader financial stability.

There are also implications for capital markets and bank valuation. Investors typically evaluate banks using reported capital, deposit costs, asset quality and earnings, but SVB demonstrated how economic losses and funding concentration can become decisive before conventional credit deterioration appears. A supervisory regime that intervenes earlier could require banks to raise liquidity, reduce duration, change funding composition or hold additional capital sooner, affecting profitability in the short run while potentially lowering tail risk. The trade-off will be closely watched by bank shareholders, creditors and asset managers because supervisory judgments can influence balance-sheet strategy even when formal regulations do not change.

At the same time, the September 2026 findings are explicitly initial findings, and Bowman said the Starling work is the first in a series. Her prepared remarks also state that the views expressed are her own and not necessarily those of other Federal Reserve Board members or the Federal Open Market Committee. That makes it important to distinguish the review’s preliminary conclusions from a final institutional judgment by the Fed. The divergence from the 2023 review—especially over the role of regulatory tailoring—will remain an important issue as the independent work continues and as the Fed decides which lessons should be incorporated into permanent supervisory policy.

The central conclusion announced on Friday is nevertheless direct: the supervisory system had enough information to recognize severe vulnerabilities at SVB well before the bank failed, but it did not act with sufficient urgency. For the financial sector, the lasting question is whether the Fed’s new operating principles, escalation channels and risk-focused supervision can convert early warnings into timely corrective action without creating a less predictable supervisory regime. The answer will shape how regional banks manage liquidity and interest-rate risk, how examiners use discretion, and how markets assess the probability that the next institution-specific vulnerability is contained before it becomes a broader financial-stability event.