Bowman answers the SVB review with a log of examiner hesitation

The operative proposal in Bowman's Friday speech is a reporting line. Examiner hesitation becomes a logged item. The reports will identify any supervisory issue or concern where an examiner was uncertain "whether the standard for taking supervisory action was met, or whether such action might be inconsistent with the expectations of Board or Reserve Bank leadership" ^[4]. "This accomplishes two goals. First, it empowers our examiners to escalate concerns without fear, and it gives leadership real-time visibility into where our teams need clearer guidance," Bowman said ^[5].

Starling's preliminary work covers the same ground as the official post-mortems that preceded it. The diagnosis underneath is old. They overlap on unrealized losses and uninsured deposits on the balance sheet, a year-long paper trail of identified issues at the bank, and supervisors unable to take prompt action on them ^[8]. Starling founder Stephen J. Scott told American Banker that much of his effort in this first report went to substantiating the prevailing narratives ^[9]. The overlap is not a surprise. "This is very much a constructive undertaking," he said. "It's not meant to be a critical undertaking" ^[10].

Two findings cut against causes that had been widely accepted. Bowman said Charles River Associates, working at Starling's request, "concluded that social media did not trigger the bank run at SVB, and there was no evidence that social media accelerated the run." She said the firm found "96% of the social media chatter regarding the run appeared after SVB's failure was inevitable" ^[13]. Four percent of it, on those numbers, came before the bank was already gone ^[14]. On her predecessor, Bowman said the former Vice Chair, Quarles, had stepped down in October 2021, well before 2022, when SVB's vulnerabilities became most apparent ^[15]. Silicon Valley Bank failed in March 2023, seventeen months after that departure ^[20][21].

Where those two causes go, the remedy follows. A run driven by depositors reading a balance sheet, and a supervisory failure dated after Quarles left, point at examiner practice, and Bowman is asking for policy changes inside the Fed ^[6]. For banks the difference is cost. Guidance and escalation reporting do not add a capital or compliance cost, and examiners who escalate a year earlier bring supervisory action a year earlier on facts the Fed already held ^[8]. The counter-case is that none of this binds yet. Bowman called the report the first in what will be a "series" ^[19]. She did not release it; it runs hundreds of pages with hundreds more pages of appendices and source materials, and she made no commitment to publish it later ^[18].

If Starling's later rounds locate the failure in capital and liquidity standards instead of examiner behaviour, the uncertainty reports shrink to a process note and the rulemaking case comes back. On the evidence so far, Bowman has asked for changes within the central bank ^[6].