Follow Topic Follow Contributor Share Feedback
<p><font color="#14ABB2">Probative Inquiry into the <br>2023 Financial Sector Stress Events</font></p><p><font size ="6"><font color="#455664">1. A “textbook case” of mismanagement</font></font></p>

Probative Inquiry into the
2023 Financial Sector Stress Events

1. A “textbook case” of mismanagement

by Starling Insights

Starling Insights Editorial Board

Sep 21, 2026

Feature

1. A “textbook case” of mismanagement

“Silicon Valley Bank (SVB) failed because of a textbook case of mismanagement by the bank,” then-Vice Chair for Supervision Michael Barr’s opening letter accompanying the Barr Report begins. “Its senior leadership failed to manage basic interest rate and liquidity risk. Its board of directors failed to oversee senior leadership and hold them accountable.”1

First among the Barr Report’s “key takeaways” is the observation that SVB suffered from “foundational and widespread managerial weaknesses.” For instance, “The full board of directors did not receive adequate information from management about risks at Silicon Valley Bank and did not hold management accountable for effectively managing the firm’s risks,” the Barr Report asserts.2

Such assertions have been oft-repeated in other postmortem accounts. Drafted in four weeks, between March 21st and April 28th 2023, the GAO’s postmortem review indicates a purposeful focus “on supervisory activities related to the banks’ liquidity and risk management.”3 Under the rubric, “Weak Liquidity and Risk Management Contributed to Bank Failures,” the GAO reiterates the Fed’s findings: 

According to Federal Reserve Board and FRBSF staff and their examination documents, SVB failed due to ineffective risk management, including the management of its deposits and assets. FRBSF documents stated that SVB’s risk-management framework was not commensurate with the bank’s size and complexity.4

In June 2022, the GAO reminds, the Federal Reserve Bank of San Francisco (“FRBSF”) downgraded SVB’s CAMELS composite rating from a 2 to a 3, its management component rating from a 2 to a 3, and its liquidity component rating from a 1 to a 2. The GAO reiterates the FRBSF’s reasoning for the downgrade:

Specifically, FRBSF examiners found that the bank’s management and board performance needed improvement and were less than satisfactory. For example, the board did not provide effective oversight of implementation of the risk-management framework and execution of the bank’s transition into the Large Financial Institution category. The board also did not hold management accountable for the root causes contributing to weaknesses in liquidity risk management and other risks.5

It should be emphasized here that these are not independent findings. These are re-statements of assertions put forward by the Federal Reserve Board and the FRBSF.

In its 2023 Material Loss Review, the Office of the Inspector General (“OIG”) of the Federal Reserve Board argued that SVB “exhibited weaknesses in corporate governance and risk management.”6 More specifically, the OIG found that “SVB exhibited several corporate governance weaknesses that created an unstable foundation to support the bank’s rapid growth and escalating risk profile. These corporate governance weaknesses included ineffective board of directors’ oversight and significant weaknesses in the bank’s lines of defense.” In sum, the OIG argues, “Management’s ability to mitigate risk did not keep pace with the bank’s rapid growth, and the board of directors failed to hold management accountable for implementing effective risk-management programs and controls commensurate with the bank’s increasing size and risk profile.”7 

And the OIG review seeks to illustrate the case: “For example, examiners noted that the bank’s incentive compensation program and executive officer performance evaluations did not meaningfully consider the bank’s risk-management deficiencies. In other words, risk-management weaknesses or deficiencies would not detract from an executive’s incentive compensation.”8

It should be noted, however, that here again, the OIG appears to have relied on findings presented by FRBSF bank examiners, rather than putting forward findings reached independently.

“The authors of the Barr, GAO, and OIG postmortem studies never sought to test their conclusions through discussion with the board of SVB.”

It is further notable that, despite the force and repetition with which findings regarding the board and management of SVB have been asserted, in the course of our Probative Inquiry, we have learned that the authors of the Barr, GAO, and OIG postmortem studies never sought to test their conclusions through discussion with the board of SVB, nor did the authors of these government studies issue any request for information to the board in the course of conducting their postmortem inquiries. 

Although the OIG and GAO conducted independent reviews within their respective statutory mandates, their conclusions regarding the conduct of SVB’s board and management were derived principally from supervisory findings already developed by the Federal Reserve System, rather than from independent inquiry. This necessarily limits the evidentiary basis upon which conclusions regarding board conduct were reached, and the established public narrative that these conclusions support therefore warrants a closer examination.

Interview with Mary Miller

Against that backdrop, Starling sought to obtain the perspectives of those who had themselves participated in risk governance at SVB, but whose views had not been sought during the preparation of the principal postmortem reviews. On March 17th 2026, Starling conducted an interview with Mary Miller, former Chair of the Audit Committee of SVB. The meeting took place at Ms. Miller’s request, conveyed through counsel, after she expressed a desire to contribute her perspective to Starling’s independent Probative Inquiry.9

Our discussion with Miller was not recorded so, unlike the majority of interviews conducted by Starling in the course of its Probative Inquiry, we do not possess a verbatim transcript reviewed and approved by her. This account was drawn from contemporaneous notes prepared by Starling personnel, and the quotations offered should be understood as language recorded in those notes.

Before joining SVB’s board in 2015, Miller had managed fixed-income, bond, money-market, international, and high-yield portfolios at T. Rowe Price. She later spent five years at the U.S. Treasury, where she was responsible for federal debt management and served on domestic-finance matters under Secretary Timothy Geithner and Deputy Secretary Jacob Lew. 

At SVB, Miller served as an independent director and chaired the Audit Committee.

Miller sought to speak with Starling because she was “shocked” by the Barr Report. She acknowledged that much of it was accurate, but regarded the account as profoundly one-sided. In her view, the Barr Report describes in detail what supervisors found, what they communicated to SVB, and what they required the bank to do; it says little about how management and the board responded.

That omission is central to Miller’s criticism. The Barr, GAO, and OIG reports portray a board that failed to oversee management, failed to demand adequate risk controls, and failed to hold management accountable for remediation. Miller does not deny that SVB received serious supervisory criticism. She instead argues that the postmortems omit the extensive — and, in her telling, increasingly urgent — efforts by the board and management to address that criticism.

For instance, Miller reported that SVB received 54 MRAs or MRIAs between 2020 and 2022, of which 40 were closed. As SVB crossed regulatory thresholds — from the Fed’s Regional Banking Organization (“RBO”) portfolio to the Large & Foreign Banking Organization (“LFBO”) portfolio — she recalled a distinct “cliff effect”: supervisory demands multiplied rapidly, many involving expectations that had not applied to SVB under its earlier supervisory regime.

Most of the early supervisory findings concerned technology and operational matters, including BSA/AML, cyber risk, and identity-access management. Their remediation consumed substantial board attention. Miller recalled a genuine sense of accomplishment when findings were closed. 

The demands eventually became so extensive that the Audit Committee risked being diverted from its core mandate, Miller contends. In response, the board established a separate Technology Committee, led by director Dick Daniels,10 to assume greater responsibility for IT oversight.

Explicit supervisory findings concerning liquidity risk arrived later. Miller recalled two MRIAs and four MRAs addressing that subject in fall 2021. She maintained that SVB submitted everything supervisors had requested in response by the deadline. Her question was direct: “Why weren’t those response plans mentioned in the Barr Report?”

Miller herself was assigned responsibility for an Internal Audit MRIA issued in 2022. She reported that SVB developed 46 areas of focus and approximately 100 associated action plans, all of which were shared with Fed bank examiners. Moreover, SVB engaged consultants to assist with other findings concerning governance and risk management, recruited senior risk professionals, and added approximately 2,000 employees during 2022, many in risk management and internal audit. 

“Supervisors repeatedly responded: ‘We can’t be prescriptive — you have to figure this out.’”

These measures do not, by themselves, establish that SVB’s response was adequate, of course. Activity is not the same as effective remediation, and the number of plans, personnel, meetings, and advisers does not reveal whether SVB addressed the risks that mattered most. But Miller’s account is difficult to reconcile with a simple narrative of an inattentive board that treated supervisory concerns with indifference. At minimum, it identifies a substantial body of evidence that must be examined before reaching conclusions about the board’s engagement and accountability.

Miller’s account also complicates the assertion that the board failed to ensure that supervisory concerns were resolved promptly. She indicated that she herself met quarterly with the San Francisco Fed and sought specific guidance on how SVB might satisfy the Internal Audit MRIA. In Miller’s telling, supervisors repeatedly responded: “We can’t be prescriptive — you have to figure this out.”

When Miller presented SVB’s proposed course of action to examiners, she recalled being told: “It looks like you’re on the right course.” Yet the process remained open-ended. SVB was expected to continue working until supervisors determined that the finding could be closed, without being given clear, measurable criteria by which such completion would be assessed.

Miller likened the resulting dynamic to a game of “cat and mouse,” with the bank repeatedly asking: “Is this enough? Is this enough?” She did not allege that supervisors had disregarded their prescribed processes. Her criticism was that those processes produced an inefficient and debilitating communications vacuum. SVB understood that supervisors regarded its response as incomplete but struggled to obtain concrete guidance on what would constitute an acceptable result.

The same pattern appeared elsewhere. Miller reported that the Compensation Committee received supervisory directives, worked to respond, and did not receive confirmation that its work was complete. Miller offered a similar example involving SVB’s resolution plan. The bank engaged an experienced resolution planning specialist and submitted a resolution plan to the bank’s regulators in December 2022. Miller received no subsequent feedback. She allowed that management might have received communications that were not relayed to her, but from her perspective as a director, the submission disappeared into a “black hole.”

The postmortems emphasize the existence and persistence of open findings as evidence that management and the board failed to remediate. Miller’s account raises a different possibility: that some findings remained open within an iterative process through which supervisors repeatedly sought further work but did not define the conditions for closure. Determining which account is more accurate requires examination of the underlying submissions, the Federal Reserve’s responses, and evidence regarding implementation — not simply a count of findings that remained open when SVB failed.

Miller was particularly troubled by the Confidential Supervisory Information released with the Barr Report. While serving as a director, she had understood that such material would remain confidential. Upon reviewing the released record, she discovered extensive internal Federal Reserve discussion of SVB’s interest-rate risk. “Why weren’t those concerns shared with me during quarterly meetings?” she asked.

Miller recalled supervisors describing the examination relationship as a “safe house” in which concerns could be discussed candidly. Yet the degree of concern visible in the Federal Reserve’s internal record was not, in her recollection, communicated to the SVB board with comparable force. She believes the Federal Reserve was itself late to appreciate the potential impact of a 500-basis-point change in interest rates, with the risk gaining urgency only during the latter half of 2022.

She also emphasized the informational advantage enjoyed by the supervisor. The Federal Reserve could compare SVB with institutions across the banking system. The board could see SVB’s position, but it could not independently see whether that position had become unusual relative to peers. Miller believes that, had supervisors clearly communicated that SVB was an outlier, the bank would have acted.

It is not our assertion that this eliminates the board’s independent responsibility to understand the risks inherent in SVB’s business model and balance sheet — directors cannot rely upon supervisors to perform the board’s risk-governance function. But the claim that the board failed to appreciate SVB’s vulnerability must be considered alongside the information supervisors possessed, the information they communicated, and the urgency with which they communicated it.

The apparent disconnect extended beyond interest-rate risk. Miller did not recall being told that Federal Reserve supervisors had raised concerns about “foundational weaknesses” with SVB’s auditor, KPMG, during meetings in early 2022. These meetings were not discussed in any of the principal postmortems but were revealed by a September 2025 Senate Permanent Subcommittee on Investigations Minority Staff Report into KPMG’s audit work for all three banks that failed in Spring 2023.11

Miller said neither KPMG nor supervisors brought those conversations to her attention. SVB understood that it needed to satisfy enhanced prudential expectations, but Miller recalled no urgent warning that the bank faced imminent danger. The bank remained comparatively well-rated when it entered receivership. In her words, it was a “dog that didn’t bark” situation.

Miller experienced SVB’s transition into large-bank supervision as awkward and unclear. When SVB became a Large Financial Institution (“LFI”) in 2021, after remaining above $100 billion in assets for four consecutive quarters, she sensed a “new sheriff in town.” Washington would play a larger role, but she did not see clarity as to what that meant in practice.

The San Francisco Fed and California regulators remained in the foreground, while unfamiliar personnel from Washington appeared in the background. It was not clear whether there was a new supervisory playbook, which officials were responsible for applying it, or with whom SVB was expected to develop relationships. Meetings continued to be arranged through San Francisco, with personnel from Washington joining episodically. 

Miller’s account also provides context for the assertion that the board allowed SVB’s growth to outrun its governance and risk-management capabilities. She reported that the bank anticipated the challenges associated with becoming a large institution. SVB engaged Sullivan & Cromwell to advise on its LFI obligations, commissioned work from McKinsey, and sought assistance from experienced advisers, including former Federal Reserve Governor Roger Ferguson and former Citi CFO John Gerspach. The board believed it was assembling the expertise required to govern a larger and more complex bank.

Management also sought to moderate balance-sheet growth, according to Miller. SVB reportedly directed approximately $150 billion of customer funds into off-balance-sheet money-market products because it did not want deposits to increase still more rapidly and trigger additional regulatory thresholds. It restrained lending because management did not believe that the influx of funds could be deployed prudently into new loans, investing instead in securities that it regarded as high quality.

With hindsight, those decisions contributed to the balance-sheet structure that made SVB vulnerable to rising rates and deposit flight. The fact that management intended to act conservatively does not establish that it managed the resulting risks effectively. It does, however, complicate the characterization of SVB as pursuing growth without regard to prudence. Miller describes management and the board as recognizing the problem of rapid growth, taking steps intended to control it, and failing to understand how those steps generated a different — and ultimately severe — combination of risks.

Miller does not believe that SVB’s failure was inevitable. She regarded the bank as solvent, while acknowledging that solvency is ultimately a question for forensic analysis. In her account, structural and operational rigidities became decisive once the run began: Federal Home Loan Bank access was curtailed, collateral could not be moved quickly enough to the Discount Window, relevant personnel at the Federal Reserve responsible for the Discount Window could not be reached after East Coast business hours, and the weekend intervened.

She emphasized the unprecedented speed of the run. Depositors in an earlier era had to stand in line; SVB experienced outflows reportedly approaching $1 million per second. Miller argues that the system needed a circuit breaker — a mechanism that would allow SVB and its supervisors to pause, assess the situation, mobilize collateral, and “get to the weekend.”

This account does not answer whether the board and management should have left SVB so vulnerable to a run. It does suggest that the ultimate failure involved more than the quality of the bank’s internal governance. In Miller’s telling, the outcome reflected an interaction among balance-sheet vulnerability, an extraordinarily fast run, supervisory and market signals, and an institutional response mechanism unable to operate at the speed of the crisis.

Miller’s account does not establish that the conclusions of the Barr, GAO, and OIG reports are mistaken. Nor does it demonstrate that SVB’s board and management acted with sufficient speed or effectiveness. It does, however, establish that the principal postmortem studies reached forceful conclusions about the board without seeking its evidence, testing its explanations, or reviewing the full record of its claimed remediation efforts.

During her interview, Miller identified specific materials capable of being tested: board and committee records, MRA and MRIA response plans, evidence relating to closed findings, consulting engagements, risk-management hiring, communications with supervisors, compensation materials, the resolution-plan submission, KPMG communications, and the Federal Reserve’s internal discussion of interest-rate and liquidity risk.

It also made clear that the perspective of a single director — recounted through contemporaneous interview notes rather than a reviewed transcript — could not provide an adequate evidentiary counterweight by itself.

For that reason, during the March 17th meeting, Starling asked Miller’s counsel to explore whether Miller and the other former Independent Directors of SVB would be willing to provide Starling with a fuller written account. The purpose of that request was to place into the record perspectives that had never been sought by past postmortem authors, so that competing accounts could be tested against one another and against the contemporaneous documentary record.

SVB Independent Directors’ Statement

On April 20th 2026, the former Independent Directors of SVB responded collectively to Starling’s request for a fuller written account of the events Miller had described.12 The resulting “Directors’ Statement” (the “Statement”) offers a retrospective account prepared by individuals whose own conduct has been criticized and must be weighed accordingly. But it provides something the Barr Report — and, by extension, the GAO and OIG reviews that drew heavily upon the same supervisory record — did not: the perspective of the Directors themselves, accompanied by specific dates, figures, actions, and references to contemporaneous communications against which their account can be tested.

The Directors opened with a direct challenge to the completeness of the established public narrative. They characterized the Barr Report as a “one-sided and selective recounting of the supervisory history of Silicon Valley Bank,” prepared “without a single conversation with or request for information from any of the Independent Directors of SVB.” 

They objected in particular that the Federal Reserve released supervisory correspondence describing its criticisms of SVB, but made no mention of the submissions SVB made in response. Nor did it share any of the records reflecting the SVB board’s own governance and risk-management activities. The result, they argued, was a public record that documented supervisory demands without documenting what the bank did in response to them.

This identifies a significant evidentiary limitation in findings — repeated across the Barr, GAO and OIG reports — that SVB management failed to remediate known deficiencies, and that the board treated supervisory findings as little more than a compliance exercise. Those conclusions cannot be assessed fully without examining both the supervisors’ criticisms and the actions taken in response.

The Directors’ Statement first addressed the Barr Report’s characterization of the SVB board as lacking sufficient banking and large-financial-institution experience. The Directors called that “sweeping characterization” incorrect, pointing to members with experience at Bank of America, Capital One, and JPMorgan, as well as service on the board of the Royal Bank of Canada, and to collective experience spanning banking, auditing, accounting, asset management, investment banking, financial technology, and government. 

More important than the Directors’ résumés, however, was their account of what they did as SVB approached the threshold for supervision as a large financial institution.

According to the Statement, the board and management began preparing for that transition in 2020, before SVB crossed the $100 billion threshold. LFI-readiness work appeared regularly in board and committee materials throughout 2021 and 2022, and the Directors received training concerning heightened supervisory expectations and large-bank governance. 

SVB retained McKinsey and Sullivan & Cromwell and later engaged former Federal Reserve Vice Chair Roger Ferguson and former Citi Chief Financial Officer John Gerspach as special advisers. Formal meetings of the board and its principal committees increased from 69 in 2020 to 86 in 2022.

Staffing increased as well. SVB grew from 3,564 employees at the end of 2019 to 8,553 at the end of 2022, including approximately 2,000 new employees added during 2022. The Statement asserts that many of those additions were made in first-, second-, and third-line risk management, internal audit and regulatory affairs, with the board’s support. These figures do not demonstrate that the resulting functions were adequately designed or sufficiently mature, of course. But they do complicate a narrative of indifference to supervisory concerns or an unwillingness to invest in risk management.

The Directors also provided context for SVB’s rapid balance-sheet growth. They maintained that the surge in deposits during 2020 and 2021 was not a strategic objective but the product of extraordinary inflows from the venture-capital and technology sectors during a period of low interest rates. They pointed to management’s placement of substantial client funds in third-party money-market funds as evidence of an effort to moderate that growth. 

By year-end 2022, they reported, approximately $168 billion in client funds were held off-balance sheet in such funds — nearly half of total client funds. The board viewed that decision as demonstrating “a willingness to slow growth during a period of rapid inflows to SVB.” That account illustrates Miller’s contention that the board understood rapid growth to present a risk and supported measures intended to constrain it.

The Statement similarly disputed the significance commonly attributed to the vacancy in the Chief Risk Officer position during much of 2022. The Directors maintained that the incumbent CRO was replaced precisely because the board and management believed SVB needed someone with experience at a large financial institution. 

While a search proceeded, SVB created an Office of the Chief Risk Officer composed of seven senior risk managers whom the Directors described as possessing significant LFI experience. FRBSF supervisors, the Directors said, knew of and supported that interim structure. By August 2022, SVB had selected Kim Olson, a risk executive with 29 years of experience, including ten years at the Federal Reserve Bank of New York. Her contractual “garden leave” delayed her formal start until late December, although she attended a board offsite in November.

The Directors’ description of SVB’s transition into the large-bank supervisory portfolio also closely tracks Miller’s account of a regulatory “cliff effect.” The Statement describes a “sea change in expectations” as SVB moved from the Federal Reserve’s regional-banking program into its large-financial-institution framework. 

In July 2021, supervisors had credited recent board appointments, increased director engagement, support for additional talent and “meaningful actions” to address regulatory concerns. Barely a year later, however, SVB received its first large-bank supervisory assessment and the bank’s composite CAMELS rating was downgraded from 2 (Satisfactory) to 3 (Less than Satisfactory).

Yet even then, the component ratings for Capital, Asset quality, Earnings, Liquidity, and Sensitivity to market risk each remained 2 (Satisfactory); the downgrade in the composite rating was driven by a downgrade in the Management component rating from 2 (Satisfactory) to 3 (Less than Satisfactory). Those ratings did not establish that SVB was free from serious financial risk. But they help explain what Miller described as the “dog that didn’t bark”: neither the formal ratings nor the supervisors’ communications to the board conveyed that SVB’s liquidity or interest-rate position placed the institution in imminent danger. Otherwise, the Federal Reserve might have been expected to downgrade the Liquidity and Sensitivity to market risk component ratings — which include interest-rate risk — from 2 (Satisfactory) to 3 (Less than Satisfactory), or perhaps even 4 (Deficient) or 5 (Critically Deficient).

The Directors’ Statement also provides context for the large number of MRAs and MRIAs later cited as evidence of management and board inattention to risk. 

The Directors acknowledged the rapid accumulation of supervisory findings but emphasized that most concerned technology, operational resilience, and process, rather than deficiencies relating to the level of SVB’s liquidity or interest-rate risk. Of the twelve MRIAs issued between 2019 and 2022, the Statement identifies three as relating to SVB’s process for managing liquidity risk, but not its level of liquidity risk. That is, the MRIAs were focused on remediating perceived weaknesses in processes for managing financial risk at SVB rather than remediating the financial risk itself. Other findings concerned vulnerability remediation, technology-risk management, IT asset management, vendor management, identity and access management, data governance and information security.

The board’s response to that emphasis was substantial. It created an IT Oversight Group in February 2021, which held 22 meetings over the following fourteen months, including three attended by federal and state supervisors. In 2022, the board established a separate Technology Committee because, as Miller recalled, the volume of technology-related work was consuming an excessive portion of the Audit Committee’s time. 

With respect to the six liquidity risk-management-related matters (two MRIAs and four MRAs) issued in November 2021, the Directors asserted that SVB submitted remediation plans for all six by their June 2022 deadlines, after review by the bank’s internal risk and audit functions. According to the Statement, the Federal Reserve did not respond to those plans. 

SVB’s Regulatory Affairs team informed the board that supervisors did not expect to do so until completing a horizontal liquidity review for a number of large banking organizations similar to SVB, planned for the first quarter of 2023.

The Directors described their attempts to obtain greater specificity from supervisors as a “cat and mouse” dynamic. At a June 2022 meeting, they recalled being told that the board would have to determine for itself what supervisors expected. 

SVB consequently retained highly reputable outside advisers, such as McKinsey and EY, and attempted to benchmark its practices against those of other large financial institutions. This experience is consistent with Miller’s recollection of asking, “Is this enough? Is this enough?” while receiving neither clear approval nor a definitive rejection.13

The Statement next addressed the three MRIAs concerning governance, risk management and internal audit delivered to the board in May 2022 — the findings most directly implicated by the OIG’s criticisms of SVB’s governance and three lines of defense. The Directors assigned responsibility for each response to specific board members: board Chair Kay Matthews led the governance work, Kate Mitchell led the risk-management response, and Miller led the internal-audit work. 

Two months later, the board gave supervisors a progress report. The lead FRBSF supervisor subsequently wrote to Matthews and CEO Greg Becker that the presentations by the committee chairs had been “quite constructive,” adding: “We have a better sense of how seriously SVB’s board understands the forward path. I appreciate the respectful approach that SVB leadership has taken on all matters related to our supervisory assessment.” 

That contemporaneous message is difficult to reconcile with an unqualified portrayal of a disengaged board treating supervisory concerns merely as formalities. The Directors understood that communication to mean that “regulators made it clear that we were on the right track.”

The Directors offered particularly detailed accounts of the risk-management and internal-audit work.

In response to the risk-management MRIA, SVB conducted a gap assessment identifying 30 deficiencies against regulatory and industry standards. Its Risk Transformation Program divided the resulting work into 13 workstreams and 174 individual action plans, each with designated owners, deliverables and target dates. The Risk Committee shifted from quarterly to monthly meetings, with sessions expanding from approximately 90 minutes to more than three hours, and a Program Oversight Office was charged with providing monthly progress reports.

The Risk Committee and full board reviewed and approved the plan before its submission to supervisors on August 31st 2022. The Statement asserts that neither the FRBSF nor the California Department of Financial Protection and Innovation (“DFPI”) responded.

“They recalled being told that the board would have to determine for itself what supervisors expected.”

The internal-audit response followed a similar pattern. A gap assessment identified 19 areas for improvement, producing 106 planned actions and 302 measurable work steps. Miller began meeting with the Internal Audit team every two weeks to monitor progress. The plan and its status were reviewed with supervisors in August and October 2022, with individual actions classified as “at risk,” “on track” or “completed.” 

According to the Directors, Federal Reserve supervisors neither provided a formal response nor expressed concern about the plan’s progress.

Later in October, Federal Reserve supervisors began a targeted examination covering many of the same subjects. The resulting December supervisory letter modified some existing expectations and introduced others, causing SVB to add another 60 actions to its plan. 

To the Directors, “it felt like supervisors were moving the goalposts.” Whether that was the product of shifting supervisory expectations, the discovery of deeper deficiencies, or the inadequacy of SVB’s initial response cannot be resolved from the Directors’ Statement alone. But the sequence provides material context missing from an account that treats an open supervisory finding as proof that the board had failed to act upon it.

The Directors also directly challenged the Barr Report’s statement that senior-management compensation through 2022 was tied to short-term earnings and equity returns without risk metrics. They asserted that risk performance had been incorporated qualitatively into compensation decisions before 2022 and that quantitative risk measures were introduced that year, resulting in significant reductions in incentive compensation for the CEO and other senior executives. They further reported that 71 percent of the CEO’s compensation was weighted toward long-term performance, compared with a peer benchmark of 63 percent. 

Taken as a whole, the Directors’ Statement does not disprove the central finding that SVB failed to manage its interest-rate and liquidity risks. The bank failed, and the ongoing remediation efforts at the time of its closure clearly failed to prevent that outcome. 

The Statement instead contests the broader characterization that the directors were inexperienced, inattentive, unwilling to invest in risk management, or content to treat serious supervisory findings as a box-checking exercise. It supplies evidence of extensive activity while leaving a more difficult question unresolved: whether that activity was appropriately directed, sufficiently urgent and capable of producing effective change before SVB’s vulnerabilities became fatal.

That question cannot be answered by counting meetings, employees, consultants, workstreams or action items. It requires an account of what the remediation work consisted of in practice: how deficiencies were identified, how priorities were set, what obstacles the teams encountered, what supervisors communicated, and how far the work had progressed by March 2023. 

The Directors’ Statement therefore points beyond the board’s account to the individuals charged with carrying out the work itself. One such individual was recruited into SVB’s risk-governance remediation effort and participated directly in the events the directors describe. It is to that operational account that we turn next.

References
  1. Ibid., 1.
  2. Board of Governors of the Federal Reserve System, Key Takeaways, in Review of the Federal Reserve’s Supervision and Regulation of Silicon Valley Bank, i.
  3. U.S. Government Accountability Office, Bank Regulation: Preliminary Review of Agency Actions Related to March 2023 Bank Failures, 3.
  4. Ibid., 14.
  5. Ibid., 21.
  6. Board of Governors of the Federal Reserve System, Office of Inspector General, Material Loss Review of Silicon Valley Bank, 11.
  7. Ibid., 13.
  8. Ibid., 14.
  9. Appendix A-8, Mary Miller, 17 March 2026.
  10. Silicon Valley Bank, “SVB Financial Group Appoints Richard Daniels to Its Board of Directors,” press release, PR Newswire, October 22 2020, https://www.prnewswire.com/news-releases/svb-financial-group-appoints-richard-daniels-to-its-board-of-directors-301158367.html.
  11. U.S. Senate Permanent Subcommittee on Investigations, “This Industry Is a Joke”: How KPMG’s Unqualified Audits of Silicon Valley Bank, Signature Bank, and First Republic Bank Allowed Troubled Banks to Hide Their Failures in Plain Sight, minority staff report prepared by Senator Richard Blumenthal (U.S. Senate Committee on Homeland Security and Governmental Affairs, September 17, 2025), https://www.hsgac.senate.gov/wp-content/uploads/2025_09_17-Regional-Bank-Failures-Report-1.pdf.
  12. Addendum Item 4 — SVB Directors’ Statement.
  13. It is notable that, throughout its efforts to address the heightened supervisory scrutiny applied to SVB as it grew, the bank’s board and management consistently sought the support of highly credentialed external advisers to include: H. Rodgin Cohen, Senior Chairman of Sullivan & Cromwell, former Federal Reserve Vice Chair Roger Ferguson, and former Citi CFO John Gerspach.

Join The Discussion

Sign in and be the first to comment.

See something that doesn't look quite right?

We strive to provide high quality and accurate content at all times. With that said, we realize that sometimes links break, new information becomes available, or there is something that you feel we may have missed.

If you see something that you think we should be aware of, we would love to hear from you. Feel free to drop us a note below and leave your name and contact info if you'd like to hear back from us.

Thank you for being a key part of the Starling Insights community!