01 June 2023
SVB and Lehman Brothers: Parallel Lives?
By Stilpon Nestor and Jacob Nolan[1]
On Wednesday the 8th of March 2023, SVB Financial Group (“SVB”) announced a $1.75 billion capital raise through a stock sale in order to fill a $1.8 billion-sized hole in its balance sheet caused by the sale of a $21 billion loss-making bond portfolio. On Thursday the 9th of March, depositors, who were primarily in the tech and VC space, withdrew around $42 billion in deposits (25% of total deposits). The next day, facing attempted cash withdrawals of nearly $100 billion (for a combined 81% of SVB’s deposits), SVB collapsed and was taken into FDIC receivership, at a cost so far of $20 billion.
Although the speed of the collapse and scale of withdrawals was unprecedented (for comparison, Washington Mutual collapsed in 2008 after a nine-day bank run that saw 9% of deposits withdrawn), the governance warning signs were there. In this short note we point to a number of similarities in the underlying governance issues between SVB and the most iconic of bank collapses, that of Lehman Brothers in 2009.
As the recent FED Review of SVB’s collapse[2] clearly demonstrates, the proximate cause of SVB’s failure was its inability to manage its interest rate risk. This is a common occurrence in times of significant interest rate movement, such as the one we are currently going through. It has also been behind the other significant stress of the last few days, that of First Republic Bank, rescued by JPMorgan Chase. In the SVB case, interest rate risk was accumulated in the securities book through unhedged positions in treasury bonds. In the case of First Republic Bank, the interest rate risk seems to have been mostly in the banking (mortgage) book.
The other side of the SVB balance sheet was also fraught with risk. Deposits were concentrated among relatively few depositors with large deposits, greatly exceeding the FDIC coverage ceiling of USD $250.000 – most of them in the tech and venture capital communities of Silicon Valley. This “lumpiness” on the liability side combined with a highly-networked depositor universe raised liquidity risk significantly for the bank.
The FED Report goes beyond the proximate causes to explain how these risks were exacerbated by significant regulatory gaps combined with less than rigorous supervision. Following a 2018 softening of the regulatory regime for banks with less than $250 billion in assets, SVB, the 16th largest bank in the US, no longer fell under the intensive regulatory and supervisory oversight mandated by the Dodd-Frank Act. Under its classification as a Category IV bank, it was subject to a less stringent set of prudential standards, less frequent stress testing, no liquidity stress testing, and less rigorous capital planning and liquidity risk management standards. It was also not subject to a liquidity coverage ratio, net stable funding ratio, or supplementary leverage ratio.
Supervisors did find deficiencies in SVB’s liquidity risk management at the end of 2021. They also commented on its ineffective board oversight, risk management, and internal audit function in May 2022, and rated its enterprise-wide governance and controls as “deficient-1” in the summer of 2022. The recent FED Review noted that they were in the process of drafting a Memorandum of Understanding as an informal enforcement action against SVB relating to its governance and controls at the time it collapsed.[3] Supervisors had also met with senior management in October and November 2022 to express concern and deliver supervisory findings on the interest rate management at SVB.
It is important to remember that good banks should not simply rely on supervisors to manage risk and correct governance deficiencies for them. They should not turn away from best practice risk management and governance simply because regulators have let them “off the hook”. Yet, it seems that this is what happened here. The supervisory findings indicate significant shortcomings in risk appetite setting and the framework for its implementation; in the way senior management and the Board followed the risk profile of the bank; and, most importantly, in the way risk was managed. Following initial supervisory findings and rebukes, SVB’s Chief Risk Officer (“CRO”) stepped down from her role in April 2022. But the Bank never managed to replace her. In nine of the twelve months preceding its collapse, SVB had no CRO. Risk management was carried out “by committee”. Presumably, SVB’s Board did little to remedy this highly abnormal situation. This is not OK in normal times but it is downright dangerous in a context of increasing risk fueled by a downturn in the tech sector (SVB’s main clientele) and the Bank’s rapid pace of growth which resulted in the size of its balance sheet increasing by 271% between year-end 2018 and year-end 2021 (compared to 29% for the banking industry as a whole).
Similar risk management deficiencies were found to underpin the collapse of Lehman Brothers back in 2009.[4] In that case, the CRO of the bank was fired in September 2007 (almost in parallel with the resignation of the head of the leveraged credit business) for having pushed back on the ever expanding risk appetite of the Bank, to be replaced a CEO loyalist from the business. The Board and its Finance and Risk Committee failed to ask why.
A parallel can also be drawn between SVB’s Board’s composition and the Board of Lehman Brothers prior to its collapse. In 2009, 90% of Lehman’s Board had no experience working in a bank. At SVB, the situation was not very different. Three members of SVB’s 12-member Board did have banking experience (although one of these Board members’ most recent experience working at a bank was over 25 years ago), which is broadly in-line with the rate seen at a peer group of high-performing North American G-SIBs that we used as back-of-the envelop benchmark (Morgan Stanley, Royal Bank of Canada, Toronto-Dominion Bank, Goldman Sachs, State Street Corporation, and JP Morgan & Chase). They averaged 23%.
However, these directors’ experience was in investment banking and SVB was primarily a corporate/commercial bank with little, if any, investment banking business. In fact our research seems to suggest that none of the twelve board members had commercial banking experience, the type of banking that constituted the vast majority of SVB’s actual business; for comparison, the directors on Goldman Sachs’ Board who have banking experience have all previously worked in investment banking, given that investment banking is Goldman Sachs’ primary business.
Moreover, our research suggests that SVB’s risk committee membership profile was less than ideal. On the one hand, there was little to no commercial (or even investment) banking risk experience. On the other, a significant number of members came from the client side (venture capital, tech, vineyards) — hardly a group to recommend restraining risk appetite.
In fact, SVB’s Board composition was far more heavily weighted towards experience in venture capital. 45% of SVB’s Board had either founded or worked at a venture capital/private equity fund. While this might be good experience to have, it also increases client power on the board with the ensuing conflict potential and “grapevine” risk that seems to have very much materialized in SVB’s case.
SVB’s Board was also rather entrenched. 50% (including the CEO) had sat on the Board for more than a decade, which is significantly closer to the 60% seen at Lehman than the 25% seen at the peers. SVB had two of the three longest serving Board members among the peers, with one director having been on the Board for 19 years and another for 18 years. The average length of time a SVB Board member had sat on the Board was 8 years and 7 months, compared to 6 years and 11 months at the peer banks. One should also note that tenure at US banks is significantly higher than in systemic banks in the rest of the world, with the average tenure at large European banks being around 5 years.[5]
In another eerie reminder of Lehman, the FED Report notes that SVB’s variable remuneration was not in any way linked to risk performance and was only tied to short-term earnings and equity returns. The FED noted that this had the effect of creating a “financial incentive [for SVB’s management] to focus on short-term profit over sound risk management”, to disastrous effect. Risk-based variable compensation is now standard for most supervised banks, including our peers above. Compensation for all bank “risk takers” is, as a matter of course, directly linked to the risks assumed by management.
In conclusion, the proximate causes of SVB’s collapse were deficient risk management at the perilous juncture of a changing interest rate environment and rapid growth. However, these deficiencies were not random events. They grew on the fertile soil of deficient governance, of a board that might have had too few competencies to understand and oversee the management of core banking risk, that was too close to the clients, and was too stale. Same old, same old…
1 Respectively Senior Advisor and Analyst, Morrow Sodali
2 Board of Governors of the Federal Reserve System, Review of the Federal Reserve’s Supervision and Regulation of Silicon Valley Bank, April 2023
3 FED Review, pages 42-43
4 Anton R. Valukas, Lehman Brothers Holdings Inc. Chapter 11 Proceedings Examiner Report, March 2010
5 Nestor Advisors, Work in Progress: A Report on the Corporate Governance of Europe’s Top 25 Banks, 2015
On Wednesday the 8th of March 2023, SVB Financial Group (“SVB”) announced a $1.75 billion capital raise through a stock sale in order to fill a $1.8 billion-sized hole in its balance sheet caused by the sale of a $21 billion loss-making bond portfolio. On Thursday the 9th of March, depositors, who were primarily in the tech and VC space, withdrew around $42 billion in deposits (25% of total deposits). The next day, facing attempted cash withdrawals of nearly $100 billion (for a combined 81% of SVB’s deposits), SVB collapsed and was taken into FDIC receivership, at a cost so far of $20 billion.
Although the speed of the collapse and scale of withdrawals was unprecedented (for comparison, Washington Mutual collapsed in 2008 after a nine-day bank run that saw 9% of deposits withdrawn), the governance warning signs were there. In this short note we point to a number of similarities in the underlying governance issues between SVB and the most iconic of bank collapses, that of Lehman Brothers in 2009.
As the recent FED Review of SVB’s collapse[2] clearly demonstrates, the proximate cause of SVB’s failure was its inability to manage its interest rate risk. This is a common occurrence in times of significant interest rate movement, such as the one we are currently going through. It has also been behind the other significant stress of the last few days, that of First Republic Bank, rescued by JPMorgan Chase. In the SVB case, interest rate risk was accumulated in the securities book through unhedged positions in treasury bonds. In the case of First Republic Bank, the interest rate risk seems to have been mostly in the banking (mortgage) book.
The other side of the SVB balance sheet was also fraught with risk. Deposits were concentrated among relatively few depositors with large deposits, greatly exceeding the FDIC coverage ceiling of USD $250.000 – most of them in the tech and venture capital communities of Silicon Valley. This “lumpiness” on the liability side combined with a highly-networked depositor universe raised liquidity risk significantly for the bank.
The FED Report goes beyond the proximate causes to explain how these risks were exacerbated by significant regulatory gaps combined with less than rigorous supervision. Following a 2018 softening of the regulatory regime for banks with less than $250 billion in assets, SVB, the 16th largest bank in the US, no longer fell under the intensive regulatory and supervisory oversight mandated by the Dodd-Frank Act. Under its classification as a Category IV bank, it was subject to a less stringent set of prudential standards, less frequent stress testing, no liquidity stress testing, and less rigorous capital planning and liquidity risk management standards. It was also not subject to a liquidity coverage ratio, net stable funding ratio, or supplementary leverage ratio.
Supervisors did find deficiencies in SVB’s liquidity risk management at the end of 2021. They also commented on its ineffective board oversight, risk management, and internal audit function in May 2022, and rated its enterprise-wide governance and controls as “deficient-1” in the summer of 2022. The recent FED Review noted that they were in the process of drafting a Memorandum of Understanding as an informal enforcement action against SVB relating to its governance and controls at the time it collapsed.[3] Supervisors had also met with senior management in October and November 2022 to express concern and deliver supervisory findings on the interest rate management at SVB.
It is important to remember that good banks should not simply rely on supervisors to manage risk and correct governance deficiencies for them. They should not turn away from best practice risk management and governance simply because regulators have let them “off the hook”. Yet, it seems that this is what happened here. The supervisory findings indicate significant shortcomings in risk appetite setting and the framework for its implementation; in the way senior management and the Board followed the risk profile of the bank; and, most importantly, in the way risk was managed. Following initial supervisory findings and rebukes, SVB’s Chief Risk Officer (“CRO”) stepped down from her role in April 2022. But the Bank never managed to replace her. In nine of the twelve months preceding its collapse, SVB had no CRO. Risk management was carried out “by committee”. Presumably, SVB’s Board did little to remedy this highly abnormal situation. This is not OK in normal times but it is downright dangerous in a context of increasing risk fueled by a downturn in the tech sector (SVB’s main clientele) and the Bank’s rapid pace of growth which resulted in the size of its balance sheet increasing by 271% between year-end 2018 and year-end 2021 (compared to 29% for the banking industry as a whole).
Similar risk management deficiencies were found to underpin the collapse of Lehman Brothers back in 2009.[4] In that case, the CRO of the bank was fired in September 2007 (almost in parallel with the resignation of the head of the leveraged credit business) for having pushed back on the ever expanding risk appetite of the Bank, to be replaced a CEO loyalist from the business. The Board and its Finance and Risk Committee failed to ask why.
A parallel can also be drawn between SVB’s Board’s composition and the Board of Lehman Brothers prior to its collapse. In 2009, 90% of Lehman’s Board had no experience working in a bank. At SVB, the situation was not very different. Three members of SVB’s 12-member Board did have banking experience (although one of these Board members’ most recent experience working at a bank was over 25 years ago), which is broadly in-line with the rate seen at a peer group of high-performing North American G-SIBs that we used as back-of-the envelop benchmark (Morgan Stanley, Royal Bank of Canada, Toronto-Dominion Bank, Goldman Sachs, State Street Corporation, and JP Morgan & Chase). They averaged 23%.
However, these directors’ experience was in investment banking and SVB was primarily a corporate/commercial bank with little, if any, investment banking business. In fact our research seems to suggest that none of the twelve board members had commercial banking experience, the type of banking that constituted the vast majority of SVB’s actual business; for comparison, the directors on Goldman Sachs’ Board who have banking experience have all previously worked in investment banking, given that investment banking is Goldman Sachs’ primary business.
Moreover, our research suggests that SVB’s risk committee membership profile was less than ideal. On the one hand, there was little to no commercial (or even investment) banking risk experience. On the other, a significant number of members came from the client side (venture capital, tech, vineyards) — hardly a group to recommend restraining risk appetite.
In fact, SVB’s Board composition was far more heavily weighted towards experience in venture capital. 45% of SVB’s Board had either founded or worked at a venture capital/private equity fund. While this might be good experience to have, it also increases client power on the board with the ensuing conflict potential and “grapevine” risk that seems to have very much materialized in SVB’s case.
SVB’s Board was also rather entrenched. 50% (including the CEO) had sat on the Board for more than a decade, which is significantly closer to the 60% seen at Lehman than the 25% seen at the peers. SVB had two of the three longest serving Board members among the peers, with one director having been on the Board for 19 years and another for 18 years. The average length of time a SVB Board member had sat on the Board was 8 years and 7 months, compared to 6 years and 11 months at the peer banks. One should also note that tenure at US banks is significantly higher than in systemic banks in the rest of the world, with the average tenure at large European banks being around 5 years.[5]
In another eerie reminder of Lehman, the FED Report notes that SVB’s variable remuneration was not in any way linked to risk performance and was only tied to short-term earnings and equity returns. The FED noted that this had the effect of creating a “financial incentive [for SVB’s management] to focus on short-term profit over sound risk management”, to disastrous effect. Risk-based variable compensation is now standard for most supervised banks, including our peers above. Compensation for all bank “risk takers” is, as a matter of course, directly linked to the risks assumed by management.
In conclusion, the proximate causes of SVB’s collapse were deficient risk management at the perilous juncture of a changing interest rate environment and rapid growth. However, these deficiencies were not random events. They grew on the fertile soil of deficient governance, of a board that might have had too few competencies to understand and oversee the management of core banking risk, that was too close to the clients, and was too stale. Same old, same old…
1 Respectively Senior Advisor and Analyst, Morrow Sodali
2 Board of Governors of the Federal Reserve System, Review of the Federal Reserve’s Supervision and Regulation of Silicon Valley Bank, April 2023
3 FED Review, pages 42-43
4 Anton R. Valukas, Lehman Brothers Holdings Inc. Chapter 11 Proceedings Examiner Report, March 2010
5 Nestor Advisors, Work in Progress: A Report on the Corporate Governance of Europe’s Top 25 Banks, 2015