Silicon Valley Bank Demise: Causes and the Path Forward

The recent demise of Silicon Valley Bank, First Republic Bank, and Signature Bank are three of the four largest bank failures in U.S. history. These bank failures have sent shock waves not only through the bank regulators and depositors, but also through the venture capital industry, commercial real estate lenders, and crypto investors. The media and researchers have focused on the role of duration mismatch between the assets and liabilities of these banks. In this paper we focus on the problems with bank fundamentals, namely, bank governance and bank capital structure, that allowed, or even encouraged, bank managers to ignore the growing mismatch between the assets and liabilities of their banks.

The problems with the fundamentals in these three banks that led to their demise are eerily similar to the fundamental problems in the big banks circa 2008 that led to the global financial crisis. The proximate cause for the big bank crisis in 2008 was the problems in valuing complex mortgage-backed securities. Once again, it was big bank governance and big bank capital structure, that enabled, or even encouraged, big bank managers to ignore the growing problems in valuing complex mortgage-backed securities that they were holding in their portfolios (Bhagat (2017)).

In the wake of the global financial crisis, attention has often focused on whether incentives generated by bank executives’ compensation programs led to excessive risk-taking. Governments worldwide have, in particular, regulated bank executives’ compensation by requiring deferral of incentive compensation, mandating clawbacks, and in some instances, even restricting compensation amounts. In earlier work (Bhagat and Romano (2010)) we recommended the following compensation structure for bank executives, with which these government initiatives are only partially consistent: incentive compensation should consist only of restricted stock and restricted stock options – restricted in the sense that the executive cannot sell the shares or exercise the options for six to twelve months after his or her last day in office; we refer to this as the Restricted Equity proposal. We contend that such an incentive compensation package will focus bank management’s attention on the long-run and discourage investment in high-risk, value-destroying projects such as using the bank’s short term deposit liabilities to fund longer-term treasury and mortgage debt assets.

Equity-based incentive programs such as our proposal may lose effectiveness in motivating managers to reduce excessive risk-taking as a bank’s equity value approaches zero. There is a moral hazard or agency cost of debt in this context arising from shareholders’ potential preference to take extreme risks when close to insolvency. This is because shareholders gain from the upside of a positive outcome, albeit low in probability, while limited liability leaves the losses, should the gamble fail, on creditors. The moral hazard problem when equity value approaches zero may well be more severe for banks, as their creditors have less interest in monitoring against risk-taking activity because the government not only stands behind retail depositors, but also often bails out other creditors as well. Properly aligning management’s incentives in this context therefore calls for focus on the interaction among bank capital structure, bank capital requirements and bank executive incentive compensation – whereas, the extant literature typically analyzes compensation reform, or bank capital reform in isolation.

In this paper we consider the executive compensation structure and the bank equity capital to understand the recent demise of Silicon Valley Bank (SVB), First Republic Bank, and Signature Bank. We find that these three bank CEOs were able to realize a significantly larger amount by selling their vested equity during January 2017 – February 2023, compared to the large losses the executives experienced on their equity stake during March and April 2023. Additionally, average stock sales by these three bank CEOs was significantly greater than average stock sales of 37 non-distressed large U.S. banks in both an absolute and relative sense (controlling for beginning of period holdings).

Based on these results and our analysis of the compensation structures at 100 of the largest U.S. financial institutions during the great financial crisis of 2008, we recommend the following compensation structure for senior bank executives: Executive incentive compensation should only consist of restricted stock and restricted stock options – restricted in the sense that the executive cannot sell the shares or exercise the options for one to three years after their last day in office. This will more appropriately align the long-term incentives of the senior executives with the interests of the stockholders.

The Restricted Equity incentive compensation proposal for bank managers logically leads to a complementary proposal regarding a bank’s capital structure: Low levels of equity capital (to total assets) will magnify the impact of losses on equity value. As banks’ equity values approach zero (as they did for some banks in 2008, and, again, in 2023), equity based incentive programs lose their effectiveness in motivating managers to enhance shareholder value. Additionally, our evidence suggests that bank CEOs sell significantly greater amounts of their stock as the bank’s equity-to-capital ratio decreases. Hence, for equity based incentive structures to be effective, banks should be financed with considerably more equity than they are being financed currently. Our recommendation for significantly greater equity in a bank’s capital structure is consistent with the recommendations of Admati and Hellwig (2013) and Bhagat (2017).

Specifically, our bank capital proposal has two components: Bank capital should be calibrated using both balance sheet (tangible equity capital) and market value of equity. Second, bank capital should be at least 20% of total assets (not risk-based assets per current regulatory requirement).

We find that the three distressed banks’ balance sheet equity variable, tangible equity (divided by total assets), is less than for the 37 non-distressed banks. Additionally, the three distressed banks’ market equity (divided by total assets), is significantly below average compared to the 37 non-distressed banks. These results suggest that banks that have more equity capital tend not to fall into financial distress.

The proximate cause of the demise of SVB, Signature Bank, and First Republic Bank was a run on the uninsured deposits in these banks. FDIC insures deposits of up to $250,000 for all depositors. However, many depositors in these banks had deposits significantly more than $250,000. In the case of these three banks, FDIC decided to use its Deposit Insurance Fund to guarantee all of the uninsured deposits; this has cost the other insured U.S. banks about $35 billion – a cost that will be passed on to the retail customers of all these other banks in the U.S.

To prevent future runs on the uninsured deposits in a bank, some have suggested unlimited deposit insurance. However, unlimited deposit insurance has costs: the government would have unlimited exposure to a bank’s deposit liabilities. Hence, the government would play a much more significant role in regulating banks’ lending and investment activities. Also, banks would have to pay a much higher deposit insurance premium; a cost that will be passed on to retail customers as higher fees.

Greater equity financing of banks coupled with the above compensation structure for bank managers will drastically diminish the likelihood of a bank falling into financial distress, making bank runs extremely unlikely – thus eliminating the need for unlimited deposit insurance.