On 10 March 2023, the US financial authorities took control of Silicon Valley Bank (SVB), the 16th largest bank in the United States with approximately $200 billion in assets. On 12 March they also took control of a smaller entity, Signature Bank, with about $100 billion in assets.

What has happened with Silicon Valley Bank (SVB) can be summarised quite simply. SVB’s clients are primarily technology start-ups. These companies, together with venture capital investors, held huge amounts of deposits that the bank used to invest in fixed income assets, mainly long-term government bonds and mortgage-backed assets. An increase in demand for deposits from its customers forced SVB to sell part of these assets to obtain the liquidity to meet these obligations. Given the current situation of rising rates in the United States, the prices of the fixed-income assets on SVB’s balance sheet had fallen significantly. This is because there are now new assets on the market similar to those that SVB once bought, but much more profitable. The sale at a discount generated losses for the bank. To cover those losses, SVB sought to raise $2.25 billion of capital in the market. This sudden demand for capital precipitated a banking panic that wiped out the bank when depositors sought to withdraw $42 billion on 9 March alone, roughly a quarter of the bank’s deposits.

At the same time as they took control of these two institutions, the US central bank (Federal Reserve Board or Fed), its deposit insurance fund (Federal Deposit Insurance Corporation or FDIC) and its Securities and Exchange Commission (SEC) issued a communiqué in which they applied the Systemic Risk Exception to these two banks. Although they were not officially classified as systemic banks, the intervention of the US authorities would treat them as if they were. 

[Recibe los análisis de más actualidad en tu correo electrónico o en tu teléfono a través de nuestro canal de Telegram]

In the United States, the resolution of a failing bank is guided by the principle of minimising the cost of such intervention. This means that, of all the possible options (selling the institution in pieces, finding buyers to buy it whole, recapitalising, taking over ownership by the state, etc.), the one that entails the lowest cost for the resolution authority, i.e. for taxpayers, is chosen. However, when an institution is considered systemic, other factors come into play that may lead to choices that do not minimise the cost to the taxpayer. In particular, if there is a risk of contagion, the costs to society of a widespread bank failure would be enormous. Therefore, the competent authority’s action may be guided by the need to minimise the risk of such contagion. More than 90% of SVB’s deposits were uninsured because they exceeded the $250,000 limit guaranteed by the FDIC. The least costly option for the government would have been to cover the guaranteed deposits and let the uninsured deposits suffer losses. The fear of the US authorities was that this option could generate a panic of uninsured deposits in other banks considered similar. To minimise this risk, it was decided to invoke the systemic risk exception and guarantee all deposits whether they were insured or not.

To understand the implications of resorting to the systemic risk exception, we have to go back to 2015. As documented in some media reports, SVB’s Chief Executive Officer (CEO), Greg Becker, was lobbying in the US Senate to change the US prudential regulation included in the Dodd-Frank Act of 2010. This law stated that for a bank to be considered systemic, it had to be larger than $50 billion in assets. This is important because, according to the Act, a systemic bank is subject to more controls and has to comply with more stringent prudential requirements than a non-systemic bank. With assets equal to 45 billion euros in 2015, SVB was approaching the threshold that would cause it to be downgraded. According to official records seen by The Lever, Greg Becker spent huge amounts of money between 2015 and 2018 lobbying US authorities to raise the threshold for what is considered systemic. That change came in 2018 when Donald Trump’s government decided to increase that limit fivefold to 250 billion euros, curiously, the amount that Greg Becker had suggested in 2015 in his speech to the US Senate. 

And, as luck would have it, in January 2019, Greg Becker was appointed a director of the Federal Reserve Bank (Fed) of San Francisco, combining this position with that of executive chairman of SVB. As SVB is located in California, the San Francisco Fed was responsible for supervising SVB within the framework of the US Federal Reserve. As it happens, according to the Wall Street Journal and the New York Times, between 2019 and 2022, the San Francisco Fed issued several warnings to SVB about deficiencies in its risk controls. These warnings prompted the central bank to conduct a full review of its supervision of SVB.

On 6 March 2023, FDIC Chairman Martin Gruenberg explained in a speech at the IIB (Institute of International Bankers) that the steep and rapid rise in US interest rates had generated some $620 billion in unrealised losses for US banks from the fall in the price of the fixed-income assets these banks held on their balance sheets. They were unrealised because the banks did not have to bear them if they did not sell the assets. But that figure gave an indication of the potential risk on which the banks were sitting. Four days later, on 10 March, SVB went bankrupt precisely because it had to sell part of its portfolio of fixed-income assets at a considerable loss. 

On the same day that SVB went bankrupt, Greg Becker resigned as a director of the San Francisco Fed. It is now known that two weeks earlier, on 27 February, Becker had sold 12,500 SVB shares at a price of $287.42 per share. He had acquired these shares on the same day by exercising a call option at a price of $105.18. This implies a profit in one day. This implies a one-day profit of $2.3 million. On 10 March, SVB’s trading had to be suspended when its price fell to $34 after massive selling of the stock. 



In the aftermath, analysts are now wondering why all the San Francisco Fed’s supervisory actions between 2019 and 2022 alluded to above did not result in concrete actions to improve SVB’s risk management. However, Martin Gruenberg’s speech to the IIB implies that the problem is not limited to SVB but may be more general. How much more general? On 13 March 2023, a joint study by four professors from Columbia, Northwestern, Southern California, and Stanford universities appeared. That study claims that 10 per cent of banks in the United States (approximately 400 banks) could have unrecognised losses greater than SVB’s associated with a decline in the price of their fixed income assets. If this asset information is combined with banks that have higher proportions of uninsured deposits, it is estimated that about 190 banks could have SVB-like difficulties. One can now understand a little better why SVB and Signature Bank were considered systemic even though, officially, they were not.

In short, SVB was on the verge of being considered a systemic bank in 2015 but a legislative change in 2018 relegates it to non-systemic. Five years later it fails and, faced with the risk of contagion to similar banks (some 190 in the United States), it is treated as a systemic bank and all its depositors are guaranteed regardless of whether their deposits are insured or not. It is important to note that the systemic risk exception was adopted in 1991. Since then, and until SVB and Signature Bank came along, it had only been applied to two banks. They were Citigroup and Wachovia in 2008. At the time, they were the third and fourth largest banks, respectively, in terms of assets.  

This whole story has enormously negative consequences for the supervision of our financial system. On the one hand, because of the consideration of what it is to be a systemic institution and what contagion means. Suppose I am a bank that is not considered systemic because of its size and therefore not subject to the increased supervision of the big banks. If I fail and am able to convince the supervisor that there are other banks with the same asset and/or liability structure as me, does this information generate a risk of contagion that would put me in the systemic category? Anticipating this situation would generate perverse incentives in such banks that would not incur the cost of stricter supervision but would benefit from a resolution more favourable to their interests. In fact, as early as 2010 the Government Accountability Office (GAO) warned of the moral hazard problems of invoking the systemic risk exception.

On the other hand, it calls into question the reasons why we have a limit on deposit insurance. Among these reasons, I would highlight three. First, the limits are linked to the premiums paid by banks. Higher limits imply higher premiums, which makes intermediation more expensive and is passed on to the bank’s customers. Second, these limits are set so that, under reasonable conditions, the deposit guarantee fund is not exhausted before all insured depositors can receive their money. Third, it is argued that having a limit makes it in the interest of large depositors to control the banks’ investments since most of their deposits are not guaranteed. Arbitrarily guaranteeing deposits that were not guaranteed implies financing that guarantee so that the cost will not fall on those who have taken the risk. It also removes the incentive for large depositors to control their banks.

'These are my principles. If you don’t like them, I have others.' It would be terrible to think that Groucho Marx’s famous quote is the road-map for the implementation of regulatory and supervisory policy in our financial system.