The swift downfall of Silicon Valley Bank (SVB) amidst the recent banking crisis in the United States has sparked renewed curiosity regarding the potential impact of central bank digital currencies (CBDCs) on financial stability. The collapse of Silicon Valley Bank has started a stampede in the outflow of deposits from banks. Bank deposits have fallen USD363 billion to USD17.3 trillion since the beginning of March, Federal Reserve (Fed) data show. In the first two weeks of March alone, overall bank deposits in the US fell by about USD161 billion, driven by outflows from smaller (community) banks. Concurrently, about USD340 billion has flown into money market fund (MMF), and assets under management (AUM) have risen to record levels of about USD5.2 trillion. Cash flowing into MMF ends up outside of the banking system because they are sterilised in the Fed’s overnight facility. More than 40 per cent of money-market fund assets are invested in the reverse repo facility. Daily usage of the Fed’s repo facility is running at about USD2.3 trillion. As of Wednesday, 5 March, more than USD2.2 trillion sat in the Fed’s reverse repo facility, paying a 4.8 per cent annualised rate. According to the Wall Street Journal, that is well above the rates on offer at most banks. Against this backdrop, the purpose of this blog is to explore the recent collapse of SVB and the implications for CBDCs.
Background on Silicon Valley Bank
Silicon Valley Bank (SVB) was a state-chartered commercial bank based in Santa Clara and was a member of the Federal Reserve System, with total assets of approximately USD209 billion and total deposits of approximately USD175.4 billion as of 31 December 2022. The bank was ranked as the sixteenth biggest in the US at the end of 2022.
SVB was founded in 1983 as a wholly owned subsidiary of Silicon Valley Bank shares (now SVB Financial Group) focusing on the needs of start-up companies. When the bank was founded, the banking industry did not have a good understanding of these companies, particularly those that lacked revenue. The bank’s customers were primarily businesses and people in the technology, life science, healthcare, private equity, venture capital and premium wine industries. SVB, which lent heavily to start-ups and tech companies, had branches in eight other countries, including China, India and Israel, according to its website.
The bank structured its loans with the understanding that start-ups do not earn revenue immediately, managing risk based on their business model. During the 1980s, the bank grew along with the local high-tech economy. The wave of computer technology start-ups during the dot-com bubble provided an influx of business for the bank, which was noted for its willingness to lend to venture-stage companies that were not yet profitable. SVB formally entered the private banking business in the early 2000s, building on prior experience and relationships with wealthy venture capitalists and entrepreneurs.
As of 31 December 2022, 56 per cent of its loan portfolio were loans to venture capital firms and private equity firms, secured by their limited partner commitments and used to make investments in private companies, 14 per cent of its loans were mortgages to high-net-worth individuals, and 24 per cent of its loans were to technology and health care companies, including 9 per cent of all loans which were to early and growth-stage start-up companies.
Despite banking a high-tech sector, the bank was criticised for having old technology and lacking biometric authentication. During the 2007–2008 Great Financial Crisis, SVB Financial Group received a USD235 million investment from the US federal government in exchange for preferred stock and warrants under the Troubled Asset Relief Program (TARP); and over the next two years, it paid USD10 million in dividends to the US Treasury, and then used the proceeds of a USD300 million stock sale to buy back the government’s interest.
What are central bank digital currencies?
CBDC is central bank-issued digital money denominated in the national unit of account, and it represents a liability of the central bank. If the CBDC is intended to be a digital equivalent of cash for use by end users (households and businesses), it is referred to as a ‘general purpose’ or ‘retail’ CBDC (rCBDC). As such, it offers a new option to the public for holding money. CBDC is different from cash, as it comes in a digital form unlike physical coins and banknotes. CBDC is also different from existing forms of cashless payment instruments for consumers such as credit transfers, direct debits, card payments and e-money, as it represents a direct claim on a central bank, rather than a liability of a private financial institution. This type of riskless claim also makes CBDC different from cryptocurrencies (such as Bitcoin) or other private digital tokens (e.g., so-called stablecoins such as Tether).
In contrast to retail CBDC, ‘wholesale’ CBDC targets a different group of eligible users. It is designed for restricted access by financial institutions and is like today’s central bank reserve and settlement accounts. Accordingly, it is intended for the settlement of large interbank payments or to provide central bank money to settle transactions of digital tokenised financial assets in new infrastructures (Bech et al. (2020)).
Amid the COVID-19 pandemic, work on CBDCs gained further momentum. After The Bahamas launched a live retail CBDC (the Sand Dollar) in 2020, Nigeria followed in 2021 with the issuance of its eNaira, and the Eastern Caribbean and China released pilot versions of their respective DCash and e-CNY. Based on the results of the 2021 BIS survey on central bank digital currencies, about 86 per cent of the world’s central banks are now exploring the benefits and drawbacks of CBDCs. As a whole, central banks are moving into more advanced stages of CBDC engagement, progressing from conceptual research to practical experimentation. Around the globe, interest in CBDCs continues to be shaped by local circumstances. In emerging market and developing economies, where central banks report relatively stronger motivations, financial inclusion and payments efficiency objectives drive general purpose CBDC work.
These efforts are a response to the declining importance of cash as means of payment and the challenges associated with the proliferation of new forms of private digital money (e.g., stablecoins). While CBDC aims to preserve the role of public money and fend off threats to monetary sovereignty, some policy makers are concerned about its potentially adverse effects on the financial system. One issue that has received particular attention is the effect of CBDC on financial stability. Unlike cash, CBDC can be remunerated, which could render it particularly attractive in crisis times and increase the risk of bank runs. To avert such scenarios, policy makers are considering specific design features, like on individual CBDC holding limits and tiered remuneration, as tools to safeguard financial stability.
CBDC and the implications for financial stability
The model of bank runs due to Diamond and Dybvig (1983) shows how banks’ mix of illiquid assets (such as business or mortgage loans) and liquid liabilities (deposits which may be withdrawn at any time) may give rise to self-fulfilling panics among depositors. The risk of bank run is aggravated when a profit-maximising bank seeks higher returns from risky long-term investment opportunities funded by uninsured deposits. The economic fundamentals of the profitability of the bank’s investment serve as a noisy private signal to depositors who must decide whether to withdraw their balances or roll them over. When funds are not kept in the bank, depositors can hold them in cash or as (possibly remunerated) CBDC.
Ahnert et al. (2023) develop a stylised model of bank runs with remunerated CBDC. A profit-maximising bank raises uninsured deposits to fund a profitable but risky long-term investment. After some time has passed, consumers receive a signal about the bank’s profitability and decide whether to withdraw their balances ‘run on the bank’ or roll them over. Importantly, funds that are not kept in the bank can be held in cash or (remunerated) CBDC. In this context, they measure the ex-ante probability of a bank run which is a measure of financial instability.
They found that an increase in CBDC remuneration has two effects in the model. First, it makes interim withdrawals more attractive by increasing the payoff from storing funds in CBDC for future consumption. This direct effect makes the bank more fragile, consistent with the line of argument underlying the ongoing policy debate. Second, higher CBDC remuneration induces the bank to offer more attractive deposit rates because consumers would otherwise not provide any funding. As a result, consumers have a lower incentive to withdraw their funds. This indirect effect renders the bank more stable. The total effect of CBDC remuneration on bank fragility thus depends on the relative strengths of these two countervailing forces. Under some parameter conditions, a U-shaped relationship arises between bank fragility and CBDC remuneration. Therefore, bank fragility is minimised (and welfare in the economy maximised) for a strictly positive level of CBDC remuneration.
They also assessed the impact of imposing holding limits on CBDC in their model. First, they showed that holding limits do not matter when CBDC is unremunerated: consumers are indifferent between withdrawing in cash or CBDC, because both yield a zero return. Second, when a central bank faces no restrictions on setting the CBDC rate, holding limits do not lead to additional benefits because the central bank can achieve the best possible outcome by setting the appropriate CBDC remuneration. The situation is different if the central bank cannot set the CBDC remuneration freely, for example because of overriding monetary policy considerations. Two possibilities emerge then. In case the central bank sets a relatively high CBDC rate, imposing holding limits will increase welfare by reducing financial stability risks. Conversely, if the central bank sets a relatively low CBDC rate, imposing holding limits can only increase the risk of bank runs, and therefore reduce welfare. Not imposing any holding limit at all would be desirable in the latter case.
What went wrong for SVB?
The demise of SVB is in some ways unique, and it boils down to bank’s business model and idiosyncratic characteristics. The bank was essentially the go-to bank for the technology sector in the US. SVB experienced spectacular growth during the past decade, following the meteoric wave of its start-up client base, that largely consisted of emerging growth and middle market companies in targeted niches (focusing on technology and life sciences industries).
As fundraising boomed over the last couple of the years, SVB experienced phenomenal deposit growth, quadrupling from USD50 billion to USD200 billion between 2018 and 2022. Concurrently, the bank’s loan book grew only from USD23 billion to USD66 billion. However, deposits dried up with a slowing economy and multi-year lows in capital funding for IPOs and start-ups. With venture capital and next-round funding drying up and thus no new cash for start-ups, deposit growth turned negative as companies burned through cash to fund daily operations.
The way a bank works is that it borrows short to lend long. The net interest margin is the spread a bank earn between the interest rate it pays on deposits and the rate borrowers pays it on loans. In the case of SVB, having a far larger deposit base than loan book was a problem. Consequently, SVB needed to acquire other interest-bearing assets and by the end of 2021, the bank had made USD128 billion of investments, mostly mortgage bonds and US Treasury securities. Now that interest rates are higher, and with the lion’s share of bank’s investments in dated bonds and treasuries, the market value of its securities substantially lower than SVB paid, which ultimately became the core issue.
SVB was running an aggressive funding gap (short commercial deposits against long-dated securities). The problem is that SVB had a very high concentration of deposit largely from commercial clients, i.e., start-up and tech firms. Only 3 per cent of SVB’s deposit balances were deposits insured by the US Federal Deposit Insurance Corporation (FDIC) which is applicable to a deposit limit of USD250,000. This was significantly lower than the 40 per cent median rate across the US banking industry. According to The Economist, some 93 per cent of SVB’s deposits were uninsured. Consequently, its customers, unlike those at most banks, had a real incentive to run and they responded to it over a 24 to 48-hour period.
Commercial and institutional depositors are generally more sensitive to pricing than retail depositors. These are likely to be customers without FDIC deposit insurance and as such would be highly motivated to quickly withdraw their deposits. In the case of SVB, it was the convergence of a community of owners and CEOs from Silicon Valley who had their life work and savings at stake that were spooked by concerns about their primary bank’s financial position. With the slowdown in the technology sector and a sharp reduction in venture capital funding, these clients have very good reason to draw down their deposits at a faster rate than other depositors.
Ultimately, SVB witnessed deposit run-off of approximately USD30 billion in the span of a few quarters. As of the end of last year, SVB said it had USD151.5 billion in uninsured deposits, USD137.6 billion of which was held by American customers. On Thursday, 9 March 2023, customers pulled USD42 billion from Silicon Valley Bank, leaving the bank with a negative cash balance of USD1 billion, the company said in a regulatory filing. To plug the hole, SVB was forced to sell substantially all of its USD21 billion available-for-sale securities which were mostly in the form of US Treasury securities, crystallising a USD1.8 billion loss on those securities on higher rates. Not only did the bank sell assets at a big loss, but it also said that clients’ cash burn rates had not slowed down as anticipated in the current economic climate. It said that deposits have been leaving the bank faster than expected this year.
As a result of these losses and the excessive deposit outflows, SVB then tried to raise equity to shore up confidence in the bank. On Thursday, 9 March 2023, shares of SVB fell 60 per cent after SVB Financial, parent of SVB, announced a plan on Wednesday evening, 8 March 2023 to raise more than USD2 billion in capital. Greg Becker, its chief executive, urged clients to “support us as we have supported you”. Under the terms of a plan released, SVB was looking to sell USD1.25 billion in common stock and another USD500 million of convertible preferred shares. SVB also announced a deal with investment firm General Atlantic to sell USD500 million of common stock, though that agreement was contingent on the closing of the other common stock offering, according to a securities filing. Unpersuaded by these developments, some venture capitalists told portfolio companies to run. Bill Ackman, a hedge-fund manager, suggested that the government should bail out the bank.
The stock fell another 60 per cent in pre-market trading Friday, 10 March 2023 before being halted as the bank was unable to find a buyer. A bank run ensued ultimately causing regulators to step in and shut down the bank. Deposit outflows outpaced the sale process, making it very difficult for a realistic assessment of the bank by potential buyers to take place. On Friday, 10 March 2023, the California Department of Financial Protection, and Innovation (DFPI) announced that, pursuant to California Financial Code Section 592, it had taken possession of Silicon Valley Bank, citing inadequate liquidity and insolvency. The DFPI appointed the FDIC as receiver of Silicon Valley Bank.
Money Market Funds
SVB’s business model and idiosyncratic features model made it particularly susceptible to a run by depositors. The bank’s specialised business model might have introduced significant correlation between large deposits and its business of revolving credit, on which it had substantial exposure. To add to that, the SVB leadership’s unorthodox approach to management of liquidity pressures clearly backfired. While the factors that befell SVB are not necessarily systemic or reflective of the wider US banking industry, the bank’s failure provides insight of what might transpire in the case of a CBDC.
Central banks around the world have been considering CBDC for reasons ranging from modernising payments, increasing financial inclusion, responding to the declining demand for cash or to the perceived risks posed by private digital currencies. Other potential benefits of CBDCs may include: better consumer protection, fully-insured deposit accounts, greater financial and macroeconomic stability, improved monetary policy transmission, streamlined regulation and regulatory structures, increased fiscal revenue arising from the recapture of economic rents from the financial sector and other.
However, the enthusiasm caused by these new possibilities needs to be tempered with caution, as the introduction of CBDCs may give rise to a host of potential risks and vulnerabilities for macroeconomic and financial stability. Depending on design options, CBDCs could have substantial impact on the financial system and the transmission of monetary policy. They could lead the central bank to engage in large-scale intermediation, thus competing with private financial institutions for deposits, with potentially negative consequences for the availability of bank credit, economic activity, and overall financial stability. Other concerns include operational risks (stemming from technological issues, including cyber-security), risks to financial integrity, privacy and governance, to name but a few.
Against this backdrop, the collapse of SVB has started a stampede in the outflow of deposits from banks. In aggregate, bank deposits have fallen USD363 billion to USD17.3 trillion since the beginning of March, Fed data show. In the first two weeks alone, overall bank deposits in the US fell by about USD161 billion, driven by outflows from smaller (community) banks. Concurrently, about USD340 billion has flown into MMF, whose AUM has risen to record levels of about USD5.2 trillion. Cash flowing into MMF ends up outside of the banking system because they are sterilised in the Fed’s overnight facility. More than 40 per cent of MMF assets are invested in the reverse repo facility. Daily usage of the Fed’s reverse repo facility is running at about USD2.3tn. As of Wednesday, 8 March 2023, more than USD2.2 trillion sat in the Fed’s reverse repo facility, paying a 4.8 per cent annualised rate. According to the Wall Street Journal, that is well above the rates on offer at most banks.
US Treasury Secretary Janet Yellen noted that “If there is any place where the vulnerabilities of the system to runs and fire sales have been clear-cut, it is money market funds, the financial stability risks posed by money market and open-end funds have not been sufficiently addressed.” With the sterilisation of the deposits exiting the banking system, banks may be disincentivised to grant credit/lend, and the implication is economic and productive activities may be adversely impacted.
Conclusion
The Diamond-Dybvig model shows how banks’ mix of illiquid assets funded by volatile uninsured deposits may give rise to self-fulfilling panics among depositors. The risk of bank run is aggravated when a profit-maximising bank seeks higher returns from risky long-term investment opportunities funded by uninsured deposits. The economic fundamentals of the profitability of the bank’s investment serve as a noisy private signal to depositors who must decide whether to withdraw their balances or roll them over. Interestingly, this is exactly what transpired and led to SVB’s failure.
SVB funded long-term investments with a high concentration of uninsured deposits from the tech start-up sector. The bank’s depositors had to decide whether to: (a) keeping their funds in SVB, and/or (b) digitally withdrawing their funds instantly. With option (b), the bank’s depositors can automatically place funds with either a systemically important financial institution (SiFi) or MMF. In this respect, it might be reasonable to argue that in the presence of rCBDC, when funds are not kept in the bank, depositors can hold them in cash or as (possibly remunerated) CBDC. In other words, a rCBDC especially one that pays interest could provide an outside option for depositors to convert their holdings private into public money, i.e., CBDC.
The introduction of CBDCs has the potential to disrupt traditional banking systems and, consequently, impact the stability of the financial industry. However, CBDCs also offer the advantage of real-time monitoring and oversight, which could help mitigate the risk of bank failures and enhance regulatory control.
As the discussion surrounding CBDCs continues, it is crucial to assess both the potential benefits and challenges they present to the banking sector, taking into account their impact on financial stability and the evolving needs of depositors and financial institutions alike.
Mark is a Senior Financial Sector Specialist in the Financial Stability, Supervision and Payments pillar at the SEACEN Centre.