Silicon Valley Bank and global banking uncertainty

Fifteen years after the Global Financial Crisis banks may be better capitalised but they are clearly still a source of risk for the world.

Picture of STANLIB


The question on investors’ minds is: are we at risk of another Global Financial Crisis following the collapse of Silicon Valley Bank?

The almost simultaneous losses of depositor and shareholder confidence in SVB, Signature and CSFB inevitably invokes memories of 2008 and stokes fears of a systemic problem in the banking system. These concerns are understandable: Signature and SVB represented two of the largest bank failures in recent history.

In fact, until its collapse this month, SVB was the 16th biggest bank in the USA by assets. It had been in business for 40 years and was seen as the premier lender to the venture capital/technology startup ecosystem in the US. According to its Q1 2023 mid-quarter update on 8 March, SVB banked nearly half of the US-based technology and life science companies backed by venture capital in 2022. SVB had grown in line with the explosion in venture capital in recent years, tripling deposits since 2020.

What does SVB’s failure mean?

Recent events have highlighted the scale of unrealised losses on long-duration securities and reminded us once again that the daily survival of the banking system relies entirely on the confidence of often unsophisticated depositors.

The Financial Times’ Martin Wolf summarised the condition of the US banking system as follows in his article of 21 March: ‘Banks are designed to fail. And so they do. Governments want them to be both safe places for the public to keep their money and profit-seeking takers of risk. They are at one and the same time regulated utilities and risk-taking enterprises. The incentives for management incline them towards risk taking, just as the incentives for states incline them towards saving the utility when risk-taking blows it up. The result is costly instability.’

The speed with which the Fed and the Swiss National Bank have acted is reassuring, even if they have put the issue of ‘moral hazard’ firmly back on the agenda in so doing.

Since the collapse of Silicon Valley Bank early in March, we have seen the worst bout of US financial turmoil since the 2008 crisis. But, there is no sign that the global banking system is in trouble. Central banks and banking regulators have been proactive and serious in their responses. The Fed has demonstrated its commitment providing $300bn of funding to the banks via the discount window ($152bn, more than at the 2008 peak, creating a new Bank Term Funding Program ($12bn) and guaranteeing all deposits at SVB and Signature ($142bn).

Importantly investors should not infer from the Fed’s support of SVB and Signature that uninsured deposits will be guaranteed in the event of any bank getting into difficulties: Treasury Secretary Janet Yellen testified in Congress that uninsured deposits will only be guaranteed for failing banks on a case-by-case basis if she sees a risk of contagion. This means that smaller banks are still riskier places to leave your money and will have to pay more to hold on to uninsured deposits, all things being equal. Depositors may be reassured but investors will now be focussed on banks’ risk management of their securities portfolios.

The political weather for the banks looks tricky. News that SVB’s CEO sold $30 million of stock over the last two years and that CSFB will still be paying bonuses this year will be ammunition for those who complain that the promised regulatory overhaul post-2008 is still woefully incomplete. Critics will remind us that deposit insurance represents a state subsidy for the banks’ cost of capital and yet executives are still being rewarded for risk-seeking behaviour that imperils the economy.

The link between the stability of the banking system and the level of interest rates also reduces the Fed’s room for manoeuvre as it attempts to get inflation under control. The 25bp hike on Wednesday, 23 March seems to have set the right tone: backing off now to support the banks may perversely increase uncertainty, since it may be taken as a sign of the severity of the underlying problem. After all no one has more transparency into the US banking system than the Fed.

Further reading

Why did SVB collapse?

Unlike some of the banks that failed in the 2008 financial crisis SVB did not fail because it had been lending recklessly, arguably though one can point to imprudent risk management and a concentrated depositor and loan book.

After years of impressive growth, in February 2023 SVB was expecting to be downgraded by Moody’s and asked Goldman Sachs in late February for advice as to how it could shore up its balance sheet.

On Wednesday, 8 March SVB announced that it had incurred a large loss selling securities to Goldman Sachs to raise cash to fund depositor withdrawals. The securities had an approximate book value of $23.9 billion but Goldman paid $21.5bn for them, creating a $1.8bn after-tax loss; the bank announced that it was raising over $2bn to fill the resulting hole in its capital base.

Neither investors nor depositors were reassured: the following day SVBs’ share price tanked and depositors tried to withdraw another $42bn (c.20% of the bank’s total deposits). The bank was declared bankrupt by the federal authorities during market hours on Friday, 10 March.

Essentially SVB’s business model had become a leveraged bet on US interest rates remaining low. The dramatic rise in US interest rates over the last 12 months has both depressed VC activity and hit the value of the securities that SVB had to sell to raise the cash required to meet deposit withdrawals. The losses demanded an influx of fresh capital, but raising equity while deposits were haemorrhaging was impossible; the depositors’ loss of confidence therefore became self-fulfilling.

What was the problem with SVB’s deposits?

SVB’s deposit base was unusually concentrated, with an average deposit of $5,000,000, far in excess of the Federal Reserve’s $250,000 guarantee. In effect 95% of the bank’s deposits were uninsured, and an uninsured depositor will be quicker to withdraw funds in the event of negative headlines.

In addition, SVB’s customer base was predominantly VC-backed startups who raise chunks of capital, put them on deposit and then draw them down over time in order to pay wages and fund research and development. SVB’s overall deposit base would therefore inevitably decline over time without new inflows of VC funding activity.

SVB’s deposit base has therefore been caught in a cross-fire since the beginning of 2022: VC funding activity shrank 22% year-on-year in 2022 while SVB’s depositors have continued to withdraw funds at twice the pace they did pre-2021.

What about SVB’s assets?

SVB’s asset base was also unusual, with loans accounting for a much smaller share than at other banks. Only about a third of SVB’s $175bn deposits were lent out as loans, the balance being invested in US Government and Agency bonds. SVB’s assets were therefore much more liquid than other banks but also more volatile in value since bonds, unlike loans, are publicly traded and their prices move up and down every day. One of the reasons SVB got into trouble was that they owned relatively long-dated bonds which yielded more than short-dated bonds but lose more value when interest rates increase as they have done.

Many banks have large portfolios of US Treasury bonds since they are regarded as liquid and zero-risk because they are the liabilities of the US government, who can simply print more dollars to pay back the bondholders if necessary. Regulators agree that bonds are sensible investments but also recognise the potential difficulty for banks of having to recognise changes in the bonds’ value every day. This is the origin of the ‘Hold To Maturity’ (HTM) accounting concept whereby banks do not have to reflect changes in the value of bonds that they intend to hold until they are redeemed at the end of their lives.

The current wave of uncertainty has been triggered by the realisation that banks have large amounts of HTM bond holdings which are worth much less than their official value in the banks’ accounts, and that if the banks had to sell those bonds (in order to generate the cash to repay depositors) the losses would punch a hole in their capital. These concerns can be self-fulfilling by triggering deposit flight.

If SVB had been forced to sell their HTM bond holdings the resulting losses would have eradicated the bank’s shareholders’ equity in its entirety.

Why have Treasuries and other securities dropped in value over the last year?

During 2022 the Fed responded to high inflation by increasing interest rates, which reduced the value of long-duration assets. From 7 March 2022 to 6 March 2023, the federal funds rate rose sharply from 0.08% to 4.57%, accompanied by quantitative tightening. As a result, long-dated assets similar to those held on bank balance sheets experienced significant value declines during the same period. For instance, the ETFs tracking residential (SPMB) and commercial (iShares CMBS ETF) mortgages each lost about 10% in value from March 2022 to March 2023. Long maturity treasury bonds were even harder hit, with 10-20 year and 20+ year Treasury bonds losing about 25% and 30% of their market value, respectively. Overall, as is evident, the FED’s monetary policy tightening caused significant value declines in long duration assets.

More insights