Review of the week: Move fast and break

When banks fail they do so suddenly and the shock can create panic that spreads trouble to other lenders. That’s why US regulators have stepped in swiftly and unambiguously.

13 March 2023

Silicon Valley Bank was taken over by regulators last week in the second-largest bank failure in American history. 

The lender had $212 billion of assets at the end of 2022, making it the 16th largest bank in the US. It had a handful of offices in other countries as well, including China, Europe and the UK. Over the weekend, in talks brokered by the Prime Minister and Bank of England, HSBC agreed to buy Silicon Valley Bank’s UK arm for $1. SVB UK has £5.5 billion in loans and £6.7 billion of deposits, mostly to companies and investors in the UK’s digital technology sector.

You tend to think of a bank run being about depositors worrying that a bank won’t have the money to return what’s in their account. Usually they smell trouble because they learn via rumours or the news that their bank has suffered stonking losses on loans they have made. Worried that there may no longer be enough money to go around, punters start lining up in huge snaking queues outside branches to drain their accounts. 

However, Silicon Valley Bank’s run has developed back to front, mainly because it focuses on a very unusual niche: highly unprofitable tech start-ups that used to be force-fed money like foie-gras geese. Now that the decade of ultra-low interest rates and cheap money is well and truly over, routine cash injections from venture capitalist backers have dried up. Meanwhile, these businesses, which make up a large proportion of Silicon Valley Bank’s deposits, are still burning through their cash at an astonishing rate. This has the same effect as a traditional bank run that’s usually driven by panicking household depositors. 

Because banks need to make more money in interest received than in the interest they pay to attract deposits, they make loans; they don’t leave lots of cash lying around just in case everyone wants to withdraw their deposits. So far, so simple. However, the quirks of Silicon Valley Bank’s tech-heavy startup clientele have again made the script of this failure a bit atypical. For years, Silicon Valley Bank’s clients received mountains of cash. They received big cheques from investors and venture capitalists and deposited them at the bank. They would then steadily drain their money as they spent and invested voraciously in pursuit of rapid sales growth. And then, if they were successful, they would secure more money from their backers. These tech customers didn’t have as much need for borrowing  though – they much preferred selling equity to investors at a steep values. That caused an imbalance: Silicon Valley Bank had a bunch of money from depositors, but not enough people and businesses to lend it to. Instead, the bank invested in huge amounts of long-dated government bonds. It had parked more than twice as much money in these bonds than it had lent out to its customers. 

As we know, the prices of government bonds with many years till maturity are very sensitive to changes in prevailing interest rates. So if interest rates rose, the value of these bonds would fall markedly. In its accounts, Silicon Valley Bank had categorised most of them as ‘held to maturity’, which meant that it didn’t have to report the big losses it was making on these bonds as interest rates rose swiftly. The reason for this is that the bond’s price will rise to its face value as the maturity date approaches, so the capital won’t necessarily be lost. Held-to-maturity assets allow companies to keep short-term fluctuations from obscuring their underlying performance. However, if the bank starts to run low of cash, it becomes a forced seller and will need to crystalise these losses today to provide the money for its depositors to scamper away with to another bank.

Silicon Valley Bank’s assets (the government bonds it bought) were highly, highly sensitive to changes in prevailing interest rates. Its deposits were also highly sensitive to interest rates, in that when rates rose startup deals and new rounds of investment melted away and its deposits started to slump dramatically. Added to this, because its customers were more business-savvy than typical depositors, Silicon Valley Bank had to increase its deposit rates more than typical banks to try to keep customers. In retrospect, it was a bomb waiting to go off.

Unlike in the UK, American regulators couldn’t find a rescuer of Silicon Valley Bank proper over the weekend. This is the preferred solution because it would mean that another bank would take on the liabilities – they would guarantee that depositors were all made whole. There are, broadly two other options. The first would be that the government steps in to bail out the bank. Given that tech companies have, in the main, been largely dismissive of regulators, and that tech founders, staff and investors have become ostentatiously wealthy, a government bailout would look very bad politically. The other solution – let the bank go bust and let the market clean up the pieces – is a risky business. Banking is a dark magic. It is great and powerful and fuels human progress, but it is highly dependent on confidence and keeping at bay the madness of crowds. That’s why modern banking has developed such an intricate web of regulation and monitoring, and why governments have given them so many tools to wield at a moment’s notice when things go wrong.

Silicon Valley Bank wasn’t the only bank to go pop either. Two other, smaller, banks that were close to the tech industry generally and the cryptocurrency boom specifically, were New York’s Signature Bank and San Diego’s Silvergate. Both are now defunct.

We have long argued that the risk of a financial crisis is slim, and we continue to believe this. As we explained above, these banks failed because of quirks in their niche markets. Unlike Lehman Brothers – which failed in 2009 – systemically important financial institutions aren’t significantly connected with Silicon Valley Bank through loans or access to money markets. The really important mega banks are also in rude health, so effects on the wider financial system seem extremely unlikely. These banks haven’t brazenly mismatched their assets and liabilities, they have more diverse sources of funding, and their shorter-term liquidity is scrutinised by fiercer regulation.

That hasn’t prevented a swift fall in indices of bank shares worldwide and much larger falls in the prices of some smaller operators, of course. People are scared, which is why US regulators have rolled out large funding schemes to ensure banks can honour withdrawals. Banks can now swap assets for cash with the US Federal Reserve, crucially at face value, i.e. what they would get at maturity, not what the market price of the asset is today. You can see how this is targeted at this problem that we identified above: long-term bonds that are trading well-below their face value.

While Silicon Valley Bank’s failure is unlikely to be the canary in a systemic banking crisis coalmine, this is a clear sign that, when interest rates are raised so sharply, things break. Expect more things to break as we head towards recession.

Budget week

UK Chancellor Jeremy Hunt will reveal the Budget for the coming year on Wednesday.

Much of the government’s financial decisions were set out back in the Autumn Statement. A large belt-tightening is planned for the coming five years, although the squeeze on government spending won’t actually start till 2024. Taxes will rise from 6 April, however, due to the winding down of tax allowances and the reduction of some thresholds.

Hunt has more room for manoeuvre now that the yield on government bonds – the country’s cost of borrowing – has fallen markedly since his first outing. Meanwhile, British GDP growth has been better than expected, which means higher tax receipts. On the other side of the equation, the government has been battling a wave of strikes, and higher wages to the tune of billions will likely be needed to put the disputes to rest. 

In interviews, Hunt has said that he wants to use the upcoming Budget to “break down the barriers that stop people here in the UK from working”. One rumour is that he will increase pension caps to encourage older, skilled workers to keep working rather than retire because of large taxes on their pension contributions. Annual pension contributions – which are tax-free – are currently capped at £40,000. More scandalously, the lifetime allowance for pensions (above which you are taxed more heavily when you withdraw your cash) is just £1,073,000 today. It was originally set almost 20 years ago at £1.5 million! This lifetime allowance is not a big number when you think about the effects of inflation and how much will be required for a comfortable retirement in 10, 20, 30 years’ time. If Hunt can address this and encourage people to both continue working longer and put money away for their retirement, that would be long overdue, in our view.

Still, eking out skills from older cohorts cannot be the only solution to the UK’s chronic skills shortage. We hope Hunt also has plans for boosting the skills of younger Brits, too, through training grants, apprenticeships and encouragement for young people to learn and boost their earning potential. The UK has a terrible productivity problem that needs addressing, which will require a multi-faceted plan.

Hunt will also have his eye on the cost-of-living crisis that continues to bubble away. While gas prices have fallen dramatically from their height, that has yet to trickle through to lower bills for households and businesses. That’s because power retailers had to buy power in advance at the high prices to ensure they weren’t exposed to further gains in power prices that would leave them making losses on every kilowatt of power they supply. The government’s price cap is set to rise next month, taking the average annual household bill to £3,000 from £2,500. If Hunt can keep the level as it is he will no doubt win some brownie points for the government.

If you have any questions or comments, or if there’s anything you would like to see covered here, please get in touch by emailing We’d love to hear from you.

Download PDF

* An earlier version of this article said that pension contributions above the lifetime allowance are taxed at your marginal income tax rate. This was wrong, we’re sorry for the mistake.