Ireland still managed to beat Scotland with five disruptive injuries in Sunday’s 6 Nations rugby showdown. Injuries are part of the game but so are substitutes who can step in to keep the team functioning. But…sometimes even the substitutes get injured and then we are into Donald Rumsfeld’s “known unknowns” territory. Ireland did survive a double-whammy loss in the specialist hooker position to win but it was hairy stuff. Now, think about the banking industry this past weekend.
The Silicon Valley Bank (SVB) collapse is arguably a “known known”, the unknown bit just being the identity of the bust bank. That didn’t stop some pretty high profile venture cap(VC) and tech founders going into meltdown mode with striking similarities to an infamous Clare Devlin finger-pointing outburst in Derry Girls. How wonderfully appropriate does Sister Michael’s withering observation sound today: “Well, I think it’s safe to say we all just lost a bit of respect for you there Clare”. Also, probably safe to politely say that the VC world needs to brush up on its banking knowledge and credibility. To be clear, the US in an average year experiences 7 or 8 bank failures. In fact, we were due a few failures as there were none in 2021 and 2022. The banking system will be fine but there are a few new challenges for the sector. So, our analytical focus, amid the blizzard of commentary, is the ‘unknown unknowns’ which have emerged.
First, let’s deal briefly with the known unknowns. The unexpected, albeit traditional, banking factors in SVB’s collapse did echo some of our experiences in the 2008-2009 credit crisis(GFC). I would pinpoint two areas which should generate GFC flashbacks:
- There was an operational/timing mismatch between what the bank’s customers deposited(liabilities) and what the bank invested in(assets) to generate a profit(excess return) and beat what they were paying customers for deposits. There’s a lot of tosh being written about long term assets being unsuited to quick cash-out/sales to pay customers withdrawing deposits. The reality was that the vast majority of SVB’s assets were in US Treasuries and other highly liquid bond securities which can be sold in a nano-second. Yes, the maturities of these securities were long-dated (eg 10 years) but had absolutely nothing to do with an inability to sell/cash out. The timing issue was far more fundamental than maturity of the assets. Thanks to rapidly rising interest rates these securities have lost value and in an ideal world the bank would be planning to hold them full term, and experience no loss. The timing of deposit withdrawal requests was unhelpful and causing losses but why the withdrawal requests?
- The customer base of SVB was massively concentrated in the tech and start-up sector. Think back to Anglo Irish and its army of property guru customers all with the same problems, and assets/loans. Back to California, and the customer concentration issue provided a new twist on balance sheet challenges for a bank. Yes, the customer concentration risk is a banking known but these customers were not big borrowers. They were minted with VC and funding-round cash. Unlike Anglo Irish, the customers’ borrowings(assets) were not the concentration problem. The cash was the problem, the unknown. A tech slow down, higher interest rates and shareholders/VCs demanding a commensurate uplift in rates of return(profit) was forcing SVB customers, all at the same time, to tap their cash deposits at an increased withdrawal rate. More disastrously, the vast majority of this cash came in to the bank in similar circumstances. The tech and VC “gold rush” through the Covid-19 pandemic can be seen quite clearly in SVB’s asset base tripling in size from $70 billion to $212 billion in the 3 years since 2019. Sure enough, when the reversal of this trend, from funding to spending, happened through 2022 and 2023 it was relatively symmetrical. Lots of customers doing the same thing at the same time, quickly. However, there’s quick and then there’s the warp speed of our digital world.
In the 24 hours after SVB told the market that the challenges above were causing manageable losses(circa $2 billion against a market capitalisation value of $45 billion) and they planned to raise some extra capital, something extraordinarily unknown to the banking regulators and industry executives happened. Almost twelve years to the day after the Tohoku earthquake and tsunami, the banking system had its own knock-on nuclear moment. Those 24 hours saw $42 billion removed from SVB deposit accounts. As somebody who watched the GFC crisis unfold in 2008 from the inside, I can tell you that this is a different galaxy of operational speed and stress.
The miracle is that SVB after this tsunami was only about $1 billion in cash shortfall when the FDIC stepped in to stabilise the banking system. But that won’t get any press right now. What will focus regulators and bank executive minds is that seamless, digital, mobile and customer-friendly banking can kill a bank at the touch of a button egged on by dubious influencers on Twitter. Was it only Friday we were writing about bad actors and “their perennial appearance on the wrong side of eventual truths”? Here’s hoping history will be very unkind. In the interim, regulators and the US Treasury will be reviewing how the customer rights’ pendulum may have swung too far in the direction of speed.
We have often cited the compounding effect of so many new technologies on the speed of our world and regularly reference the Diamandis and Kotler book, “The Future is Faster Than you Think”. The uncomfortable truth is that things could become much much faster. Every bank in the world is looking at blockchain to cut out costs and intermediaries(read time) and crypto/digital currencies are a logical technology development in that race. Time to pause, and think about unknown speed? Then we may need to think about our second unknown unknown; where banking services and banks themselves go from here. You have read so many times our mantra about the dangers of a business model dependent on “other people’s money” and how everything can be fine, until it isn’t.
As we write, Signature Bank and First Republic Bank are about to enter the banking graveyard despite the US government and monetary authorities guaranteeing all SVB deposits, irrespective of insurance coverage. We have also regularly suggested that we think of financial services as a feature rather than a standalone business in the future. Think Amazon and delivery. Once upon a time retail and delivery were separate. Then Jeff B said that’s BS. Now I’m seeing Apple get into buy now/pay later (BNPL) activities and I’m still wondering about Elon’s PayPal formative years and his 368 million Twitter-user platform. So, will the following be “unknown” in the banking world in the coming decades……?
Global Platforms: Just as the US government realised this weekend technology (start-ups or not) impacts all businesses (payroll, security, data etc), we may need to think about global banking mirroring Big Tech and consolidating rapidly. A world with just 5-10 big banking platforms with very diversified customer bases looks like the required trade-off for increased operational speed and technological complexity(blockchain, digital currencies etc). The disappearance of smaller banks and customer focus(SVB was very much part of the tech sector success story) could curb innovation but….
Global Capital: I could drone on about yield curve shapes and duration here but let’s be very clear. Forever and a day, the bank analyst orthodoxy was that higher interest rates would be good for banks and their profit margins. So why did SVB struggle? Remember all those cash deposits? In the good old days banks were able to sell customers new products(capital) in the form of loans to buy houses, cars, credit cards, holidays etc (assets). My personal view is that it is not just banks selling capital to consumers – think BNPL(Klarna), car finance(VW, Ford banks), crowdfunding(Spark), Big Tech(Apple Card). The other side of that capital availability coin is that banks face increased competition and are struggling to match assets(loan products) with their customer service liabilities (deposits). Traditional banking “assets” will increasingly become just a feature of various consumer platforms, be it Big Tech, social media platforms, e-commerce, metaverse or whatever. The other killer for the banks is that many of their non-traditional competitors are in the ‘fast lane’ of data analytics. Better data means not just better products, but better risk management. The irony of SVB’s unmatched connection to technology and failure is that even basic banking data would have foretold a highly combustible risk environment.
SVB may be just the canary in the banking coal mine for massive concentration of banking platforms amid product/asset proliferation for consumers. We shall see, but we might just mention one other canary. In the musings above we referenced higher interest rates(true) and asset-liability mismatch (not so true). Yes, there was a timing mismatch at SVB but the assets (bonds, Treasuries etc) were easy to sell and raise cash. There is one part of the financial ecosystem which does, in fact, meet that definition of mismatch; real estate can’t be sold quickly. We note that specialist fund managers like Blackrock, Blackstone, M&G and Schroders have restricted investor withdrawals from some of their property funds. These are not traditional banks but I worry there will be a few banks who have serious concentration risk. I’m also worried about Credit Suisse trading at a value implying 85% of its book value is wiped ie the equity and bond holders are facing big losses.
Anyway, who knew higher interest rates and the capital tide going out would reveal some naked cheeks? Actually, lots of people. But if you’re really cheeky, you’d ask those champions of smaller government, socialist scare-mongering and social division in the Tory and Republican parties why the two interest rate explosions so far – pensions and SVB – could only be sorted by sensible state intervention? Clearly, they are still in the slow lane……