Seemingly every day, somebody, somewhere, publishes a new postmortem on last month’s turmoil in the US banking sector.
Humanity’s perverse fascination with calamity, which I lament at regular intervals, means all market mishaps are monetizable events.
That doesn’t mean all financial journalism is a manifestation of the profit motive, but then again, what else is there at the end of the day? If you have any overhead or non-trivial fixed costs (and I don’t, by the way), keeping the proverbial lights on isn’t cheap. Catastrophes, collapses and crashes pay the bills and the margins on disaster coverage are usually fat.
Pulitzers are great, but good reporting requires a lot of time and investment. Sure, you could change the world, but you have to tie up resources, often for quite a while. There are usually legal risks involved, and depending on what kind of story it is, and where your journalists have to go to get it, people could die, in a worst-case scenario.
Meanwhile, any new grad from a decent journalism program can write another superfluous sequel to an extant body of crisis coverage, particularly if there’s an easy hook. It won’t cost you anything beyond their meager salary, and it can be done and published in a day.
That’s not to disparage incremental Silicon Valley Bank coverage, but… well, mistakes were made at SVB and also around SVB. There’s plenty of blame to go around. We’ve established that by now, and I’m not sure there’s much utility in reapportioning the blame every 12 hours.
In fairness to the media, I suppose “exposing” each and every oversight in the oversight process (if you will) is useful to the extent it could make relevant parties more vigilant in the future, although I’d point to the recurring nature of crises and the commonalities between them, as evidence to the contrary.
On Monday, in one of the latest backward-looking exposés, the Wall Street Journal informed us that when KPMG gave SVB “a clean bill of health” a mere two weeks prior to the bank’s implosion, the audit opinion “flagged potential losses on loans as a critical audit matter [but] was silent on [the] unrealized bond losses and ability to hold them given a reliance on potentially flighty deposits.”
The linked article is worth reading, particularly on a slow financial news day. And yet, I can’t help but feel like everyone is ignoring the central problem at the heart of last month’s drama, because if we acknowledge it head on, we’re admitting that the system is a house of cards.
The Journal quoted a University of Michigan professor who described KPMG’s process as an example of missing “the fire in the basement or the box of dynamite on the first floor,” even as the audit opinion was keen to “point out the peeling paint on the flower box.” But the real issue is that SVB, like all banks, was built on a shaky foundation.
Banking is a business that relies on maturity and liquidity transformation, which is just a sophisticated-sounding way of saying that only a fraction of the money banks say is readily available to depositors actually is. The rest of it is tied up somewhere, and often for a very long time.
Yes, everyone with even a passing interest in finance knows that, and sure, a majority of the educated public knows it too, but we do tend to ignore it as a matter of course, and not just because of the FDIC and what we assume is the government’s aversion to saddling bank depositors with losses. We ignore it for the same reason most of us don’t go around all day pondering what death means — it’s terrifying, so we don’t think about it.
We habitually describe SVB’s deposit base as “flighty” and “concentrated,” where the former is an oblique criticism of the bank’s depositors and the latter is a criticism of management. But if we put aside any visceral aversion to venture capitalists (and frankly, we shouldn’t treat VCs as a monolith to be uniformly demonized), how much sense does it make to blame (implicitly or otherwise) depositors? If you put your money in a bank, and you want it all back, are you “flighty”? Maybe. I guess. But really, you’re just someone (or some company) who thinks it’s your money, and therefore it should be available to you whenever you want it, in its entirety.
But it’s not. Available to you, at any time, in its entirety, or at least not if everyone else comes wanting their money too, and at the same time you want yours.
SVB’s bond portfolio wasn’t a problem because of the unrealized losses, it was a problem because if everyone suddenly called bulls–t on the maturity transformation charade, those losses might have to be crystallized, opening the door to insolvency.
It was an anarchist’s dream come true, albeit not at the scale necessary to bring about actual anarchy. Not all crypto fans are anarchists, of course, but, as JPMorgan’s Nikolaos Panigirtzoglou wrote, “For crypto supporters, the US banking crisis exposed the weaknesses of the traditional financial system given banks’ maturity mismatch is susceptible to runs.” He went on:
Crypto supporters have been arguing for a long time that the crypto ecosystem is superior not least because deposits are held in entities such as stablecoins which as a digital form of money market funds are 100% backed with high quality liquid assets and are thus less susceptible to runs. The US banking crisis and the intense shift in US bank deposits to US money market funds is viewed by crypto supporters as a vindication of the crypto ecosystem.
It’s supremely ironic, in that context, that Circle (the sponsor of the USDC stablecoin) briefly lost its peg due to short-lived ambiguity around more than $3 billion in deposits held at SVB.
Of course, without the maturity mismatch, the credit creation mechanism at banks would be severely impaired, and if you forced all depositors into term deposits, many of them would balk given that such an arrangement would effectively rule out transaction accounts.
As discussed here on any number of occasions over the years, without credit, there’s no growth. Credit is, in essence, just a manifestation of faith in the future. Without credit, there’s no future. Credit isn’t an all or nothing phenomenon. Between ultra-easy credit and no credit, there’s a continuum. The harder credit is to obtain and the more onerous the terms, the less growth there’s likely to be. That’s part and parcel of the current US macro zeitgeist in the wake of SVB’s failure.
At the end of the day, SVB’s real “problem” (note the scare quotes) was the same problem all banks have: If everyone wants their money at the same time, it’s over. But the paradox is that if banks can’t borrow short to lend long, there’s no growth. Then we’re back to the pre-modern dilemma: Growth is generally a function of credit creation, and without credit creation, growth is constrained, and the more constrained growth is, the less willing people will be to extend credit.


Well said. The bank business has been roughly the same since the Dutch goldsmiths invented it 500 years ago. Traders who owned ships and shipments would deposit their wealth with goldsmiths who had strong vaults and good reputations for honesty before departing on a long voyage. It didn’t take long for the smiths to discover that during long voyages the gold deposits the owners had entrusted to the them just laid in the vaults, earning nothing for anyone. So the smiths lent some of it to others who needed it for a time and could pay it back later. The entrusted gold still belonged to its owners but they didn’t need it so it could be used by others until the owners wanted it back. That worked great as long as only a small percentage needed their money back on a given day. Every day some specie went in and some went out. If some reason arose that folks suddenly felt their money wasn’t safe there could be a run and there would be trouble. Here we are 500 years later and the same existential issue still exists. Banks can’t really successful unless we are willing to trust them. When accountants who don’t really understand banking do audits they don’t bother with the roots of that problem. Banks I worked for would generally get three audits a year (taking a bunch of time and costing the bank a lot of money. They would get a typical accounting audit, along with at least one from the Comptroller of the Currency and one from at least one other regulator. The current mark-to-market rules created the problems at SVB directly because of rising rates.
Crypto in its present form isn’t competition for banks, but a Fedcoin might be – something for the govt to weigh.
Regulations were relaxed, the system was awash in funds and the interest rate regime changed. So even if the bank did nothing further on its own (which it did), the sea level was already rising. But what DOES seem to have changed is the ease in and our comfort with moving large amounts of money rapidly, combined with the instant rumor machines (Teams, Slack, G-Chat, Twitter even) to quickly galvanize that even speedier money. That alone would seem to make runs more likely than in the past. I alost had to choke back a chuckle seeing news coverage of depositors gathered at the already locked doors of an SVB branch looking to empty their accounts before it was too late. It’s not really how it happens anymore … instead we’ve got bigger, faster, suddenly and all at once.
My takeaway is that everything needs to happen way slower than it has been. If the Fed wants to print money, raise rates or lower rates- better move a lot slowwwwwer than they have been.
I agree with everything you say here, but the idea that its okay to buy a longer term treasury paying 55 bps is insane. So is saying that the bank couldn’t buy some long dated treasury puts …It’s all gambling at some level, and there is also the question of why the spread between a savings account and short term treasuries is so high…beware of guys who wear collar pins…
Congrats on you capital light model, H.
Let asymmetry reign!