Silicon Valley Bank and the Politics of Digital Casino Chip Insurance
When tech startups holding bank chips become Too Big To Fail
In November 2022 I released The Casino Chip Society. My aim was to popularise a simple metaphor that could help us defeat the OTOMI - the One Type of Money Illusion. Too many people go through life believing that what they call ‘bank deposits’ are basically just repositories of government money that they’ve stashed in a bank. Part of the reason for this is a linguistic convention in the English language, in which the verb ‘deposit’ typically creates a noun ‘deposit’ (such as when we say “the flood deposited silt into the river, and now there’s a silt deposit in the river”). We often believe that because we deposit (verb) money into a bank, the subsequent ‘deposit’ (noun) must be the thing we put into the bank, but this is very wrong.
In reality, the thing we deposit into a bank is taken away from us, and we are given something entirely different in return, which is why I use casino chips as a metaphor. The thing I deposit into a casino is cash, but the thing I get given in return is chips. The casino owns the cash. I own the chips. Now for the linguistic glitch: imagine the casino started calling the chip a ‘deposit’. This is exactly what happens in the banking sector. All those units in our bank accounts are bank-issued digital chips, but we have to contend with this god-awful terminology in which they’re referred to as ‘bank deposits’. In reality, bank deposits are not the thing we ‘put in’ to a bank. They’re the thing issued out by banks. This is why I believe it’s much more useful, and less confusing, to just talk about Layer 1 government-issued money and Layer 2 bank-issued chips.
So what does this have to do with Silicon Valley Bank? Silicon Valley Bank was very skillful at marketing itself to tech startups, and many of them began depositing (verb) money into the bank over the last several years, which means Silicon Valley Bank’s Layer 1 reserves were going up while they issued out a whole lot of Layer 2 digital chips, or ‘deposits’ (noun) to those startups. For a beginner’s guide to some of the dynamics of how banks build up their balance sheets, check out the new version of my Emotional Guide to ‘Fractional Reserve Banking’, but the very crude version of the SVB story is that the bank was accumulating Layer 1 reserves as assets, but those aren’t very profitable to hold, so they were using those to obtain various long-term bonds as assets instead.
The twist is that recent interest rate rises pushed down the value of older bonds that they were holding, because those were issued in a lower-interest rate environment (this is a technical topic that we need not get into here). This means jitters began to spread about the asset side of SVB’s balance sheet, and this means the people and companies holding SVB-issued digital casino chips began to edge towards the door.
Silicon Valley isn’t just an economic powerhouse. It’s a relatively close-knit cultural community with people who spend a lot of time networking to make professional connections and to build reputation. This is an economy of agglomeration, where industry blooms precisely because everyone is in close proximity to each other. This means a dense network of people all start receiving the same WhatsApp, Telegram and Discord notifications when some news is hitting the ground, such as concerns about a specialist bank that many of them had accounts with.
This is especially the case when a bunch of those people are part of startups that are funded by the same ecosystem of venture capital firms. In my last piece Zen and the Art of CBDC Analysis I talked about the contradictions of capitalism, a Marxist term for an economic fallacy of composition in which capitalists assume that what makes sense at an individual level must make sense at a collective level too (or, more realistically, they don’t even think about the collective level). Here we have a beautiful example: venture capitalists told their startups to pull their money from SVB in order to protect their short-term interests, seemingly unaware that in doing so they’d trigger a chain reaction that would undermine the collective interests of all the startups in whole ecosystem.
This means that all the people with accounts at SVB started stampeding for the exits, like a bunch of gamblers holding chips in a casino that they start to have doubts about. And that’s when the crisis hit and the the government stepped in.
But this takes us to one of the most controversial topics in the whole saga: bank deposit insurance. As mentioned, ‘bank deposit’ is an attrocious term, so we really should refer to this as ‘bank-issued digital chip insurance’. Digital chip insurance is run by the FDIC (the Federal Deposit Insurance Company), a US government agency that’s supposed to dampen the chance of bank runs by providing assurance to all digital chip holders that even if the bank goes bust, its chips can be redeemed at par for government money again. Historically there was a limit to this of $250,000, but the problem for tech startups is that they tend to raise tens of millions at a time from venture capitalists (to do stuff like building platforms for AI-generated cat pictures and so on), and because they’re not very focused on financial risk management, they just might keep $10,000,000 worth of Layer 2 digital casino chips in their account, rather than diversifying.
So when Silicon Valley Bank goes bankrupt, all those Layer 2 chips are called into question, and given that those firms use those chips to pay their developers, all that progress on building AI-generated cat picture platforms, or killer drone technology, or mind-control devices, or surveillance tech is called into question too. It turns out that the US government sees the startup ecosystem as being collectively Too Big To Fail (TBTF), so is swooping in to bail them out by overriding the traditional limit on digital chip insurance.
This raises a question I briefly touched upon in my Casino Chip Society piece. In that piece I made a clear distinction between Layer 1 state-issued money and Layer 2 bank-issued money, but if Layer 2 money is implicitly guaranteed to be redeemable into Layer 1 money, is it in fact Layer 2, or is it really a variant of state money that’s had its issuance delegated to the banking sector?
I suspect that’s a question that will be discussed in the coming months, but I’d like to reiterate that in most countries Layer 2 bank-issued digital chips are in fact not legal tender, even if they are widely used for so-called ‘cashless’ payments, and that’s a meaningful legal difference. If you want to delve deeper into the technical weeds of this, I’d recommend checking out this interesting piece from my friend Nathan Tankus, who makes a suggestion that concessions could be extracted from the banking sector by giving all Layer 2 chips insurance in future. I’d also encourage you to listen to this Odd Lots podcast episode for some solid analysis of the realpolitik of bank bailouts from Dan Davies.
Let me conclude with one final point that few commentators will pick up on. The FDIC insures bank deposits, but we know that these are actually bank-issued digital chips, and these Layer 2 chips are the core component of what’s referred to as the ‘cashless society’. This means the FDIC is also acting as a Federal Cashless Society Insurance Corporation, providing an implict government backstop to Layer 2 bank domination of the US monetary system. This feeds into the parallel debate about CBDCs: as mentioned in Part 1 of my Zen and the Art of CBDC Analysis, many libertarians are convinced that the state wishes to take over the digital money system via CBDC, and yet it’s pretty obvious that the state spends a helluva lot of its time trying to protect confidence in the privately-run bank digital payments system. I’ll try get into that in Part 2 of that piece, but until then I encourage you to stop using the term ‘deposits’, and to call them for what they are.
:D:D:D "all that progress on building AI-generated cat picture platforms, or killer drone technology, or mind-control devices, or surveillance tech is called into question too." Oh nooo! 😱
That gloating description of SV got me by surprise. Had to laugh loud.
Hey Brett, hope you're doing well and I'm glad to see you writing on Substack again.
Re: "the contradictions of capitalism" and "seemingly unaware that in doing so they’d trigger a chain reaction that would undermine the collective interests of all the startups in whole ecosystem.".
This is somewhat of a pet peeve of mine. Although I am not part of the SV ecosystem and don't have access to any of their group chats, I very much doubt it happened the way you describe. I suspect everyone, especially the most connected/powerful knew exactly that mass withdrawals could undermine the bank and that it would be better for them, collectively, that they not do that.
That does NOT mean they had the power to stop it or even that attempting self-sacrifice by not withdrawing themselves was a good idea..
Collective action problems are hard, actually. Nor are they limited to 'capitalism', they happen in any social system. There are countless historical examples of armies that had a good chance of winning if they stuck together breaking on contact with the enemy and getting routed.
Knowledge alone is not a solution to a collective action problem. Actual solutions require either very strong interpersonal solidarity (not easy to get, not the kind of thing you can build in group chats) or some sort of institutional power to enforce the collective good.
For instance, if the casino chip owners of SVB had previously organized a "withdrawal institution" (WI) that had the sole right to request withdrawals from SVB, they could have used this to stymie the bank run by having that institution reject withdrawals that were unnecessary and save SVB.
Of course, this WI could have led to a lot of other issues. More paperwork for normal withdrawals, the costs of needed withdrawals being flagged as 'unnecessary', abuse of power by the WI, capture of the WI by SVB, etc.
Institutions are hard, actually. Concentrating power to enforce solutions to collective action problems safely is something people have been trying to do for thousands of years. With very mixed success.
"If only they had organized and worked together instead!" says the left. "They could have easily solved their problems".
"What a brilliant insight!" I say, "Dangling off the edge of this cliff would be a lot less dangerous if I could just sprout wings and fly off. How about that huh?"