Felix challenged me:
'uncapped deposit insurance makes runs more likely" is one of the all-time great dsquared slatepitches and I really need to see it spelled out in your new newsletter
I promised in the Manifesto that I was going to stop doing contrarianism, but who can resist a challenge? I will do this twice. First by cheating and defending the thing I wish I’d said rather than what I actually said, and then properly. There is a point of general importance here though. In a financial system you never want to create zeros. Things which have zero risk, zero capital requirement or zero cost, because these are the things which have zero restrictions on their ability to grow.
My first cheaty response to Felix is to pretend that I’m answering the question “uncapped deposit insurance makes bank failures more likely”. Which it really would, and I have evidence because uncapped deposit insurance has actually existed in the USA and it was a disaster.
It was in the years of the S&L crisis and they were called “brokered deposits”. A financial firm would do the admin for you of splitting up your big deposit into a lot of little deposits and spreading them out across the system. Deposit brokers became the financing source of choice for bad or fraudulent S&Ls, precisely because they were completely risk-insensitive money that would go straight to any bank, no matter how crappy, if it promised a little bit more of an interest rate. This is also the reason why you don’t hear much about the Federal Savings & Loan Insurance Corporation (FSLIC) any more. Zero risk with zero cost = eventual disaster.
Brokered deposits still exist, but they’re subject to regulation. So if you want uncapped deposit insurance, you can still have it. You just need to a) go through the hoops of helping to make sure that deposit funding is actually diversified small accounts, rather than a couple of dozen VCs concentrated in one bank. And b) accept that the deposit broker is going to be making some cursory checks to make sure they’re not sending money to one of the banks that’s prohibited from accepting brokered deposits because it’s on the FDIC naughty list. And c) pay a modest fee for doing so, reflecting the fact that it is you, not the system as a whole, that is benefiting from the deal. Brokered deposits are no longer zero cost, so it doesn’t matter so much that they’re zero risk.
My more arguable, not necessarily right but non-cheaty response is to thump the table and emphasise that I could win this bet at *very low absolute levels* of bank runs. Classic deposit runs are extremely rare – most banks which fail, in the USA or elsewhere, fail because they made bad loans. And uncapped deposit insurance would mean that there wouldn’t be any benefit (in funding terms) to diversifying your deposit base. It would put the deposit broker industry out of business, because you could get the zero risk at zero cost again.
Which would mean you’d get some – not necessarily very many, but some, and I don’t need a large absolute number – banks which wouldn’t diversify their deposit base *at all*, and who would get their funding from a very small number of large depositors, probably connected parties.
That’s intrinsically an unstable situation; if the connected parties all move at once (possibly for some economic reason, but quite possibly just because the CEO of the bank did some objectionable tweets) then it’s entirely possible that the bank wouldn’t be able to replace their deposits; it’s a small bank, it doesn’t have other customers or a marketing operation to get them, and ex hypothesi in this world, there’s no brokered deposits because deposit brokers don’t exist in a world of uncapped insurance.
My intuition here is that in a world of uncapped deposit insurance, there’s no incentive to take anything remotely seriously; people might stage a bank run for YouTube content, or to make a point about wokeness or as a promotional stunt for a remake of It’s A Wonderful Life. If there’s no cost and no friction, then there’s no cost to doing anything.
The point of the slogan about zeros is that “regulation” is not just a rulebook. The financial system (and lots of other systems) are meant to be “regulated” in the sense of being subject to negative, stabilising feedback. If you remove a limit, or make something free, or set a zero capital requirement, then you’re removing a constraint, and you need to be confident that what you’re doing isn’t going to cause the system to become unregulated in that sense.